Jean-Christophe Chiem , Hatem Alharazin , Everardo D Saad , Koji Oba , Masaru Muto , Hisakazu Yamagishi , Junichi Sakamoto , Takaki Yoshikawa , Marc Buyse Author Notes

Content published in The Oncologist, Volume 31, Issue 4, April 2026, oyag081, https://doi.org/10.1093/oncolo/oyag081.

Abstract

Background

Given the superior relapse-free survival (RFS) and different safety profiles of 1 year of adjuvant S-1 or uracil/tegafur (UFT) for stage II/III rectal cancer, the benefit-risk of these 2 regimens was formally assessed using the net treatment benefit (NTB).

Patients and methods

Individual patient data from the Japanese Foundation for Multidisciplinary Treatment of Cancer (JFMC) 35-C1 trial were used. S-1 and UFT were compared regarding RFS, incidence of grade ≥3 symptoms, and incidence of grade ≥3 laboratory abnormalities reported as adverse events (AEs). Laboratory abnormalities and symptoms were analyzed as binary variables and as counts. Univariate and multivariate NTBs were computed for various ways of prioritizing the outcomes.

Results

The univariate NTB for RFS was 9.2% (95% CI, 3.4%-15.2%, P = .005) in favor of S-1. The univariate NTB was not statistically significant for any symptom. For grade ≥3 laboratory AEs, only thrombocytopenia was statistically significant in favor of UFT (NTB = −0.8%; 95% CI, −1.6% to −0.02%; P = .044). In the multivariate analysis considering RFS as the outcome of first priority, the incidence of grade ≥3 symptoms as second, and the incidence of grade ≥3 laboratory abnormalities as third, the multivariate NTB was 8.8% (95% CI, 2.7%-14.9%, P = .014) in favor of S-1. In sensitivity analyses according to age group, the NTB was generally positive for patients <70 years but nonsignificant for those ≥70 years old.

Conclusion

The reanalysis of the JFMC 35-C1 trial suggests that S-1 has a superior benefit-risk to UFT when RFS is considered as the outcome of first priority, followed by the incidence of grade ≥3 symptoms and of grade ≥3 laboratory abnormalities.

Highlights

• S-1 improves RFS vs UFT in patients with rectal cancer, but this alone does not reflect the benefit-risk balance.
• A novel statistical methodology can provide a patient-centric benefit-risk assessment called the net treatment benefit (NTB).
• The NTB prioritizing RFS > Symptoms > Lab abnormalities, shows a significant superiority of S-1.
• The NTB shows age-based difference for S-1: superiority in patients <70 but no significant difference in those ≥70.

Implications for practice

In Japan, both S-1 and UFT are chemotherapy options in the adjuvant treatment of rectal cancer. The benefit-risk balance between these 2 competing treatments can be assessed formally using generalized pairwise comparisons. When this method is used, the superiority of one of the interventions depends on how individual stakeholders prioritize efficacy and safety outcomes based on their personal preferences. Our results suggest that choosing between these 2 agents will depend on the selection and relative prioritization of outcomes that matters to patients with rectal cancer who are candidates to adjuvant chemotherapy.

Continue reading in The Oncologist, Volume 31, Issue 4, April 2026, oyag081, https://doi.org/10.1093/oncolo/oyag081.

Original content published in pharmaphorum here.

By definition, rare disease trials are conducted under obvious structural constraints. Patient populations are extremely limited, making recruitment challenging, and the clinical profile of patients is often heterogeneous. An analysis of 199 discontinued rare disease trials found that insufficient patient accrual was the leading cause of discontinuation, accounting for up to one third of cases.¹ In this context, trials should be designed to maximise the information obtained from each enrolled patient.

Despite this, many rare disease trials continue to rely on a single outcome to assess efficacy and determine success or failure. While multiple outcomes are collected, they typically play limited or no formal role in the treatment efficacy conclusion or regulatory approval. This conventional approach does not always reflect how treatment benefit is experienced in rare diseases, where effects are multidimensional and trade-offs between efficacy, safety, and quality of life are central to decision making.

Limitations of single endpoint designs in rare diseases

Traditional superiority trials are powered on one primary outcome. If that endpoint does not meet its predefined significance threshold, the trial is considered negative, even if improvements are observed in other clinically meaningful domains.

In rare diseases, this creates two structural challenges.

First, focusing on a single endpoint underutilises available data. Trials frequently collect information on functional status, patient-reported outcomes, adverse events, and event-based outcomes such as hospitalisations. When excluded from the primary analysis framework, their contribution to the overall assessment of benefit/risk profile is minimal.

Second, powering a study on one outcome may require larger sample sizes than are realistically achievable in small populations. When recruitment is difficult and patient numbers are fixed by epidemiology, improving statistical efficiency becomes essential.

These constraints have led to growing interest in methodologies that formally integrate multiple outcomes into a single assessment of treatment effect.

Outcome selection and prioritisation

A multidimensional analysis begins with structured outcome selection. Clinically meaningful endpoints must be identified based on disease biology, treatment mechanism, and stakeholder input. These may include survival, organ function, symptom burden, functional capacity, patient reported quality of life, or safety outcomes.

However, selection alone is insufficient. Outcomes must also be prioritised. Not all endpoints carry equal clinical importance. In progressive conditions, mortality or irreversible organ damage may take precedence. In other settings, maintaining independence or minimising severe adverse events may rank higher.

Prioritisation requires explicit decisions. Sponsors increasingly engage patient advocates, clinicians, and sometimes payers to define a hierarchy reflecting shared views of clinical value. Once established, this hierarchy can be embedded directly into the statistical analysis plan, either as a multidimensional primary endpoint or as a key secondary endpoint complementing a conventional primary analysis.

The ordering of outcomes directly determines how treatment benefit is evaluated, making stakeholder priorities operational within the design.

What is the Net Treatment Benefit?

Net Treatment Benefit (NTB), estimated using the Generalised Pairwise Comparisons (GPC) methodology, is designed to implement such a hierarchy.

The method forms all possible pairwise comparisons between patients in the treatment group and patients in the control group of a randomised clinical trial. For each pair, outcomes are assessed sequentially according to the predefined priority order. The comparison starts with the highest ranked endpoint. If a clinically meaningful difference is observed, the pair is classified as favourable or unfavourable for treatment. If not, the comparison proceeds to the next outcome in the hierarchy.

After evaluating all pairs, NTB is estimated as the difference between the probability that a patient randomly selected from the experimental group has a more favorable overall outcome than a randomly selected patient from the control group. The result is a single summary metric reflecting the overall benefit risk profile across all prioritised endpoints. The statistical framework and properties of GPC have been described extensively in the methodological literature.²

Statistical efficiency in small populations

By incorporating multiple clinically relevant outcomes into one unified analysis, NTB can improve statistical efficiency. Rather than relying on a single endpoint, the method leverages information across prioritised outcomes. Clinically meaningful thresholds can be specified to distinguish trivial from important differences, improving interpretability.

This is particularly relevant in rare diseases, where expanding sample size is often not feasible. A design that extracts more information per participant can reduce the number of patients required to achieve adequate power, or increase the probability of detecting an effect within a fixed population.

An example from the rare disease field illustrates this potential. In a post hoc analysis of the Phase 3 COMET trial in Pompe disease, investigators applied a prioritised multidimensional approach incorporating forced vital capacity and the 6-minute walk test. While the original analysis did not demonstrate statistical superiority on the primary endpoint alone, the prioritised analysis provided evidence favouring avalglucosidase alfa over alglucosidase alfa.³ This case shows how integrating clinically relevant outcomes can alter interpretation of treatment effect in small samples.

Implications for regulatory and HTA evaluation

Regulatory authorities increasingly require comprehensive benefit risk assessments. A multidimensional endpoint defined through explicit prioritisation provides a structured way to present the overall clinical profile of a therapy.

For regulators, NTB offers a framework that integrates mortality, functional outcomes, patient reported outcomes, and safety within a single estimate. For health technology assessment bodies and payers, the same framework can clarify how a therapy performs across outcomes that influence long term resource use and quality of life.

When outcome selection and prioritisation are defined early in development, the evidence base supports continuity from trial design through regulatory submission and reimbursement discussions. The same hierarchy used to evaluate efficacy can underpin value dossiers and payer negotiations, aligning clinical evidence with market access strategy.

Rare disease trials require approaches that reflect both scientific rigour and structural constraints. Single endpoint designs may not fully capture multidimensional treatment effects or optimise statistical efficiency in small populations.

Net Treatment Benefit provides a formal framework to integrate multiple prioritised outcomes into a single, interpretable measure of overall treatment effect. By requiring explicit selection and ordering of clinically meaningful endpoints, it enables incorporation of patient and clinician priorities into trial design. By leveraging information across outcomes, it can improve efficiency when sample size expansion is limited.

As rare disease research evolves, methodologies that align endpoint design with stakeholder priorities and make fuller use of available data are likely to play an increasingly important role in regulatory and health technology assessment contexts.

References

(1) Rees, C. A. P., Pica, N., Monuteaux, M. C., & Bourgeois, F. T. (2019). Noncompletion and nonpublication of trials studying rare diseases: A cross sectional analysis. PLoS Medicine, 16(11), e1002966.

(2) Buyse, M., Verbeeck, J., Saad, E. D., De Backer, M., Deltuvaite Thomas, V., & Molenberghs, G. (Eds.). (2025). Handbook of Generalized Pairwise Comparisons: Methods for Patient Centric Analysis. Chapman and Hall/CRC.

(3) Verbeeck, J., Dirani, M., Bauer, J. W., Hilgers, R. D., Molenberghs, G., & Nabbout, R. (2023). Composite endpoints, including patient reported outcomes, in rare diseases. Orphanet Journal of Rare Diseases, 18, 262.

About the author

Tom Mann is clinical solutions engagement lead at One2Treat. He brings over 15 years of experience in tech start-ups and scale-ups, where he played a pivotal role in driving customer engagement, marketing initiatives, and strategic partnerships. With a strong background in SaaS companies, Mann has a deep understanding of customer needs.

Original content published on JAMAnetwork.com here.

Key Points

Question  Can generalized pairwise comparisons be used to assist with shared decision-making between patients and clinicians?

Findings  This comparative effectiveness study used patient-level data from a randomized clinical trial comparing the outcomes of antibiotics vs appendectomy. Using generalized pairwise comparison, the net treatment benefit significantly favored antibiotics, was neutral, or significantly favored appendectomy, depending on the patient’s order of priority.

Meaning  This study found that prioritized outcomes are a powerful tool to assess the benefit-risk of a new treatment compared with standard of care in a mathematically rigorous way, providing the outcomes are prioritized to reflect patient-specific choices.

Abstract

Importance  Shared decision-making (SDM) can be made difficult by the multifaceted nature of outcome assessment. A rigorous method for analyzing results from multiple outcomes is called generalized pairwise comparisons (GPC), which could assist in SDM.

Objective  To examine whether GPC can be useful in SDM by using individual-patient data from the Comparison of Outcomes of Antibiotic Drugs and Appendectomy (CODA) trial.

Design, Setting, and Participants  This comparative effectiveness study used data from participants in the multicenter US CODA trial (conducted between May 2016 and March 2020). All possible pairs of patients (one from each arm) were formed to analyze each of 7 outcomes of interest sequentially. Data were analyzed between February 2020 and early 2024.

Exposures  Three scenarios of priorities related to a different order of outcomes were considered. The first scenario came from a consensus exercise with patients that favored antibiotics, whereas the other 2 were arbitrarily chosen to illustrate the range of possible outcomes depending on prioritizations. Scenario 2 favored neither treatment, and scenario 3 favored appendectomy.

Main Outcomes and Measures  The primary outcome was the net treatment benefit (NTB), a formal measure of benefit-risk, which is the net probability that a randomly selected patient from the antibiotic-assigned arm would have a more favorable outcome than a randomly selected patient from the appendectomy-assigned arm.

Results  A total of 1552 patients were included in the CODA trial, with 776 (mean [SD] age, 38.3 [13.4] years; 286 [37%] female) in the antibiotic arm and 776 (mean [SD] age, 37.8 [13.7] years; 290 [37%] female) in the appendectomy arm. The NTB of antibiotic treatment was 12.8% (95% CI, 7.1% to 18.3%; P < .001) for the first scenario, 3.2% (95% CI −2.4% to 8.7%; P = .27) for the second, and −14.5% (95% CI. −20.2% to −8.8%; P < .001) for the third. These results respectively favored antibiotics, neither treatment, or appendectomy, thus illustrating that benefit-risk varies considerably according to individual priorities.

Conclusions and Relevance  This comparative effectiveness study of antibiotics and appendectomy illustrates that the GPC method is a flexible yet mathematically rigorous quantitative analysis of benefit-risk balance. This method provides a more exhaustive and nuanced quantitative assessment of the differences between 2 treatment modalities in terms of prioritized outcomes. Furthermore, GPC could support SDM by considering individual prioritizations of the multiple outcomes.

Introduction

Once the purview of physicians alone, clinical decision-making has evolved into a shared process in which a patient’s preferences, circumstances, and priorities are elicited to determine a preferred treatment or course of action.¹ Shared decision-making (SDM) improves patient satisfaction, can help avoid decisional regret, may improve adherence to treatment, and most importantly, honors the ethical principle of autonomy in several medical fields, including surgery.^1–4 As a best practice, the SDM process includes a standardized decision aid that provides the needed information about the treatment options and elicits patient preferences.^5,6 A challenge in creating such decision aids is that treatments often have different effects on multiple outcomes and presenting results related to multiple benefits and risks can lead to confusion and cognitive overload.⁷ A variety of methods have been identified for comparing patient preferences between specified alternatives; such preferences may relate to an individual’s ranking of outcomes as part of the SDM process, and this can be done in a rigorous manner with specific quantitative techniques.⁸ In addition, multicriteria decision analysis (MCDA) is a set of powerful tools that allow formal assessment of multiple outcomes by individual and group-level stakeholders.^9,10 However, a common feature of existing quantitative benefit-risk analyses such as MCDA is that multiple outcomes are analyzed separately, and the summary measures of these marginal analyses are then aggregated.

Generalized pairwise comparisons (GPC) are an emerging class of statistical methods for the comparison of 2 samples of patients (eg, in randomized clinical trials) in terms of several outcomes, possibly prioritized.¹¹ The win ratio and the desirability of outcome ranking (DOOR) are 2 popular instances of such GPC analyses.^12,13 In this article, we use GPC exactly as in a win ratio analysis of prioritized outcomes, but using the net treatment benefit (NTB) instead of the win ratio as a measure of treatment effect, given some limitations in interpreting the win ratio that make it inappropriate for benefit-risk analyses.¹⁴ Briefly, GPC assumes an order of preference among the multiple potential outcomes that may result from 2 competing interventions. Note that this order of preference may be patient dependent, as illustrated later in this article. By doing this, different outcomes can be jointly analyzed using pairs of patients that are formed by taking each patient from the experimental group and comparing them with each patient from the control group. The ordered outcomes are sequentially compared in each pair, from the outcome considered most important to that considered least important. When outcomes are related to benefits and harms, GPC provides a formal benefit-risk analysis of a given treatment.¹⁵ In each pairwise comparison, a pair is classified as favorable to the experimental treatment, unfavorable to the experimental treatment, or neutral. The GPC method has been used in cardiology for several conditions, including amyloid heart disease,¹⁶ heart failure,¹⁷ and myocardial infarction,¹⁸ with the win ratio as the measure of treatment effect.^13,19 Here we use the GPC method to assess the net benefit of competing interventions in terms of the NTB, an absolute measure of effect that has a more intuitive interpretation to patients and is adequate to combine treatment effects on multiple outcomes having different baseline risks (ie, the risks expected in the control group).^11,20–23 Unlike MCDA and similar quantitative methods of benefit-risk analyses, GPC analyses implicitly take the correlation between the outcomes into account, which is useful to distinguish situations in which patients who develop toxic effects are also those more likely to derive benefit from a given treatment, arguably a more desirable situation than if patients with toxic effects would not derive such benefit.¹⁵

While SDM has most commonly been used for elective procedures, there is an increasing interest in its application to other conditions, including those that present acutely, such as uncomplicated appendicitis.²⁴ Over the past 20 years, multiple randomized clinical trials (RCTs) have compared antibiotics alone vs appendectomy and found an antibiotic strategy to be safe, effective, and noninferior for overall health status,^25,26 even though doubts remain about the risk of recurrent appendicitis and hospital readmission within 1 year.²⁷ Since there are many outcomes that are associated with the treatment of appendicitis, many of which with different perceived impacts on different individuals,^28,29 it was hypothesized that the GPC method applied to these outcomes might help future patients identify a treatment strategy based on their personal prioritization of different outcomes. We applied the GPC method to the individual-patient data from the Comparison of Outcomes of Antibiotic Drugs and Appendectomy (CODA) trial, a US multicenter RCT comparing appendectomy vs antibiotics in adults.²⁶ If the GPC method was effective in distinguishing treatment strategies based on patient prioritization of outcomes, it might have value in complementing existing SDM-based decision aids for appendicitis.

Methods

Data Source

The current work is based on individual patient data from the CODA trial (NCT02800785), an RCT that investigated the noninferiority of a 10-day course of antibiotic therapy as an alternative to surgery for the treatment of uncomplicated appendicitis. The CODA trial met the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.²⁶ In total, 1552 patients were stratified by the presence or absence of appendicolith and were randomly assigned in a 1:1 ratio to receive antibiotics (776 patients) or appendectomy (776 patients) at 25 US centers.²⁶ The primary end point was 30-day health status, which was assessed by the European Quality of Life–5 Dimensions (EQ-5D) questionnaire.³⁰ Higher scores on this instrument indicate better health status; a minimal clinically important difference of 0.05 points has been established for posttraumatic stress disorder and was used in the CODA trial.³¹ Secondary end points of that trial included the total numbers of workdays missed by the patient and by the caregiver within 30 days of enrollment, the duration of hospital stay, symptom resolution (absence of the following: pain in lower right quadrant, tenderness in lower right quadrant when pressed, fever, shaking, and chills) at 2 weeks, any additional overnight hospitalization within 30 days, and any drainage procedure within 30 days (eTable 1 in Supplement 1).²⁶ No additional institutional review board approval or informed consent were sought given that we reanalyzed already-existing data.

Prioritized Outcomes

We formed all possible pairs of patients, one from each arm of the CODA trial, to analyze sequentially each outcome of interest, starting with the outcome considered highest priority to the one considered lowest priority under different scenarios (eTable 2 in Supplement 1). Each pairwise comparison was evaluated starting from the outcome of highest priority and could result in 3 possible classifications: favorable or unfavorable to antibiotics when a difference was observed (eg, favorable to antibiotics if the patient from the antibiotic arm of a pair had a higher EQ-5D score) or neutral (when it was not possible to determine which patient within the pair had a better outcome). The latter situation could happen when (1) both patients presented the same values for the outcome of interest, (2) there was incomplete information for either patient (due to missing data), or (3) the difference did not reach a predefined threshold of clinical similarity (eg, pairwise comparisons on the EQ-5D score that were within a 0.05 absolute difference were considered neutral). For pairs that were classified as neutral for a given outcome, the pairwise comparisons were carried over to the next prioritized outcome. This sequence was repeated for all pairs until they were either classified as favorable to the antibiotic arm, unfavorable to the antibiotic arm, or all outcomes had been considered (Figure 1).

Figure 1.  Schematic View of the Multivariate Generalized Pairwise Comparison Analysis of Scenario 1

View LargeDownload

(opens in new tab)

Each pair is classified hierarchically through the list of outcomes. The pair is assessed on the first outcome, here the European Quality of Life–5 Dimensions (EQ-5D) score at 30 days. If the result favors or disfavors the treatment, the pair is classified accordingly. Here, 23.9% of all pairs favored antibiotics, while 17.0% of the pairs favored surgery. However, if the pair is neutral, the assessment is carried over to the subsequent outcome for further classification, here symptom resolution. Here 59.1% of all pairs were neutral for EQ-5D at 30 days and were therefore assessed by symptom resolution. The lines in the figure illustrate how neutral or tied matches are resolved at the next level of the hierarchy.

We considered 3 different scenarios among the many that could be envisaged, each related to a different order of priorities of outcomes (eTable 2 in Supplement 1). Scenario 1, which favors antibiotics, stems from responses to a survey on preference ranking provided by 443 of 3066 patients who accessed a decision support website.³² The preference ranking of these patients used ratings on a 3-point scale for 7 outcomes of appendicitis management they consider most relevant to their well-being: 1 indicated not important; 2, somewhat important; and 3, extremely important (eTable 3 in Supplement 1). Given our goal of illustrating how the GPC method might help future patients with different individual priorities from those arising from the survey, we chose scenarios 2 and 3 as plausible choices of clinically relevant orderings chosen to illustrate that the results could also be neutral (scenario 2) or favor the alternative treatment, ie, appendectomy (scenario 3).

Statistical Analysis

The method of GPC is an extension to multiple outcomes of the Mann-Whitney form of the nonparametric Wilcoxon test.¹¹ The results from all pairwise comparisons across all outcomes are aggregated in a single statistic, the NTB statistic, which is computed as the proportion of all favorable pairs minus the proportion of all unfavorable pairs. For a more complete description of this method, refer to the eAppendix in Supplement 1. This statistic estimates the NTB, which can be interpreted as the net probability that a randomly selected patient taken from the antibiotic-assigned arm would have a more favorable outcome than a patient taken randomly from the appendectomy-assigned arm, given a specific order of prioritized outcomes. As a net probability (ie, difference between 2 probabilities), NTB ranges from −1 to 1, with 0 indicating no difference between the 2 treatment groups. Univariate GPC analyses were performed to estimate NTB for each outcome individually, assessing their independent impact on the treatment effect. For multivariate analyses, multiplicity was tackled with sequential testing, starting with overall NTB and then proceeding sequentially until a nonsignificant P value (P > .05) was found. Statistical inference for the NTB was performed using the large-sample distribution of the GPC test statistic, which is a U statistic.³³ All analyses were run using the software package buysetest version 3.0 in R version 4.2.3 (R Project for Statistical Computing). The presence of an appendicolith, solid or calcified material inside the appendix seen on imaging, was used as a stratification factor in the CODA trial because of its association with complications.²⁶

Results

A total of 1552 patients from the CODA trial were included in this analysis. There were 776 (mean [SD] age, 38.3 [13.4] years; 286 [37%] female) in the antibiotic arm and 776 (mean [SD] age, 37.8 [13.7] years; 290 [37%] female) in the appendectomy arm.²⁶

Univariate Analyses

Results of the univariate analyses are depicted in the Table. GPC results were consistent with those from marginal statistical analyses presented in the original publication of the CODA trial.²⁶

Table.  Univariate NTB for Each Outcome Considered

View LargeDownload

(opens in new tab)

Outcome	Proportion of total pairs			NTB (95% CI)	P value
	Favor antibiotics	Favor surgery	Neutral
EQ-5D at 30 d	0.239	0.170	0.591	0.069 (0.026 to 0.112)	.002
Symptom resolution at 2 weeks	0.179	0.167	0.654	0.012 (−0.029 to 0.053)	.58
Any overnight hospitalization within 30 d	0.026	0.109	0.865	−0.083 (−0.107 to −0.059)	<.001
Any drainage procedure within 30 d	0.008	0.021	0.972	−0.013 (−0.024 to −0.002)	.02
Workdays missed by patient within 30 d	0.218	0.123	0.660	0.095 (0.066 to 0.125)	<.001
Workdays missed by caretaker within 30 d	0.158	0.101	0.742	0.057 (0.029 to 0.086)	<.001
Length of hospital stay	0.584	0.416	0.000	0.168 (0.106 to 0.228)	<.001

Multivariate Analyses

Figure 1 shows a schematic view of the multivariate GPC analysis for scenario 1. The overall NTB for the primary prioritization scheme, scenario 1, was positive and statistically significant in favor of antibiotic treatment (12.8%; 95% CI, 7.1%-18.3%; P < .001). Figure 2 shows the information proportion, which is the weight of each outcome in this analysis (sum of favorable plus unfavorable proportions for each outcome), while the individual contribution provides the contribution of each outcome to NTB (difference of favorable minus unfavorable proportions for each outcome). All outcomes contributed to increasing the NTB in favor of the antibiotic group, except any hospitalization within 30 days and any drainage procedure within 30 days, although the latter only classified a negligible proportion of all possible pairs (0.6%). More than half of the overall NTB (6.9% of 12.8%) was contributed by EQ-5D at 30 days.

Figure 2.  Multivariate Generalized Pairwise Comparison Analysis of Scenario 1

View LargeDownload

(opens in new tab)

Information proportion (IP) was calculated as the percentage of pairs favoring antibiotics plus those favoring surgery; information contribution (IC), percentage of pairs favoring antibiotics minus those favoring surgery; net treatment benefit (NTB), sum of ICs. The size of the squares is proportional to the cumulative pairs classified (CPC). EQ-5D indicates European Quality of Life–5 Dimensions.

Figure 3 shows that the overall NTB for scenario 2 was also positive but not significantly different from zero in favor of antibiotic treatment (3.2%; 95% CI, −2.4% to 8.7%; P = .27). Here, the 6.9% contribution to the NTB due to EQ-5D at 30 days was counterbalanced by negative contributions of any hospitalization within 30 days (−4.6%), length of hospital stay (−1.3%), and any drainage procedure within 30 days (−0.2%).

Figure 3.  Multivariate Generalized Pairwise Comparison Analysis of Scenario 2

View LargeDownload

(opens in new tab)

Information proportion (IP) was calculated as the percentage of pairs favoring antibiotics plus those favoring surgery; information contribution (IC), percentage of pairs favoring antibiotics minus those favoring surgery; net treatment benefit (NTB), sum of ICs. The size of the squares is proportional to the cumulative pairs classified (CPC). The size of the squares is proportional to the cumulative number of pairs classified. NC indicates not calculated. EQ-5D indicates European Quality of Life–5 Dimensions.

Finally, the overall NTB was negative and statistically significant in favor of appendectomy in scenario 3 (−14.6%, 95% CI, −20.2% to −8.8%; P < .001) (Figure 4). The major contributors to this negative NTB were length of hospital stay (−8.9%) and any hospitalization within 30 days (−8.3%). Of note, there were no more pairs for the last 3 outcomes in the hierarchy, as the first 4 outcomes classified all pairwise comparisons as favorable or unfavorable.

Figure 4.  Multivariate Generalized Pairwise Comparison Analysis of Scenario 3

View LargeDownload

(opens in new tab)

Information proportion (IP) was calculated as the percentage of pairs favoring antibiotics plus those favoring surgery; information contribution (IC), percentage of pairs favoring antibiotics minus those favoring surgery; net treatment benefit (NTB), sum of ICs. The size of the squares is proportional to the cumulative pairs classified (CPC). The size of the squares is proportional to the cumulative number of pairs classified. EQ-5D indicates European Quality of Life–5 Dimensions.

Discussion

The CODA trial investigated antibiotic therapy as an alternative to surgery for the treatment of appendicitis.²⁶ The trial contributes to the body of knowledge suggesting that antibiotics are noninferior to appendectomy,^25,26 despite remaining doubts about specific outcomes.²⁷ However, this type of conclusion is based on group-level data and typically ignores personal prioritization of outcomes. By analyzing data from the CODA trial using GPC, we have illustrated how a formal quantitative analysis of the benefit-risk relationship of antibiotics compared with surgery can provide a more exhaustive and nuanced picture of the differences between the 2 treatment modalities. This approach also takes individual priorities into account. The NTB, an absolute measure of the treatment effect, estimates the net probability that a patient taken randomly from the antibiotic group would have a better outcome than a patient taken randomly from the appendectomy group, given a certain order of priorities of outcomes.¹¹ It is worth emphasizing that the NTB is a trial-level treatment effect (although potentially based on the personal preferences of a specific patient). As such, it cannot be interpreted as the probability for a specific patient to do better after receiving the treatment condition than the control condition, which would be a causal individual-level treatment effect.³⁴ GPC allows for an exhaustive benefit-risk analysis by accounting for dependencies between the outcomes (ie, using conditional probabilities), rather than analyzing outcomes at the group level (ie, using marginal probabilities).¹⁵ Of note, thresholds of clinical similarity can be used to define a favorable pairwise comparison for a given outcome.

It is expected that patients are differentially impacted by the outcomes included in the analysis and the estimated NTB, based on their selected prioritization, reflects the overall benefit-risk that they would face by choosing one treatment over the other. The usefulness of individual prioritization rests on the assumption that individuals have a good understanding of the outcomes being considered. Thus, before conducting a ranking exercise with as part of SDM, patient training and discussions with a physician should take place, especially considering the large number of possibilities regarding prioritization. In the presence of 7 different outcomes, a total of 5040 permutations of outcomes are possible, each reflecting one order of priorities. For the present GPC analysis, we first considered the prioritized outcomes that were endorsed as most relevant in a survey of patients (scenario 1). The corresponding NTB for patients receiving antibiotic therapy, compared with patients undergoing surgery, was estimated at 12.8%, the difference between the probability of a better outcome for antibiotics (56.4%) and the probability of a better outcome for surgery (43.6%) using this order of priorities. As an additional interpretation, the inverse of the NTB can be translated to the number needed to treat. Here, approximately 8 patients (1 of 12.8% = 7.8 patients; 95% CI, 5.5-14.3 patients) would need to be treated on average for 1 patient to benefit from the antibiotic treatment.

Two other possible scenarios of outcomes ordering were used as illustrative sensitivity analyses. In scenario 2, raising drainage procedures and the length of hospital stay in the outcome hierarchy reduced the NTB to 3.2% and made it nonstatistically significant. Hence, for patients with this individualized prioritization of the outcomes, there is no overall statistical advantage of choosing one strategy over the other. Finally, scenario 3 placed hospitalizations within 30 days and drainage procedures as the outcomes of highest priority. With this prioritization, the NTB favored appendectomy, with a statistically significant negative NTB of −14.5% (corresponding to a number needed to treat of approximately 7 patients). Although these are only 3 of many possible scenarios, they reflect a range of variation that clearly shows that marginal comparisons between 2 interventions answer only part of a much more nuanced question about treatment benefit, one that can be explored in a rigorous and informative way using GPC.

Limitations

This study has limitations, including that this approach may not be consistent with the more nuanced way in which people actually make decisions, often in a gestalt-oriented manner that extends beyond rational definitions of priorities.³⁵ In addition, a prerequisite for the usefulness of GPC as a support to SDM would be willingness to undergo antibiotics or appendectomy. Such willingness is essential for SDM (and potential support from GPC) to be relevant.^29,36 A second limitation is that we could not include appendectomy as an outcome, since almost all patients in the appendectomy arm received surgery by design; nevertheless, salvage appendectomy is an important outcome for a patient choosing antibiotics, and our analysis could not take that outcome into account. Third, some of our results are heavily influenced by the EQ-5D; in scenario 1, approximately half of the net benefit (6.9% of 12.8%) from antibiotics was attributable to an EQ-5D higher by at least 0.05 at 30 days. The use of EQ-5D as a patient-reported outcome warrants caution, as its relevance may vary by clinical context. This also highlights a fourth limitation, of a more general nature, which is the fact that with GPC, the analysis of the first outcome is identical to a univariate analysis of this outcome, while the contributions of other outcomes are conditional on all previous outcomes being neutral. Similarly, the hierarchical nature of GPC may lead outcomes with lower probabilities of equivalence to disproportionately influence the analysis, potentially affecting interpretability of the overall NTB. Finally, it is worth noting that the application of GPC to indirect comparisons, like matching adjusted indirect comparisons for cross-trial analyses, generally require population adjustments which were not required in our reanalysis of the CODA trial.³⁷

Conclusions

In this benefit-risk analysis of antibiotics vs appendectomy, the NTB was used as a tool to collaboratively engage patients in SDM using statistical summaries tailored to their individual preferences. There is considerable interest in the development of support tools for SDM in uncomplicated appendicitis.^7,24,38 We surmise that the GPC methodology could contribute to decision support tools by incorporating a multivariate dimension to such tools. To fully explore the potential of adding GPC to decision support tools for SDM, work remains to be done toward communicating probabilities to patients, a key undertaking in medicine in general, and in the setting of appendicitis in particular.³⁹

Article Information

Accepted for Publication: January 28, 2025.

Published: March 31, 2025. doi:10.1001/jamanetworkopen.2025.2484

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2025 Salvaggio S et al. JAMA Network Open.

Corresponding Author: Samuel Salvaggio, PhD, One2Treat, 25 Bd Baudouin 1er, 1348 Louvain-la-Neuve, Belgium (samuel.salvaggio@one2treat.com).

Author Contributions: Dr Flum had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Salvaggio, Monsell, Heagerty, Chiem, Buyse, Flum.

Acquisition, analysis, or interpretation of data: Salvaggio, Monsell, Heagerty, De Backer, Barre, Saad, Flum.

Drafting of the manuscript: Salvaggio, Monsell, Saad, Buyse.

Critical review of the manuscript for important intellectual content: Salvaggio, Monsell, Heagerty, De Backer, Barre, Chiem, Flum.

Statistical analysis: Salvaggio, Monsell, Heagerty, De Backer, Barre, Chiem, Buyse.

Obtained funding: Heagerty, Flum.

Administrative, technical, or material support: Salvaggio, Monsell, Chiem, Flum.

Supervision: Salvaggio, Monsell, Heagerty, Chiem.

Conflict of Interest Disclosures: Dr Salvaggio reported receiving grants from BioWin during the conduct of the study and being an employee of One2Treat. Dr Monsell reported receiving grants from the Patient-Centered Outcomes Research Institute (PCORI) during the conduct of the study. Dr Heagerty reported receiving grants from the National Institutes of Health during the conduct of the study. Dr Saad reported having a patent for 18/653,133 pending and being an employee of IDDI during the conduct of the study. Dr Buyse reported stock ownership in IDDI and One2Treat during the conduct of the study and outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported in part by the Government of Wallonia, Belgium (BioWin Consortium Agreement No. 7979). The CODA Trial was funded by a PCORI Award (No. 1409-240099).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 2.

References

1. Elwyn  G, Frosch  D, Thomson  R,  et al.  Shared decision making: a model for clinical practice.   J Gen Intern Med. 2012;27(10):1361-1367. doi:10.1007/s11606-012-2077-6PubMedGoogle ScholarCrossref

2. Mitropoulou  P, Grüner-Hegge  N, Reinhold  J, Papadopoulou  C.  Shared decision making in cardiology: a systematic review and meta-analysis.   Heart. 2022;109(1):34-39. doi:10.1136/heartjnl-2022-321050PubMedGoogle ScholarCrossref

3. Niburski  K, Guadagno  E, Abbasgholizadeh-Rahimi  S, Poenaru  D.  Shared decision making in surgery: a meta-analysis of existing literature.   Patient. 2020;13(6):667-681. doi:10.1007/s40271-020-00443-6PubMedGoogle ScholarCrossref

4. Niburski  K, Guadagno  E, Mohtashami  S, Poenaru  D.  Shared decision making in surgery: a scoping review of the literature.   Health Expect. 2020;23(5):1241-1249. doi:10.1111/hex.13105PubMedGoogle ScholarCrossref

5. Stacey  D, Volk  RJ, IPDAS Evidence Update Leads (Hilary Bekker, Karina Dahl Steffensen, Tammy C. Hoffmann, Kirsten McCaffery, Rachel Thompson, Richard Thomson, Lyndal Trevena, Trudy van der Weijden, and Holly Witteman).  The International Patient Decision Aid Standards (IPDAS) Collaboration: evidence update 2.0.   Med Decis Making. 2021;41(7):729-733. doi:10.1177/0272989X211035681PubMedGoogle ScholarCrossref

6. Agoritsas  T, Heen  AF, Brandt  L,  et al.  Decision aids that really promote shared decision making: the pace quickens.   BMJ. 2015;350:g7624. doi:10.1136/bmj.g7624PubMedGoogle ScholarCrossref

7. Rosen  JE, Flum  DR, Davidson  GH, Liao  JM.  Randomized pilot test of a decision support tool for acute appendicitis: decisional conflict and acceptability in a healthy population.   Ann Surg Open. 2022;3(4):e213. doi:10.1097/AS9.0000000000000213PubMedGoogle ScholarCrossref

8. Soekhai  V, Whichello  C, Levitan  B,  et al.  Methods for exploring and eliciting patient preferences in the medical product lifecycle: a literature review.   Drug Discov Today. 2019;24(7):1324-1331. doi:10.1016/j.drudis.2019.05.001PubMedGoogle ScholarCrossref

9. Thokala  P, Devlin  N, Marsh  K,  et al.  Multiple criteria decision analysis for health care decision making–an introduction: report 1 of the ISPOR MCDA Emerging Good Practices Task Force.   Value Health. 2016;19(1):1-13. doi:10.1016/j.jval.2015.12.003PubMedGoogle ScholarCrossref

10. Dodgson  JS. Multi-criteria analysis: a manual. Department for Communities and Local Government. January 2009. Accessed December 24, 2024. https://assets.publishing.service.gov.uk/media/5a790545e5274a2acd18b975/1132618.pdf

11. Buyse  M.  Generalized pairwise comparisons of prioritized outcomes in the two-sample problem.   Stat Med. 2010;29(30):3245-3257. doi:10.1002/sim.3923PubMedGoogle ScholarCrossref

12. Evans  SR, Rubin  D, Follmann  D,  et al.  Desirability of outcome ranking (DOOR) and response adjusted for duration of antibiotic risk (RADAR).   Clin Infect Dis. 2015;61(5):800-806. doi:10.1093/cid/civ495PubMedGoogle ScholarCrossref

13. Pocock  SJ, Ariti  CA, Collier  TJ, Wang  D.  The win ratio: a new approach to the analysis of composite endpoints in clinical trials based on clinical priorities.   Eur Heart J. 2012;33(2):176-182. doi:10.1093/eurheartj/ehr352PubMedGoogle ScholarCrossref

14. Butler  J, Stockbridge  N, Packer  M.  Win ratio: a seductive but potentially misleading method for evaluating evidence from clinical trials.   Circulation. 2024;149(20):1546-1548. doi:10.1161/CIRCULATIONAHA.123.067786PubMedGoogle ScholarCrossref

15. Buyse  M, Saad  ED, Peron  J,  et al.  The net benefit of a treatment should take the correlation between benefits and harms into account.   J Clin Epidemiol. 2021;137:148-158. doi:10.1016/j.jclinepi.2021.03.018PubMedGoogle ScholarCrossref

16. Maurer  MS, Schwartz  JH, Gundapaneni  B,  et al; ATTR-ACT Study Investigators.  Tafamidis treatment for patients with transthyretin amyloid cardiomyopathy.   N Engl J Med. 2018;379(11):1007-1016. doi:10.1056/NEJMoa1805689PubMedGoogle ScholarCrossref

17. Voors  AA, Angermann  CE, Teerlink  JR,  et al.  The SGLT2 inhibitor empagliflozin in patients hospitalized for acute heart failure: a multinational randomized trial.   Nat Med. 2022;28(3):568-574. doi:10.1038/s41591-021-01659-1PubMedGoogle Scholar

18. Berwanger  O, Pfeffer  M, Claggett  B,  et al.  Sacubitril/valsartan versus ramipril for patients with acute myocardial infarction: win-ratio analysis of the PARADISE-MI trial.   Eur J Heart Fail. 2022;24(10):1918-1927. doi:10.1002/ejhf.2663PubMedGoogle Scholar

19. Finkelstein  DM, Schoenfeld  DA.  Combining mortality and longitudinal measures in clinical trials.   Stat Med. 1999;18(11):1341-1354. doi:10.1002/(SICI)1097-0258(19990615)18:11<1341::AID-SIM129>3.0.CO;2-7PubMedGoogle Scholar

20. Bodemer  N, Meder  B, Gigerenzer  G.  Communicating relative risk changes with baseline risk: presentation format and numeracy matter.   Med Decis Making. 2014;34(5):615-626. doi:10.1177/0272989X14526305PubMedGoogle Scholar

21. Péron  J, Lambert  A, Munier  S,  et al.  Assessing long-term survival benefits of immune checkpoint inhibitors using the net survival benefit.   J Natl Cancer Inst. 2019;111(11):1186-1191. doi:10.1093/jnci/djz030PubMedGoogle Scholar

22. Péron  J, Roy  P, Ozenne  B, Roche  L, Buyse  M.  The net chance of a longer survival as a patient-oriented measure of treatment benefit in randomized clinical trials.   JAMA Oncol. 2016;2(7):901-905. doi:10.1001/jamaoncol.2015.6359
ArticlePubMedGoogle Scholar

23. Deltuvaite-Thomas  V, De Backer  M, Parker  S,  et al.  Generalized pairwise comparisons of prioritized outcomes are a powerful and patient-centric analysis of multi-domain scores.   Orphanet J Rare Dis. 2023;18(1):321. doi:10.1186/s13023-023-02943-8PubMedGoogle Scholar

24. Patterson  KN, Deans  KJ, Minneci  PC.  Shared decision-making in pediatric surgery: an overview of its application for the treatment of uncomplicated appendicitis.   J Pediatr Surg. 2023;58(4):729-734. doi:10.1016/j.jpedsurg.2022.10.009PubMedGoogle Scholar

25. Sallinen  V, Akl  EA, You  JJ,  et al.  Meta-analysis of antibiotics versus appendicectomy for non-perforated acute appendicitis.   Br J Surg. 2016;103(6):656-667. doi:10.1002/bjs.10147PubMedGoogle Scholar

26. Flum  DR, Davidson  GH, Monsell  SE,  et al; CODA Collaborative.  A randomized trial comparing antibiotics with appendectomy for appendicitis.   N Engl J Med. 2020;383(20):1907-1919. doi:10.1056/NEJMoa2014320PubMedGoogle Scholar

27. Herrod  PJJ, Kwok  AT, Lobo  DN.  Randomized clinical trials comparing antibiotic therapy with appendicectomy for uncomplicated acute appendicitis: meta-analysis.   BJS Open. 2022;6(4):6. doi:10.1093/bjsopen/zrac100PubMedGoogle Scholar

28. Knaapen  M, de Wind  A, van der Lee  JH,  et al.  Implementing nonoperative treatment strategy for simple pediatric appendicitis: a qualitative study.   J Surg Res. 2022;279:218-227. doi:10.1016/j.jss.2022.06.011PubMedGoogle Scholar

29. Hanson  AL, Crosby  RD, Basson  MD.  Patient preferences for surgery or antibiotics for the treatment of acute appendicitis.   JAMA Surg. 2018;153(5):471-478. doi:10.1001/jamasurg.2017.5310
ArticlePubMedGoogle Scholar

30. EuroQol Group.  EuroQol—a new facility for the measurement of health-related quality of life.   Health Policy. 1990;16(3):199-208. doi:10.1016/0168-8510(90)90421-9PubMedGoogle Scholar

31. Le  QA, Doctor  JN, Zoellner  LA, Feeny  NC.  Minimal clinically important differences for the EQ-5D and QWB-SA in post-traumatic stress disorder (PTSD): results from a doubly randomized preference trial (DRPT).   Health Qual Life Outcomes. 2013;11:59. doi:10.1186/1477-7525-11-59PubMedGoogle Scholar

32. Appendicitis treatment options. Accessed February 19, 2025. https://appyornot.org/

33. Bebu  I, Lachin  JM.  Large sample inference for a win ratio analysis of a composite outcome based on prioritized components.   Biostatistics. 2016;17(1):178-187. doi:10.1093/biostatistics/kxv032PubMedGoogle Scholar

34. Fay  MP, Brittain  EH, Shih  JH, Follmann  DA, Gabriel  EE.  Causal estimands and confidence intervals associated with Wilcoxon-Mann-Whitney tests in randomized experiments.   Stat Med. 2018;37(20):2923-2937. doi:10.1002/sim.7799PubMedGoogle Scholar

35. Gladwell  M.  Blink: The Power of Thinking Without Thinking. Black Bay Books; 2005.

36. Rosen  JE, Agrawal  N, Flum  DR, Liao  JM.  Willingness to undergo antibiotic treatment of acute appendicitis based on risk of treatment failure.   Br J Surg. 2021;108(11):e361-e363. doi:10.1093/bjs/znab280PubMedGoogle Scholar

37. Signorovitch  JE, Sikirica  V, Erder  MH,  et al.  Matching-adjusted indirect comparisons: a new tool for timely comparative effectiveness research.   Value Health. 2012;15(6):940-947. doi:10.1016/j.jval.2012.05.004PubMedGoogle Scholar

38. Minneci  PC, Cooper  JN, Leonhart  K,  et al.  Effects of a patient activation tool on decision making between surgery and nonoperative management for pediatric appendicitis: a randomized clinical trial.   JAMA Netw Open. 2019;2(6):e195009. doi:10.1001/jamanetworkopen.2019.5009
ArticlePubMedGoogle Scholar

39. Rosen  JE, Agrawal  N, Flum  DR, Liao  JM.  Verbal descriptions of the probability of treatment complications lead to high variability in risk perceptions: a survey study.   Ann Surg. 2023;277(4):e766-e771. doi:10.1097/SLA.0000000000005255PubMedGoogle Scholar

Original content published on ACRPnet.org here, written by Rudradev Sengupta.

For many early-career clinical researchers, trial design can feel rigid. A primary endpoint is selected, the study is powered around it, and success is determined by that single measure. But patients do not experience treatment effects through one outcome. They experience a combination of efficacy, side effects, symptoms, and quality of life. Reducing this to a single outcome can risk missing the bigger picture. A more patient-centered approach, when integrated into trial design, might help to better capture this complexity, particularly from Phase II onward when key development decisions are made.

Moving beyond single outcome starts with how trials are designed. A single measure of efficacy is typically considered the most important outcome and determines success or failure. Other outcomes are collected but often have limited influence on conclusions. This can create a disconnect between what the study aims to show and what patients really experience.

One treatment may improve efficacy but come with meaningful toxicity, while another may offer similar efficacy with better tolerability or quality of life. These differences matter to patients but are not always reflected in traditional analyses.

A patient-centric trial begins during the design of the trial. Instead of asking what the single best outcome is, a more relevant question is: What combination of outcomes reflects how patients judge benefit? This leads to multidimensional endpoints, where several outcomes are considered together.

Designing these endpoints requires prioritization. There is usually one measure of efficacy in a traditional primary endpoint that is considered the most important outcome, but what follows depends on the population. Some patients may prioritize other efficacy outcomes whereas others may value minimizing severe side effects or a better quality of life.

Patients, clinicians, investigators, and patient advocates can all contribute to defining which outcomes matter the most and in what order. This prioritization becomes central to how treatment benefit is evaluated.

Understanding Net Treatment Benefit

The Net Treatment Benefit (NTB) framework provides a practical way to analyze these multidimensional endpoints. It compares all possible pairs of patients taken from the treatment and the control group across a prioritized list of outcomes. Each comparison starts with the most important outcome. If one patient does better, the comparison stops. If not, it moves to the next outcome. This continues until a patient pair is classified or all outcomes are considered.

This process is repeated across all possible patient pairs. For example, with 10 patients in each group, 100 pairwise comparisons are performed. The result is a single metric that reflects how often patients on the new treatment do better overall, considering all prioritized outcomes. The key idea is simple: NTB allows for capturing trade-offs between benefits and risks in a way that mirrors real clinical decision-making.

This links directly to the concept of “totality of the evidence.” Trials collect large amounts of clinically meaningful data, but much of it is underused when only one endpoint drives the analysis.

The NTB approach integrates multiple outcomes into a single assessment of treatment benefit. For patients, this better reflects real experience. For sponsors, it improves the ability to detect meaningful effects, which is particularly important in Phase II, where decisions are critical for future development. For researchers, it provides a clearer and more complete understanding of treatment impact.

From Phase II onward, this approach supports better decision-making. Early trials often involve small sample sizes and difficult choices around dose selection or progression to Phase III. Evaluating multiple prioritized outcomes provides a more balanced view of benefit and risk, reducing the chance of advancing ineffective treatments or stopping promising ones too early.

By Phase III, this framework creates continuity. The same prioritized outcomes can be carried forward—either as the primary endpoint or as a key secondary endpoint—ensuring consistency from early development through to confirmatory trials.

Far-Reaching Impacts of NTB

The impacts of NTB extend beyond clinical development. Regulators are increasingly focused on benefit-risk assessment and patient-relevant outcomes. A multidimensional approach supported by NTB offers a structured way to present this information.

Health technology assessment bodies and payers also require evidence of value beyond efficacy. This includes quality of life, safety, and broader impact on healthcare systems. By summarizing multiple outcomes into a single measure, NTB aligns well with these expectations and strengthens the overall evidence package.

Examples from cardiovascular research show that this approach is already established. In transthyretin amyloid cardiomyopathy, treatment benefit has been assessed by prioritizing survival followed by hospitalization, reflecting outcomes that matter the most to patients and clinicians. This type of analysis has already been used to support regulatory decision-making and demonstrate the value of integrating multiple endpoints into a single framework.{1,2}

A similar approach is valuable in more complex populations. In elderly patients with rectal cancer, treatment decisions often involve balancing efficacy with tolerability and quality of life. A single endpoint does not capture these trade-offs. Evaluating outcomes such as survival, disease progression, and severe toxicity together provides a more realistic assessment of treatment benefit. This has been applied in trials designed for older rectal cancer patients, where maintaining quality of life could be considered by the patients as important as extending survival.{3}

Conclusion

For those starting in clinical research, the takeaway is straightforward. Trial design is not just about selecting a statistically acceptable endpoint; it is about generating evidence that reflects how treatments are experienced and valued in real life.

Multidimensional endpoints and NTB offer a practical way to do this. They allow patient preferences to be incorporated, make better use of collected data, and create a consistent strategy from Phase II through approval and market access. This approach does not complicate trials unnecessarily; it makes them more relevant, more informative, and better aligned with the needs of patients and the broader healthcare system.

References

1. Maurer MS, Schwartz JH, Gundapaneni B, Elliott PM, Merlini G, Waddington-Cruz M, … Rapezzi C. 2018. Tafamidis treatment for patients with transthyretin amyloid cardiomyopathy. New England Journal of Medicine 379(11):1007–16.
2. Gillmore JD, Judge DP, Cappelli F, Fontana M, Garcia-Pavia P, Gibbs S, … Fox JC. 2024. Efficacy and safety of acoramidis in transthyretin amyloid cardiomyopathy. New England Journal of Medicine390(2):132–42.
3. Saúde-Conde R, Vandamme T, De Backer M, Martinive P, Covas A, Deleporte A, … Sclafani F. 2024. Efficacy and safety of short-course radiotherapy versus total neoadjuvant therapy in older rectal cancer patients: a randomised pragmatic trial (SHAPERS). ESMO gastrointestinal oncology 4:100067.

OSE Immunotherapeutics, a biotechnology company that develops immunotherapies for cancer and autoimmune diseases and One2Treat, a fast-growing technology company transforming how the patient voice is integrated into clinical development, have won the 2025 Clinical Trials Arena Excellence Award for Innovation in Endpoint Design for a practical idea: involve investigators upfront to define what matters most, then use those priorities to shape a multidimensional endpoint.

In ARTEMIA—a global Phase III trial comparing OSE2101 (Tedopi®), a neoepitope-based cancer vaccine, with docetaxel in HLA-A2 positive metastatic non-small cell lung cancer. OSE Immunotherapeutics partnered with One2Treat and used the One2Treat Voice platform to collect preferences quickly and consistently. In March 2025, 29 investigators completed a short, secure exercise judging simulated patient cases and identifying which outcomes should carry the most weight. Those inputs were translated into a pre-specified exploratory analysis that combines overall survival, quality of life, disease control, and safety into a single, interpretable measure of the Net Treatment Benefit.

Bringing real-world clinical judgment into endpoint selection

The innovation lies in making clinical judgment part of clinical trial design, rather than an afterthought. Investigators viewed anonymized, simulated patient pairs and chose who was better overall, noting which outcome drove each choice. This simple format captured the real trade-offs clinicians make every day and turned it into structured data tied to the protocol’s outcomes.

Speed and ease strengthened the result. In one week in March 2025, 29 investigators completed the exercise in about nine minutes each via a secure link.

The group settled on a clear priority order: overall survival (OS), EORTC QLQ-C30 global quality of life (QoL), progression-free survival (PFS), side-effect burden, and serious adverse events (SAEs). OS was selected as the top outcome in 72% of choices. The platform also captured practical thresholds that make the differences meaningful: three months for OS, one level for QoL, and six months for PFS. Physical and role functioning were considered but did not make the final five, keeping the focus on the outcomes investigators judged most relevant for decisions.

By moving from informal feedback to a transparent, reproducible process, the collaboration gave endpoint definition a clearer clinical foundation.

A multidimensional, interpretable measure of treatment impact

ARTEMIA keeps overall survival as the primary endpoint. The added innovation is a pre-specified exploratory analysis—Net Treatment Benefit (NTB)—that combines multiple outcomes into one readout. Using Generalized Pairwise Comparisons, each patient on Tedopi® is compared with each patient on docetaxel, following the investigator-defined order and thresholds. If one patient lives at least three months longer, that counts toward survival; if not, the comparison moves to global QoL, then to PFS, and, if needed, to side-effect burden and SAEs.

This method produces a single, straightforward probability: the chance that a randomly selected patient on Tedopi® has a better overall outcome than a randomly selected patient on docetaxel, based on the agreed priorities. It makes trade-offs visible and avoids piecemeal interpretation across separate endpoints. For decision-makers, it offers a pre-specified, transparent way to understand overall impact. The clarity and rigor of this multi-outcome assessment were key factors behind the award.

Speed, scale, and a replicable framework with minimal operational burden

The One2Treat Voice approach worked because it was rapid, lightweight, and scalable. Investigators completed the One2Treat Voice exercise in minutes, and results were compiled in days, leaving trial operations uninterrupted. The adaptive questionnaire and simulated cases reduced burden while still capturing nuanced clinical preferences.

The process is also repeatable. It is digital, adaptive, and outcome-agnostic, so it can be applied in other studies where considering the overall treatment effect across several outcomes is important. The steps are clear: elicit preferences, set meaningful thresholds, and link them directly to a predefined NTB analysis. That end-to-end pathway turns assumptions into explicit design inputs and provides a template others can follow. The ability to scale this method without adding complexity was another important reason for the recognition.

“Collaborating with OSE Immunotherapeutics on defining a prioritized list of outcomes for this innovative endpoint has been a meaningful step forward. By grounding multi-dimensional endpoint design in the perspectives of investigators, we’re helping ensure that trial design reflects what really matters for both patients and clinical decision-makers.”

Sebastien Coppe, CEO, One2Treat

About OSE Immunotherapeutics

OSE Immunotherapeutics is an integrated biotechnology company focused on developing and partnering therapies to control the immune system for immuno-oncology and immuno-inflammation.

About One2Treat

One2Treat is dedicated to transforming the way biopharmaceutical companies design, analyze and interpret randomized clinical trials, and how the overall medical value of a treatment may be communicated. Through its innovative software platform and patient-centered methodologies, One2Treat helps sponsors incorporate the patient’s voice into early trial design, endpoint selection, and treatment value assessment. By aligning clinical evidence generation with patients and clinicians’ needs, One2Treat supports more meaningful clinical trials, efficient decision making and enhanced regulatory discussions, leading to faster market access.

Download the PDF poster here.

Background

ARTEMIA is a phase III trial comparing OSE2101, a cancer vaccine, to docetaxel in HLA-A2-positive patients with metastatic NSCLC and secondary resistance to Immune Checkpoint Inhibitor (ICI), with overall survival (OS) as primary endpoint. To consider additional outcomes in a single analysis, an exploratory endpoint estimating the Net Treatment Benefit (NTB) of OSE2101 versus docetaxel is planned. Hereby we report the preferred outcomes and their priority order appraised by the investigators for the NTB analysis.

Methods

The elicitation software One2Treat Voice captured which outcomes were prioritized by the investigators, their relative importance and meaningful thresholds. Investigator could choose up to 5 out of 7 outcomes selected from the study’s main endpoints: OS, Progression-Free survival (PFS), QLQ-C30 Global Quality of Life (QoL), physical QoL, role QoL, side-effect burden QoL, and serious adverse events (SAEs) occurrence. The software used an adaptive algorithm to present simulated pairs, distinguishing between two treatments without disclosing investigational products. For each pair, investigators selected which patient had the better overall situation. Preferences were inferred and aggregated to inform the design of the NTB analysis. Participation was voluntary and anonymized.

Results

From March 17 to 24, 2025, investigators (N=29) completed the questionnaire in an average of 9 minutes [min: 4, max: 24]. Five outcomes were selected by investigators, with overall survival (OS) preferred by 72% of investigators, followed by Global QoL, PFS, side effects burden, and SAEs. The corresponding thresholds were 3 months for OS, 6 months for PFS, and 1-level improvement for QoL and side effects.

Conclusions

The One2Treat Voice software enabled rapid collection of investigator preferences in the OSE ARTEMIA trial for relevant outcomes, their prioritization, and the corresponding thresholds. This approach ensures that the planned Net Treatment Benefit analysis is fully aligned with clinical practice and actively involves investigators in the trial design process.

Clinical trial identification

NCT06472245.

Legal entity responsible for the study

OSE Immunotherapeutics SA.

Louvain-la-Neuve, Belgium — 4^th September 2025 

One2Treat SA, a fast-growing technology company transforming how the patient voice is integrated into clinical development, announced today the successful completion of a planned seed extension to accelerate its software platform development.

In parallel, One2Treat has relocated to a larger office in Louvain-la-Neuve to support its continued growth.

Shareholder confidence accelerates One2Treat’s investment in software platform development.

The conclusion of One2Treat’s planned seed extension funding round marks a key milestone in the company’s growth. This investment follows the successful deployment of the One2Treat Insights^® module across multiple live projects and the recent release of the One2Treat Voice^® module. It reflects the confidence of initial shareholders in One2Treat’s unique approach to answer the growing needs of the biopharma industry to both increase productivity and better reflect what matters to patients and clinicians.

The funding supports the continued development and scaling of One2Treat’s cloud-based software platform. This platform integrates diverse patient-focused outcomes into a single, comprehensive assessment of treatment effects, enabling a clearer understanding of the Net Treatment Benefit. The platform answers strategic needs for clinical development decisions, but also supports the overall medical value of a new treatment during regulatory discussions, HTA and commercialization.

“Based on multiple recent successes, our shareholders were keen to accelerate the investment in our platform development.  In addition, we offered every employee the opportunity to become a shareholder, and all chose to take part, highlighting a deep, shared commitment to our mission of putting patient voices at the center of clinical research.”

— Sébastien Coppe, CEO, One2Treat

Move to larger office supports One2Treat’s planned team expansion

Alongside the capital increase, One2Treat has moved to a new office in Louvain-la-Neuve, a dynamic academic and tech ecosystem just outside Brussels.

“The new space provides a larger, more modern and collaborative environment for the expanding team, fostering innovation and facilitating closer partnerships with key academic and industry stakeholders.”

• Marc Buyse, Founder One2Treat

-END-

About One2Treat

One2Treat is dedicated to transforming the way biopharmaceutical companies design, analyze and interpret randomized clinical trials, and how the overall medical value of a treatment may be communicated. Through its innovative software platform and patient-centered methodologies, One2Treat helps sponsors incorporate the patient’s voice into early trial design, endpoint selection, and treatment value assessment. By aligning clinical evidence generation with patients and clinicians’ needs, One2Treat supports more meaningful clinical trials, efficient decision making and enhanced regulatory discussions, leading to faster market access.

For more information about One2Treat and its innovative approach to clinical trial design and market access, please visit: www.one2treat.com.

Media Contact

Tom Mann 

Clinical Solutions Engagement Lead, One2Treat

Tom.Mann@one2treat.com

Original interview with ARCP for Clinical Trials Day 2025 available here. Author: Mathilde Tournay

My clinical research power is…

…bridging the gap between statistical rigor and clinical development. As a biostatistician, I bring deep expertise in data and complex statistical methods together with a solid understanding of science and medicine to ensure that clinical evidence is not just methodologically sound but also meaningful. My goal is to transform data into actionable insights that truly support better decisions for patients and clinicians alike.

The greatest challenge I see to clinical research right now is…

…designing trials that can manage growing complexity while staying focused on patient needs. As trials evolve to include more data, endpoints, and subgroups, it becomes harder to reflect the diverse preferences of the people they’re meant to serve. Too often, key outcomes are chosen without patient input, leading to evidence that misses the mark on real-life impact.

The greatest opportunity I see for clinical research in the near future is…

 …embedding patient preferences very early into the design of trials—starting with how outcomes are selected and prioritized. Tools that enable this shift will drive more meaningful evidence generation, reduce trial inefficiencies, and accelerate the delivery of treatments that patients truly value. We must do better at integrating patient priorities from the outset and making full use of the data we collect to ensure research reflects what matters most to those living with the condition.

Original content written for Discover Pharma, available here. Author: Pascal Piedbois.

As oncology research continues to evolve, the focus is increasingly shifting from traditional endpoints to patient-centered outcomes. Ahead of ASCO 2025, Pascal Piedbois, discusses how methodologies like Generalized Pairwise Comparisons (GPC) are transforming clinical trials to better reflect the realities of patient care. In this insightful Q&A, he shares his expectations for ASCO, explains the practical impact of GPC, and outlines how One2Treat is helping sponsors build more meaningful and efficient oncology studies.

What are you most looking forward to at ASCO 2025, and how do you see it shaping the future of oncology research?

I am most looking forward to seeing the latest advancements in personalized medicine and immunotherapy. But more than that, I expect ASCO 2025 to highlight a growing emphasis on comprehensive approaches that could revolutionize oncology research strategy. Approaches that not only aim to extend life but also respect patient priorities and preferences. The future of oncology lies in combining clinical efficacy with a deeper understanding of what matters most to patients. That mindset is starting to shape how we design trials, evaluate benefits, and ultimately guide treatment decisions.

As One2Treat attends ASCO, are there any particular scientific trends or sessions you’re especially excited about?

I’d like to attend sessions that go beyond the science of new treatments to address how we design research that truly reflects patient needs. There is growing interest in understanding patient preferences and in translating those insights into concrete changes in trial design, whether it’s how we define meaningful benefit, how we prioritize outcomes, or how we engage patients early in the protocol development process. ASCO is increasingly becoming a forum not just for drug innovation, but for rethinking how we include the patient voice in the way evidence is generated.

Traditional oncology trials often focus on tumor shrinkage and survival metrics. Why do you believe it’s time for a shift toward more patient-centric approaches like Generalized Pairwise Comparisons (GPC)?

As medical oncologists, we rely on endpoints like tumor response and overall survival for good reason. They’re critical indicators of efficacy. But patients experience treatment through a broader lens: side effects, daily functioning, and long-term quality of life are often just as impactful as how long they live. What’s needed isn’t a replacement of our traditional endpoints, but an evolution, one that allows us to assess treatment benefit more holistically.

Generalized Pairwise Comparisons (GPC) supports that evolution. It respects the hierarchy of what matters most in oncology — often survival first — while also making room for additional outcomes that reflect the reality of patient experience. It’s a way to enrich our evidence without compromising scientific rigor.

Can you explain, in simple terms, how GPC works and what makes it more reflective of patient priorities like safety and quality of life?

GPC, or Generalized Pairwise Comparisons is a statistical method that allows us to compare treatments across several outcomes, in a way that mirrors clinical decision-making.

At its core, GPC works by forming all possible pairs between patients in the treatment group and those in the control group. For example, if there are 10 patients in each group, we evaluate 100 unique comparisons. For each pair, we assess who had the better outcome based on a ranked list of priorities — typically starting with survival, followed by quality of life, side effects, and so on. If one patient clearly does better on the most important outcome, that pair is counted accordingly. If not, we move down the list.

What makes GPC powerful is that it doesn’t force us to choose a single primary endpoint, it allows us to integrate several outcomes, based on their clinical relevance. The result is a single, interpretable measure called Net Treatment Benefit (NTB), which is simply the net difference between the probability that a random patient in the treatment group has a better outcome than a random patient in the control group, and the probability of the opposite occurring. This kind of quantitative analysis gives a more complete picture of treatment impact, one that aligns with how patients and clinicians weigh trade-offs in real life.

How receptive are regulators and sponsors to using GPC in real-world trials — and what progress have you seen?

I think that regulators are increasingly receptive to methodologies like GPC, especially when they help capture clinical benefit in a way that reflects the totality of patient experience. What matters most to agencies like the EMA or FDA isn’t whether a method is novel, but whether it is clearly pre-specified, well-justified, and clinically interpretable. GPC has already been included in several regulatory submissions, often as a key secondary or exploratory analysis, and fits well within current initiatives focused on patient-centric evidence generation — such as FDA’s Project Patient Voice or Project Optimus.

On the sponsor side, we’ve seen growing adoption, particularly in trials where traditional endpoints alone may not tell the full story — for example, in rare diseases or complex oncology settings. What’s shifting is the mindset: the question is no longer whether we should consider patient-centered methodologies, but how we can implement them in a rigorous, operationally feasible way. GPC offers a structured framework to do just that.

What are some of the key challenges to adopting alternative trial methodologies in oncology?

One of the biggest challenges is cultural. As clinicians and researchers, we’re trained to think in terms of traditional endpoints — overall survival, progression-free survival, objective response — and those remain essential. But broadening our view to include prioritized, multi-dimensional outcomes requires a shift in mindset. It means rethinking what we consider “primary” and accepting that benefit can’t always be reduced to a single number or curve.

Another key challenge lies in the practical setup. For methods like GPC to work, we need to define a clear hierarchy of outcomes, and that’s not always straightforward. It requires early, structured input from both clinicians and patients to understand which outcomes truly matter and in what order. These preferences need to be captured before the trial begins, and doing so in a consistent, rigorous way across sites and stakeholders takes time, planning, and the right tools.

Finally, there’s an educational gap. Many clinicians are unfamiliar with these approaches, and sponsors may hesitate to deviate from conventional designs unless there’s a clear regulatory or strategic incentive. Overcoming that inertia takes time, but it starts with building confidence in methods that are both scientifically rigorous and clinically meaningful.

How does One2Treat help companies overcome those barriers — technically, scientifically, or culturally?

At One2Treat, we support sponsors in making early strategic decisions that place patient and clinician priorities at the center of trial design. A recurring challenge is identifying which outcomes to measure and how to prioritize them. To address this, we’ve developed One2Treat Voice, a software solution that facilitates the structured collection of preferences from key opinion leaders, investigators and patient advocates, enabling prioritization that’s evidence-based and operationally feasible.

This is part of a broader suite of tools we’re building to support modern trial design. One2Treat Insights helps sponsors to leverage the totality of evidence from prior trials to inform endpoint selection and data interpretation, while One2Treat Design enables simulation-based planning and sample size calculation to complement methodological guidance. Alongside these tools, we provide scientific and strategic advice to ensure that trial designs incorporate a more comprehensive and clinically relevant estimation of the Net Treatment Benefit, helping ensure that trial outcomes better reflect patient priorities, making the resulting evidence more clinically meaningful.

Beyond trial interpretation, NTB can also inform early R&D go/no go decisions by quantifying benefit-risk trade-offs across multiple outcomes, helping teams prioritize compounds with the most patient-relevant impact. In HTA contexts, NTB offers a transparent, interpretable metric that aligns with payer expectations for comprehensive value assessment.

Our goal is to make the adoption of patient-centered methodologies not just possible, but practical, by combining robust tools with the right scientific framework early in the trial design stage.

Could you share any case studies or examples where GPC made a significant impact on trial outcomes or interpretation?

GPC has already been used to provide a much richer understanding of treatment effects, particularly in complex clinical scenarios where relying on a single primary endpoint would have fallen short. In these situations, where multiple clinical outcomes may play a decisive role, GPC enables estimating a quantitative benefit risk balance to support decision-making.

This was exactly the challenge addressed in the SHAPERS trial for older patients with rectal cancer. The investigators needed to evaluate whether a shorter, less intensive treatment could offer a better overall balance of benefit and harm compared to total neoadjuvant therapy. Rather than using a traditional non-inferiority design, which would have required a large sample size to detect equivalence based on a single endpoint, the team leveraged GPC to estimate a Net Treatment Benefit based on multiple prioritized outcomes: survival, tumor progression, neuropathy, and toxicity. This approach allowed the trial to test for superiority, not just equivalence, while also ensuring that the primary analysis reflected trade-offs meaningful to patients. It offered a more efficient, clinically relevant framework that traditional endpoints alone could not provide

Another compelling use of GPC comes from re-analyses of existing trials. In a JAMA Oncology study, Péron et al. showed that the Net Treatment Benefit provided clearer, more intuitive insight than hazard ratios, especially when survival curves crossed or treatment effects emerged late. By reframing the question as “What’s my chance of living meaningfully longer with treatment?”, GPC helped uncover clinically relevant benefits that standard analyses often obscure. This approach allows us to extract more patient-relevant evidence from data we already have.

Finally, we’re now exploring how GPC can support dose selection in Phase II clinical trials, in alignment with FDA’s Project Optimus. By comparing multiple doses not just on efficacy but also on tolerability and quality of life, NTB offers quantitative evidence towards the most balanced treatment to support discussions with regulatory bodies.

Where do you see the biggest opportunities for improving oncology trials over the next five years?

We have a tremendous opportunity to rethink how we define and measure success in oncology trials. Over the next five years, I believe the biggest impact will come from integrating patient-centered methodologies into earlier stages of trial design, not just as an afterthought, but as a core design principle. By embedding structured patient and clinician input into endpoint selection and benefit-risk balance evaluation, we can generate evidence that is not only scientifically rigorous but also clinically meaningful.

Another key area is the smarter use of prior clinical trial data. Leveraging evidence through quantitative frameworks like NTB allow us to leverage prior evidence to inform endpoint prioritization, reduce sample size requirements, and design trials that are more focused and efficient. Taken together, these strategies can accelerate development while ensuring that trials answer the questions that matter most to patients and clinicians alike.

If you could change one thing about how oncology trials are currently designed or evaluated, what would it be?

I would change the way we think about what constitutes “meaningful benefit.” Too often, trial designs are anchored to endpoints that may not always reflect all that matters to patients. If I could change one thing, it would be to make the explicit prioritization of outcomes — guided by both clinical insight and patient perspective — a standard part of trial design.

This means going beyond survival curves and response rates to also consider factors like symptom burden, treatment-related toxicity, and quality of life outcomes. It also means valuing methods that allow for the integration of these outcomes in a structured, interpretable way, such as GPC. Redefining what constitutes benefit is essential if we want oncology clinical trials to keep pace with the growing emphasis on patient centricity. It’s not enough to demonstrate efficacy, we need to ensure that evidence reflects the outcomes patients value and the trade-offs they are actually willing to make.

Original content written for DIA available here. Author: Samuel Salvaggio

Patient-centricity has become a focal point in the evolution of clinical trials, emphasizing outcomes meaningful to patients. The challenge lies in balancing rigorous scientific evaluation with incorporating patient experiences and preferences.

Current approaches to evaluating treatment efficacy often fail to capture the complexity of patient needs, particularly in conditions with diverse symptomatic or functional impacts. Conventional statistical analyses typically assess outcomes in isolation, neglecting their interactions and the multidimensional nature of treatment effects. This narrow focus can overlook clinically relevant improvements in symptoms, function, or tolerability, leading to an incomplete understanding of a treatment’s true benefits and risks.

Balancing the demands of scientific rigor with the reality of how patients experience illness is now a central challenge in trial design—one that requires solutions capable of integrating clinical relevance with lived experience. Today, market authorizations for new treatments are based on demonstrating superiority on a single clinical outcome. This paradigm can disregard the constellation of diverse effects a treatment may produce and the multidimensional needs and preferences of patients. As a result, it often fails to capture the complexity of patient experiences and overlooks important interactions and broader implications of treatment effects that matter greatly to those living with illness.

As clinical research moves toward greater patient-centricity, trials must evolve to reflect outcomes genuinely meaningful to those they seek to help.

The Net Treatment Benefit (NTB), a robust statistical measure derived from the Generalized Pairwise Comparisons (GPC) methodology, addresses this challenge. NTB integrates multiple prioritized outcomes into a single, interpretable metric, comprehensively reflecting treatment impacts. By aligning trial designs more closely with patient preferences and clinical realities, NTB enhances clinical relevance, optimizes trial efficiency, and ensures that trials truly reflect patient needs and priorities. Adopting such innovative methodologies will drive the next generation of clinical trials toward greater alignment with patient realities and preferences.

Towards Patient-Centric Clinical Trials

In recent years, regulatory agencies have been increasingly encouraging such approaches. The US Food and Drug Administration (FDA) has underscored the importance of incorporating patient perspectives into the drug development process, notably through its Patient-Focused Drug Development (PFDD) guidance series. The European Medicines Agency (EMA) has similarly called for the integration of patient perspectives in benefit-risk assessments and clinical trial designs. These documents encourage sponsors to systematically capture and integrate patient input when defining clinical outcomes, recognizing that patients may prioritize aspects of treatment beyond traditional clinical endpoints.

However, moving from intention to implementation remains a major challenge. Despite clear guidance, most trials continue to fall back on conventional endpoints, in part because there is no standard method to translate patient preferences into structured, regulatory-grade evidence. Sponsors are often left asking:

• Which outcomes should we prioritize?
• How do we quantify their relevance?
• And how can we analyze them together in a way that remains statistically sound and acceptable to regulators?

Introducing the Net Treatment Benefit

The Net Treatment Benefit (NTB) offers a scientifically grounded and patient-centered way to evaluate treatment effects. Developed over more than a decade of academic research and rooted in the Generalized Pairwise Comparisons (GPC) framework, NTB provides a robust and validated alternative to single-endpoint analyses. It allows multiple outcomes—of any type, including efficacy, safety, and patient-reported measures—to be integrated into a single, interpretable hierarchical composite endpoint.

Rather than analyzing outcomes marginally, NTB considers them jointly, respecting both their clinical importance and the order in which they matter. This makes it particularly well suited for trials that aim to reflect patient preferences and experiences. It is both statistically rigorous and adaptable to the complexity of modern clinical research.

Applications in Cardiovascular and Rare Disease and Regulatory Considerations

The practicality and real-world relevance of the Net Treatment Benefit have been demonstrated across several therapeutic areas, most notably in oncology, cardiovascular disease, and rare disorders, three settings where conventional approaches often fall short.

In the cardiovascular domain, NTB played a pivotal role in the approval of two treatments for transthyretin amyloid cardiomyopathy, a progressive and life-threatening condition. In both of these phase 3 trials, the NTB methodology was used as the prespecified primary analysis, prioritizing survival over hospitalization. This allowed the trials to more effectively capture the composite clinical benefit of treatment—not just prolonging life, but reducing the frequency of hospital stays, which are highly relevant to patients and healthcare systems alike. Importantly, numerous trial designs incorporating this methodology are now widely accepted by regulators, establishing NTB as a standard approach that meets evidentiary requirements for market authorization while providing a more patient-relevant assessment of treatment effects. Reflecting this growing acceptance, a search of the NIH-affiliated website ClinicalTrials.gov in May 2025 identified 24 phase 3 industry-sponsored trials in cardiometabolic diseases and obesity that employed the methodology for their primary endpoint.

Similarly, in the rare disease space, NTB has offered important insights where traditional methods have fallen short. For example, in Pompe disease—a progressive neuromuscular disorder—the phase 3 COMET trial originally failed to meet its primary endpoint using a conventional, single-outcome analysis focused on forced vital capacity. However, a subsequent GPC-based analysis, which integrated additional functional outcomes such as the six-minute walk test, revealed a statistically significant advantage for the experimental therapy. This reanalysis captured a broader picture of benefit, better aligned with patient experiences and expectations. In rare diseases, where trials are often constrained by small sample sizes and symptom heterogeneity, NTB provides a valuable way to maximize information gained from each participant while highlighting clinically meaningful effects.

Together, these examples illustrate how NTB can strengthen the case for therapeutic benefit by offering a more complete, patient-centered assessment—one that is not only methodologically sound but also compatible with regulatory expectations.

From Principles to Practice

As clinical research embraces patient-centricity, trial methodologies must evolve to meet the moment. Relying on univariate primary endpoints is no longer sufficient in a landscape where patients demand, and regulators support, a fuller picture of the benefits and risks of treatment. The NTB can offer a statistically robust, flexible, patient-aligned solution—one capable of integrating what matters most to patients into a single, interpretable assessment.

With case studies in different therapeutic areas, including cardiovascular and rare diseases, and demonstrated acceptability by regulators, NTB moves beyond theory into practical impact. It not only strengthens the scientific integrity of trials but also ensures that the evidence generated is truly reflective of patient priorities.

Realizing the full potential of NTB requires active collaboration across stakeholders to determine which clinical outcomes should be prioritized and incorporated into the primary endpoint. Patients, investigators, and clinicians should be engaged as early as possible to capture the preferences of this diverse group of stakeholders. Embracing this shift toward patient-centered outcome prioritization can offer sponsors a strategic advantage, while regulators and HTA bodies can play a key role by clarifying how such composite endpoints will be evaluated, supporting broader adoption in clinical development. Finally, sponsors must embrace this shift—not just as a compliance obligation but as a strategic advantage. Trials designed with NTB can be faster, leaner, and more aligned with payer expectations, increasing the likelihood of both regulatory success and patient impact.

As the expectations of patients, regulators, and health systems evolve, so must our tools. Methodologies like NTB represent a necessary step forward—transforming patient-centricity from a guiding principle into a measurable standard in clinical trial design.

Louvain-la-Neuve, Belgium — June 17th, 2025 

One2Treat SA, a fast-growing tech company incorporating the patient voice in all strategic decisions about treatment assessments in Pharma R&D, today announced the launch of the One2Treat Voice app, a new module of their software platform. 

This new milestone supports One2Treat’s commitment to developing its software platform to better express the overall medical value of a treatment.  One2Treat Voice is an innovative software designed to identify and prioritize the clinical outcomes that matter most from a patient or a clinician perspective. It enables the trial sponsor to incorporate the patient voice in the definition of endpoints in clinical trials, but also to support more holistic and clinically meaningful treatment assessments based on randomized clinical trials.  

This strategic development supports One2Treat’s mission to transform how clinical trials are designed and how treatment value is assessed, anchoring every decision in the perspective of what matters most for patients. 

“This past year has shown that patient-centricity and scientific rigor not only coexist but strengthen each other in clinical developments. With strong traction from our partners and the release of One2Treat Voice, we are accelerating a shift in the industry toward clinical development that is more relevant – and transparent – to what matters most to patients, clinicians, regulators and pharma sponsors.” 

Marc Buyse – One2Treat Founder 

Launch of One2Treat Voice: capturing stakeholder priorities to guide endpoint selection 

One2Treat has officially launched One2Treat Voice, a software module designed to capture and prioritize inputs from patients, clinicians, regulatory bodies and clinical trial sponsors. The result is a structured, data-driven prioritization of clinical outcomes, enabling the definition of multi-dimensional primary endpoints using the Net Treatment Benefit (NTB) methodology.  

“By embedding diverse stakeholder voices early in the protocol development process, One2Treat Voice helps sponsors design trials that are not only scientifically robust but also aligned with clinical development’s stakeholder values and expectations, while often allowing for significant sample size reductions.”  

Pascal Piedbois, Chief Medical Officer, One2Treat 

One2Treat Voice may also help physicians and patients express what matters most to them, leading to a clearer decision about the most appropriate approved treatment – based on past randomized clinical trial data. 

Expanding a cloud-based platform for multi-dimensional treatment assessment 

One2Treat Voice is part of a broader cloud-based platform that supports strategic decision-making throughout the clinical development process, from protocol design to data analysis. The platform empowers sponsors to assess the totality of the evidence and understand the overall medical value of a treatment. One2Treat’s platform is also used to support trial design, go/no-go decisions, dose selection, market access planning, and commercialization based on patient-and clinician-prioritized outcomes. 

“One2Treat Voice integrates quantitative and qualitative approaches to collect and prioritize patient preferences. Its adoption by pharma sponsors to collaborate with patient advocates, site investigators, and regulators highlights the industry’s commitment to more patient-focused drug development, by focusing treatment assessment on what matters most.” 
 

Sébastien Coppe, Chief Executive Officer, One2Treat 

About One2Treat 

One2Treat is a fast-growing tech company incorporating the patient voice in all strategic decisions about treatment assessments in Pharma R&D. One2Treat is dedicated to transforming the way biopharmaceutical companies design and evaluate clinical trials, and how the overall medical value of a treatment may be communicated, by integrating multiple meaningful dimensions in the treatment assessment 

For more information about One2Treat and its innovative approach to clinical development and market access, please visit: www.one2treat.com. 

Media Contact 

Tom Mann   

Clinical Solutions Engagement Lead, One2Treat 

Tom.Mann@one2treat.com 

www.one2treat.com 

Original content written for The Journal for Clinical Studies, available here. Author: Tom Mann

Introduction

Rare disease clinical trials face a confluence of challenges: limited patient populations, heterogeneity in disease progression, and often a lack of established outcome measures. Yet the stakes involved are exceptionally high. For the over 300 million people living with a rare disease worldwid[SS1] e (1), most of whom lack access to effective therapies, each trial represents a vital opportunity—not just to generate evidence, but to shape treatments that meaningfully improve the patients’ lives.

Traditional clinical trial designs, typically focusing on a single primary endpoint, are often ill-suited for this complex task. They frequently simplify the multidimensional reality of how patients, caregivers, and clinicians define meaningful treatment benefit into a single dimension. For instance, a therapy may slow disease progression but negatively impact quality of life; it may show modest improvement in the main endpoint yet substantially improve fine motor functions or reduce intolerable side effects. In rare diseases, where patient numbers are limited and the burden of participation is high, trials must do more than test hypotheses—they must produce data that reflects what matters most to those affected.

The Net Treatment Benefit (NTB) emerges as a patient-centric, statistically rigorous approach that allows for the prioritization and integration of multiple outcomes into a single, interpretable measure of treatment effects (2). When coupled with early engagement from patients, investigators, and key experts to define outcome hierarchies, NTB offers a practical path to trials that are both more efficient and more aligned with real-world needs.

The Challenge of Endpoint Selection in Rare Diseases

One of the most persistent bottlenecks in rare disease trial design is the selection of an appropriate primary endpoint. In common conditions, regulatory precedent and existing clinical guidelines typically point the way. In rare diseases, the path is often uncharted.

Consider Pompe disease or Duchenne muscular dystrophy. Patients, families, and clinicians may prioritize very different outcomes depending on the disease stage: respiratory function, ambulatory capacity, ability to feed independently, fatigue, or even cognitive symptoms in syndromic variants. Designing a trial around one of these clinical outcomes risks overlooking the others—and worse, dismissing a therapy that offers multidimensional benefit simply because it falls short on a single axis.

This issue becomes more acute when regulators require “hard” clinical outcomes, such as time to death or forced vital capacity, that may not be the most relevant for early- or mid-stage patients. Many rare diseases progress slowly or unpredictably, making it difficult to observe changes in a single outcome within the limited duration of a trial.

By forcing sponsors to choose one outcome as the sole measure of success, traditional designs risk misrepresenting the true value of an intervention. This not only complicates regulatory evaluation but can discourage further investment in promising therapies.

Why Net Treatment Benefit Is a Game-Changer

Net Treatment Benefit, grounded in the methodology of Generalized Pairwise Comparisons (GPC), offers a solution to these challenges. Rather than selecting a single endpoint, NTB enables trials to incorporate multiple outcomes—each assigned a position in a pre-defined hierarchy reflecting clinical and patient priorities.

In essence, NTB calculates the difference between the probability that a randomly selected patient in the treatment group does better across the prioritized outcomes than a randomly selected patient in the control group, and the reverse. This yields a single, interpretable statistic that reflects the totality of the evidence.

 The statistical advantages are compelling. By incorporating multiple relevant outcomes into the analysis simultaneously, NTB makes fuller use of the collected patient data, effectively capturing more comprehensive information about treatment effects. This is especially critical in rare disease trials, where small sample sizes are the norm. More efficient use of available data means improved power to detect clinically meaningful differences—potentially with fewer patients or shorter trial durations.

Specifically in the rare disease domain, a post-hoc analysis of the randomized, double-blind, phase 3 COMET trial, prioritizing the primary (forced vital capacity) and secondary outcome (6MWT), provided evidence of efficacy of avalglucosidase alfa therapy (n = 51) over alglucosidase alfa (n = 49) in Pompe disease, while the original analysis failed to significantly show superiority of treatment on the primary endpoint (3).

Prioritizing Outcomes with Stakeholder Input

What truly sets NTB apart is not just its statistical sophistication, but its ability to formalize clinical and patient preferences in the design phase of a trial.

In rare diseases, the need for such an approach is acute. Disease burden varies widely across individuals, and the diversity of symptom trajectories makes a one-size-fits-all endpoint inadequate. Engaging stakeholders early—patients, caregivers, site investigators, and treating clinicians—enables trial sponsors to co-create outcome hierarchies that reflect the lived experience of the disease.

 Structured preference elicitation methods, such as discrete choice experiments or ranking exercises, can yield clear insights into which outcomes matter most and in what order. However, these traditional approaches can be cumbersome, often requiring large numbers of respondents. Innovative methods are therefore needed to simplify the process and reduce the burden, especially in rare diseases with limited patient populations.

By building consensus around outcome prioritization upfront, sponsors not only create trials that are more meaningful—they reduce the risk of post-hoc disputes about relevance and increase the likelihood that trial data will resonate with regulators, payers, and clinicians.

Reducing the Burden on Patients and Families

Rare disease trial participants and their families often carry a disproportionate burden: frequent travel, complex assessments, and uncertainty around the value of their contribution. Any opportunity to streamline trials without compromising scientific integrity is not just a design consideration—it’s imperative.

NTB can reduce this burden in two important ways. First, by increasing statistical efficiency, NTB-based designs may require fewer patients to reliably detect whether a treatment is truly effective. Second, by allowing multiple outcomes to contribute to the primary analysis, NTB helps ensure that more of the collected data is meaningfully used, reducing waste and enhancing the value of each patient assessment.

Moreover, NTB allows the inclusion of clinically meaningful thresholds—minimum differences that matter to patients—in the analysis. This means that only differences considered meaningful are used to distinguish between outcomes, while smaller, less relevant differences are treated as neutral. This helps the analysis focus on what truly matters and adds another layer of patient-centricity, ensuring that the trial’s conclusions reflect not just differences, but meaningful ones.

Supporting Regulatory and HTA Pathways

 While the NTB has yet to become a standard primary analysis method in rare disease regulatory submissions, it is already well established and familiar to regulators in other therapeutic areas.

The ATTR-ACT trial for transthyretin amyloid cardiomyopathy used an NTB-like approach to prioritize time to death over time to hospitalization—highlighting how multidimensional benefit-risk profiles can be formalized in regulatory-grade evidence (4).  As regulatory agencies continue to emphasize patient-focused drug development (PFDD) (5), particularly for conditions where unmet need is high, there is a growing appetite for approaches that reflect the real-world complexity of treatment benefit.

Importantly, NTB is also well-suited for health technology assessments (HTAs). These bodies are increasingly requiring quantitative evidence of value beyond clinical efficacy—especially in Europe and Canada, where quality-adjusted life years (QALYs) and other composite measures are common. Because NTB summarizes multiple prioritized outcomes into a single interpretable measure, it aligns well with the demands of HTA dossiers and payer value frameworks.

In rare diseases, where treatments are often high-cost and subject to scrutiny, demonstrating comprehensive benefit-risks balance quantitatively is critical not only for approval but for access.

Fostering Adoption and Continuation of Development

An often-overlooked benefit of NTB in rare diseases is its potential to de-risk development decisions. When phase 2 trials are underpowered due to small sample sizes, NTB can detect more signal from limited data. Sponsors can make better-informed go/no-go decisions, reducing the likelihood of prematurely abandoning promising therapies or investing heavily in interventions with narrow appeal.

In turn, this supports better engagement with investors and partners. A clear, well-structured NTB analysis—grounded in patient and clinician priorities—can be a persuasive element in fundraising and partnership discussions. It also supports clinicians in understanding which patients are most likely to benefit, based on outcomes that mirror their own treatment goals.

Conclusion: Making Rare Disease Trials Work for Patients

For decades, rare disease trials have struggled under the weight of conventional clinical trial methodologies not designed for their constraints. The use of a single endpoint often obscures meaningful multidimensional benefits. It increases the likelihood of inconclusive results, slows development, and most importantly, can fail to serve the patients who volunteer their time, energy, and hopes.

Net Treatment Benefit, supported by robust stakeholder engagement in the selection and prioritization of outcomes, offers a viable, scalable, and scientifically rigorous solution. It allows for the integration of what matters most—survival, function, quality of life, and tolerability—into a single evaluative framework. And in doing so, it makes trials more efficient, more informative, and more aligned with real-world treatment decisions.

As the rare disease community continues to push for faster, more meaningful innovation, the integration of NTB into early trial design is not just a statistical refinement. It is a strategic imperative—one that places patients, not endpoints, at the center of progress.

---

References:

• (1) Rare Diseases International – https://www.rarediseasesinternational.org/living-with-a-rare-disease
• (2) Buyse, M., Verbeeck, J., Saad, E.D., Backer, M.D., Deltuvaite-Thomas, V., & Molenberghs, G. (Eds.). (2025). Handbook of Generalized Pairwise Comparisons: Methods for Patient-Centric Analysis (1st ed.). Chapman and Hall/CRC https://doi.org/10.1201/9781003390855
• (3) Verbeeck, J., Dirani, M., Bauer, J. W., Hilgers, R. D., Molenberghs, G., & Nabbout, R. (2023). Composite endpoints, including patient reported outcomes, in rare diseases. Orphanet Journal of Rare Diseases, 18(1), 262.
• (4) Maurer, M. S., Schwartz, J. H., Gundapaneni, B., Elliott, P. M., Merlini, G., Waddington-Cruz, M., … & Rapezzi, C. (2018). Tafamidis treatment for patients with transthyretin amyloid cardiomyopathy. New England Journal of Medicine, 379(11), 1007-1016.
• FDA, Patient Focused Drug Development Series – https://www.fda.gov/drugs/development-approval-process-drugs/fda-patient-focused-drug-development-guidance-series-enhancing-incorporation-patients-voice-medical