Enhancing Empowerment in Patient-Focused Research

Enhancing Empowerment in Patient-Focused Research

  Back

How the inclusion of measures centered on net treatment benefit can drive an effective multifaceted approach.

Patient-centricity has become a guiding principle for clinical research and treatment evaluation. There is a strong movement in clinical research to actively involve patients at the center of decision-making, with their preferences, quality of life, and individual values increasingly recognized as key elements in assessing treatment success. The net treatment benefit (NTB) embodies this patient-centered approach by allowing integration of diverse clinical outcomes into a single comprehensive metric. By prioritizing outcomes based on patient input, NTB offers a more personalized and transparent way to evaluate treatment effects, empowering patients to play an active role in their healthcare journey.

The robust methodology behind NTB

NTB is a measure of treatment effects estimated from generalized pairwise comparisons (GPC), a statistical method that stems from the well-known Mann-Whitney Wilcoxon non-parametric test.1 The primary innovation of GPC is to provide a global assessment of treatment effects through the hierarchical integration of any type of outcomes, such as efficacy, safety, and quality of life. By allowing the integration of multiple outcomes within a single analysis, this approach can overcome the limitations of traditional methods that focus on a single primary endpoint. Indeed, analyzing outcomes hierarchically rather than marginally better reflects their interactions and dependencies.

An advantage of GPC is that its logic is quite simple. The approach sequentially compares each clinical outcome based on a predefined list of priorities. For instance, in a cancer trial, overall survival may be the highest-ranked outcome, followed by cancer progression outcomes, treatment-related adverse events, and quality of life measures. It also allows the use of thresholds of clinical relevance that define a minimum clinical difference required for classifying two pairs. Once the priorities and thresholds are specified, each patient from the experimental arm is compared with all possible patients from the control arm, starting with the highest-priority outcome. The possible outcomes of these comparisons are classified as a “favorable” (if the patient in the experimental group fares better), “unfavorable” (if the control patient does better), or “neutral” (if there is no significant difference between the two patients).

If a pair is classified as neutral, the comparison moves to the next outcome, which might be toxicity or quality of life. Each pair initially classified as “neutral” is then compared on the next outcome and reclassified as “favorable,” “unfavorable,” or “neutral.” This continues until all hierarchical outcomes have been considered. The NTB can be calculated as the difference between the probability of being favorable to the experimental treatment and the probability of being favorable to control. This creates an easily interpretable absolute metric where a NTB value greater than zero indicates that the experimental treatment provides a net benefit over the control treatment, while a value less than zero indicates net harm.

Possible applications and case studies

This emerging class of statistical analysis is sometimes referred to as hierarchized composite outcome and is gaining increased traction in clinical research. One of the most recognized is an alternative estimator called the win ratio (WR). However, as a relative measure, the WR has several limitations. First, it lacks transparency in illustrating how individual outcomes contribute to the overall treatment effect, making it difficult to understand the specific factors driving the observed benefit or harm. Moreover, the WR disregards neutral pairs by assuming they behave similarly to classified pairs, which can be misleading if this assumption does not hold. Recently, two treatments have been approved by FDA’s cardiovascular division using GPC and the WR, the most well-known being tafamidis.2

In contrast, the NTB provides a clearer understanding of treatment effects by explicitly detailing each outcome’s contribution. This makes NTB a more transparent and patient-centered approach, ensuring that each relevant clinical outcome is considered and that the overall estimate reflects a comprehensive picture of patient impact. Recent case studies have demonstrated the benefits of NTB in highlighting differences across a wide range of patient-relevant outcomes, ultimately leading to more informed and patient-aligned decision-making in clinical trials.

The NTB can also address some of the shortcomings associated with non-inferiority trials. As discussed in a recent published paper in Lancet Oncology, non-inferiority trials often face criticism for relegating outcomes such as toxicity and cost to secondary considerations, and for setting very narrow statistical margins that can make it difficult for alternative, potentially patient-friendlier treatments to demonstrate equivalence. These trials frequently require large sample sizes and expensive resources to prove that an alternative is “not worse” than the standard, even when the clinical differences are minimal.3 The NTB, on the other hand, can transform this approach by incorporating multiple outcomes into a superiority benefit-risk analysis, offering a more nuanced and holistic assessment. By prioritizing and integrating different endpoints, NTB helps redefine the evaluation from simply establishing non-inferiority to understanding the full spectrum of benefits and risks, thereby supporting more patient-centered, cost-effective, and meaningful treatment decisions.

Incorporating patient preferences into NTB

One of the most powerful aspects of NTB is its ability to accommodate patient preferences directly into the clinical trial design and analysis. In the traditional trial framework, only outcomes such as overall survival or disease-free survival are used for market approval, while others—such as quality of life, symptom burden, or adverse effects—are secondary considerations. This approach can fail to reflect what truly matters to patients, particularly when dealing with chronic conditions where long-term side effects or daily functioning may outweigh the benefit of the former outcomes. By focusing solely on clinical endpoints, traditional methods may overlook the nuanced needs and desires of patients, limiting the comprehensiveness of the treatment effect evaluation.

The NTB, however, can be estimated based on the outcomes that are most important to individual patients. This patient-centric approach empowers individuals by giving them a voice in the evaluation process. It enables patients to express their preferences and offers a more personalized view of treatment effects. For example, patients may prioritize quality of life and minimal side effects over overall survival, depending on their unique circumstances. By allowing for such individual differences, NTB provides a more holistic understanding of treatment benefits that align closely with patients’ lived experiences.

To facilitate the incorporation of patient preferences, software tools designed to elicit these preferences are essential. These tools can help quantify and integrate patient values, allowing researchers to develop personalized NTB metrics. Such tools exemplify how patient input can be incorporated into the NTB methodology, transforming the traditionally rigid framework of clinical trials into one that is more adaptable to individual patient needs. The FDA currently encourages medical researchers to engage with patients as early as the design phase of clinical trials,4 underscoring the importance of understanding patient priorities from the outset.

Furthermore, relevant methods are needed to elicit patient preferences in the context of treatment effect measures. Current tools and techniques, such as surveys, discrete choice experiments, or conjoint analysis, offer promising approaches but still lack standardization across the industry. While the FDA, in past guidance documents, acknowledges that “quantitative patient preference assessment is an active and evolving research area,”5 no standard process has been validated yet. This lack of standardized tools poses a challenge, as pharmaceutical companies may be hesitant to invest in patient-focused methodologies without clear regulatory frameworks.

Significant efforts are needed from the industry to address this unmet need and support patient-focused clinical research. This includes developing and validating standardized methods to elicit patient preferences and ensuring these methods are robust enough to be used in regulatory submissions. By taking these steps, the industry can foster a more patient-centered approach to clinical trials, ultimately leading to treatments that better reflect patient needs and enhance their quality of life.

Tom Mann is Clinical Solutions Engagement Lead; Sarah Kosta, PhD, is Clinical Trial Solutions Analyst; and Samuel Salvaggio, PhD, is Senior Trial Design Lead; all with One2Treat

References

1. Buyse, M. Generalized Pairwise Comparisons of Prioritized Outcomes in the Two‐Sample Problem. Stat Med. 2010. 29 (30), 3245-3257. https://pubmed.ncbi.nlm.nih.gov/21170918/

2. Maurer, M.S.; Schwartz, J.H., Gundapaneni, B., et al. Tafamidis Treatment for Patients with Transthyretin Amyloid Cardiomyopathy. N Engl J Med. 2018. 379 (11), 1007-1016. https://www.nejm.org/doi/full/10.1056/NEJMoa1805689

3. Tannock, I. F.; Buyse, M.; De Backer, M.; et al. The Tyranny of Non-Inferiority Trials. The Lancet Oncol. 2024. 25 (10), e520-e525. https://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(24)00218-3/abstract

4. FDA, FDA Patient-Focused Drug Development Guidance Series for Enhancing the Incorporation of the Patient’s Voice in Medical Product Development and Regulatory Decision Making (February 14, 2024)https://www.fda.gov/drugs/development-approval-process-drugs/fda-patient-focused-drug-development-guidance-series-enhancing-incorporation-patients-voice-medical

5. FDA, Patient Preference Information – Voluntary Submission, Review in Premarket Approval Applications, Humanitarian Device Exemption Applications, and De Novo Requests, and Inclusion in Decision Summaries and Device Labeling (August 2016). https://www.fda.gov/regulatory-information/search-fda-guidance-documents/patient-preference-information-voluntary-submission-review-premarket-approval-applications