============================================================ TITLE: Are Cancer Trial PROs Actually Measuring Treatment Tolerability? TYPE: article VERSION: 1 VERSION_ID: 968f8215-21de-437e-854b-1289f75df1b5 GENERATED_AT: 2026-06-19T15:08:23.220Z SUMMARY: Dropout rates of 15-30% in AOM trials threaten regulatory submissions. Evidence-based strategies to improve participant retention in anti-obesity trials. AUTHOR: Bill Byrom DATE PUBLISHED: June 19, 2026 DATE MODIFIED: June 19, 2026 READING TIME: 14 min WORD COUNT: 2678 KEYWORDS: Key Points SOURCE URL: https://signanthealth.com/resources/are-cancer-trial-pros-actually-measuring-treatment-tolerability ============================================================ KEY TAKEAWAYS: * Key Points * AUTHOR BIO Back to Resource Hub Article Are Cancer Trial PROs Actually Measuring Treatment Tolerability? Signant Health Jun 19, 2026 Article Measuring the severity and frequency of treatment side effects in cancer trials is not the same as measuring tolerability. That distinction, clear in concept but unresolved in methodology, is the subject of this peer-reviewed commentary published in Cancer Control (2026) by Bill Byrom, PhD, of Signant Health, and John Devin Peipert, PhD, of the University of Birmingham's Centre for Patient Reported Outcomes Research.   The paper identifies three specific methodological gaps that limit the value of patient-reported outcomes data currently collected in oncology trials and calls for the research needed to close them.  Key Points  PRO-CTCAE describes toxicity severity, frequency, and interference, but does not directly assess whether a participant is willing or able to continue treatment given the side effects experienced, which is the definition of tolerability under ICH E9. An analysis of three solid tumor clinical trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported FACT GP5 bother scores of 2 to 4 at baseline, before receiving study treatment, creating a measurement problem for change-from-baseline analyses. Tolerability thresholds, defining the point at which reported toxicity severity becomes intolerable for a specific cancer population, do not yet exist in validated form. The authors propose anchoring PRO-CTCAE and GP5 scores against observable tolerability indicators such as treatment pauses, dose reductions, and early discontinuation. A simple direct tolerability measure, asking participants whether they would be willing to continue treatment as it is, may offer immediate utility as a complement to existing toxicity measures and is proposed as a starting point for formal qualitative instrument development. The FACT GP5 item has appeared in FDA labelling for selpercatinib in thyroid cancer, where participant-reported overall side effect impact results were supported by lower rates of treatment discontinuation due to adverse reactions, demonstrating its regulatory utility while also illustrating its indirect relationship to tolerability itself. The paper, published open access in Cancer Control (doi: 10.1177/10732748261432279), contains the complete methodological analysis, exposure-response analysis discussion, and the authors' specific calls for research that this page does not reproduce in full.  Why It Matters Now: The Methodological Gap Has Regulatory Consequences       For a participant in an oncology trial managing fatigue, nausea, and neuropathy simultaneously, the question of whether they can tolerate continued treatment is not answered by recording each symptom as low-grade.   Tolerability is a personal, contextual judgement that weighs experienced toxicity against perceived or anticipated benefit, shaped by disease severity, treatment expectations, and individual life circumstances. One participant in a non-small cell lung cancer qualitative study captured this directly: the willingness to endure symptoms in exchange for the possibility that the cancer will recede is a calculation that varies between individuals and cannot be read from a severity score alone.  The FDA's October 2024 final guidance on core patient-reported outcomes in cancer clinical trials, alongside Project Optimus and the August 2024 final guidance on optimizing dosage for oncological treatments, has raised the evidentiary bar for tolerability characterization in oncology submissions. The field has responded by collecting more PRO data, more frequently, and across more domains.   What this commentary argues is that collecting more of the same data does not resolve the underlying methodological problem. Sponsors designing oncology programs today are investing in patient-reported outcome strategies that may not yield the tolerability evidence they need for regulatory and health technology assessment decision-making unless the gaps this paper identifies are addressed.  That challenge is active now, for trials in design and execution in 2026.   What the Paper Argues: Three Problems That Require Research Attention    The oncology PRO research community has treated PRO-CTCAE and the FACT GP5 as tolerability measures. This paper, co-authored across Signant Health and the University of Birmingham's Centre for Patient Reported Outcomes Research, argues they are not, and that the distinction matters practically. The first problem is that existing toxicity measures cannot be translated into tolerability inferences without threshold values that define when reported severity becomes intolerable for a specific cancer population. Existing analytical approaches, including time-to-event analyses using severity thresholds, area under the curve calculations, and exposure-response analysis, are more valuable when anchored against observable tolerability indicators. The authors call for studies that establish those anchors using meta-analyses of existing trial data and qualitative vignette methodologies. The second problem is the absence of direct tolerability measures. The authors propose that a single willingness-to-continue question, administered alongside GP5 and PRO-CTCAE, could provide an immediately actionable starting point, while acknowledging that dispositional factors beyond toxicity will also influence the response. Formal qualitative work to develop a multi-item direct tolerability measure is encouraged. The third problem is the GP5 baseline assessment challenge. Between 11.8% and 23.9% of participants report non-zero GP5 bother scores before treatment begins, whether due to prior treatment effects or negative expectations. The authors propose the development and patient-validated instructional text to accompany the GP5 at baseline, with specific suggested language included in the paper. For sponsors and CROs planning PRO tolerability strategy for upcoming oncology programs, the full paper provides the methodological foundation for understanding what current measures can and cannot support in regulatory submissions.   "Patient-reported outcomes have become indispensable in oncology clinical trials, yet critical methodological gaps remain that limit the value we can derive from the data patients provide. Through collaborative, systematic methodology research we can ensure that patient-reported outcome measures collected in cancer clinical trials fulfil their promise of truly patient-centred drug development." - Bill Byrom, PhD, eCOA Science, Signant Health; Cancer Control, 2026 (doi: 10.1177/10732748261432279)  What is the difference between toxicity and tolerability in cancer clinical trials? Toxicity measures the frequency, severity, and functional interference of treatment side effects, as captured by instruments such as PRO-CTCAE. Tolerability, as defined by ICH E9, is the degree to which overt adverse effects can be tolerated by the individual participant, a judgement that integrates toxicity with perceived treatment benefit, disease severity, and personal life circumstances. Current PRO measures characterize toxicity and its impact but do not directly assess whether a participant would be willing to continue treatment despite those effects. Why is the FACT GP5 baseline assessment problematic in cancer trials? The GP5 asks participants how bothered they are by the side effects of treatment. Before treatment has begun, this question is ambiguous. Published data from three solid tumor trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported non-zero GP5 bother scores at baseline. This creates bias in change-from-baseline analyses. The authors propose developing and validating instructional text, developed with patient input, to accompany the GP5 at pre-treatment assessments.  How should tolerability thresholds be developed for PRO measures in oncology trials? Tolerability thresholds define the point at which reported toxicity severity becomes intolerable for a specific patient population. They do not currently exist in validated form for PRO-CTCAE or GP5. The authors recommend anchoring severity scores against observable tolerability indicators such as treatment pauses, dose reductions, and early treatment discontinuation, using meta-analyses of existing trial data as a starting point. Qualitative approaches using hypothetical adverse event combination vignettes may help define multidimensional thresholds for composite measures. AUTHOR BIO  Name: Bill ByromTitle and Credentials: PhD, Vice President of Product Intelligence and Positioning at Signant HealthBio: Bill Byrom, PhD, is Vice President of Product Intelligence and Positioning at Signant Health, specializing in ePRO methodology, clinical outcome assessment strategy, and oncology PRO implementation.  He is the corresponding author of this open-access commentary, published in Cancer Control in 2026 (doi: 10.1177/10732748261432279), co-authored with John Devin Peipert, PhD, Professor of Health Outcomes Measurement at the University of Birmingham and a researcher at the Centre for Patient Reported Outcomes Research. The paper was accepted in February 2026 and published open access under a Creative Commons Attribution-Noncommercial 4.0 license.   READ THE FULL PAPER Designing PRO tolerability strategy for an oncology program and working through what current measures can and cannot support in a regulatory submission? Speak to Bill Byrom, PhD, about tolerability characterization methodology, GP5 implementation, and PRO-CTCAE strategy for cancer clinical trials. # Are Cancer Trial PROs Actually Measuring Treatment Tolerability? Jun 19, 2026 Measuring the severity and frequency of treatment side effects in cancer trials is not the same as measuring tolerability. That distinction, clear in concept but unresolved in methodology, is the subject of this peer-reviewed commentary published in Cancer Control (2026) by Bill Byrom, PhD, of Signant Health, and John Devin Peipert, PhD, of the University of Birmingham's Centre for Patient Reported Outcomes Research.   The paper identifies three specific methodological gaps that limit the value of patient-reported outcomes data currently collected in oncology trials and calls for the research needed to close them.  Key Points  PRO-CTCAE describes toxicity severity, frequency, and interference, but does not directly assess whether a participant is willing or able to continue treatment given the side effects experienced, which is the definition of tolerability under ICH E9. An analysis of three solid tumor clinical trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported FACT GP5 bother scores of 2 to 4 at baseline, before receiving study treatment, creating a measurement problem for change-from-baseline analyses. Tolerability thresholds, defining the point at which reported toxicity severity becomes intolerable for a specific cancer population, do not yet exist in validated form. The authors propose anchoring PRO-CTCAE and GP5 scores against observable tolerability indicators such as treatment pauses, dose reductions, and early discontinuation. A simple direct tolerability measure, asking participants whether they would be willing to continue treatment as it is, may offer immediate utility as a complement to existing toxicity measures and is proposed as a starting point for formal qualitative instrument development. The FACT GP5 item has appeared in FDA labelling for selpercatinib in thyroid cancer, where participant-reported overall side effect impact results were supported by lower rates of treatment discontinuation due to adverse reactions, demonstrating its regulatory utility while also illustrating its indirect relationship to tolerability itself. The paper, published open access in Cancer Control (doi: 10.1177/10732748261432279), contains the complete methodological analysis, exposure-response analysis discussion, and the authors' specific calls for research that this page does not reproduce in full.  Why It Matters Now: The Methodological Gap Has Regulatory Consequences       For a participant in an oncology trial managing fatigue, nausea, and neuropathy simultaneously, the question of whether they can tolerate continued treatment is not answered by recording each symptom as low-grade.   Tolerability is a personal, contextual judgement that weighs experienced toxicity against perceived or anticipated benefit, shaped by disease severity, treatment expectations, and individual life circumstances. One participant in a non-small cell lung cancer qualitative study captured this directly: the willingness to endure symptoms in exchange for the possibility that the cancer will recede is a calculation that varies between individuals and cannot be read from a severity score alone.  The FDA's October 2024 final guidance on core patient-reported outcomes in cancer clinical trials, alongside Project Optimus and the August 2024 final guidance on optimizing dosage for oncological treatments, has raised the evidentiary bar for tolerability characterization in oncology submissions. The field has responded by collecting more PRO data, more frequently, and across more domains.   What this commentary argues is that collecting more of the same data does not resolve the underlying methodological problem. Sponsors designing oncology programs today are investing in patient-reported outcome strategies that may not yield the tolerability evidence they need for regulatory and health technology assessment decision-making unless the gaps this paper identifies are addressed.  That challenge is active now, for trials in design and execution in 2026.   What the Paper Argues: Three Problems That Require Research Attention    The oncology PRO research community has treated PRO-CTCAE and the FACT GP5 as tolerability measures. This paper, co-authored across Signant Health and the University of Birmingham's Centre for Patient Reported Outcomes Research, argues they are not, and that the distinction matters practically. The first problem is that existing toxicity measures cannot be translated into tolerability inferences without threshold values that define when reported severity becomes intolerable for a specific cancer population. Existing analytical approaches, including time-to-event analyses using severity thresholds, area under the curve calculations, and exposure-response analysis, are more valuable when anchored against observable tolerability indicators. The authors call for studies that establish those anchors using meta-analyses of existing trial data and qualitative vignette methodologies. The second problem is the absence of direct tolerability measures. The authors propose that a single willingness-to-continue question, administered alongside GP5 and PRO-CTCAE, could provide an immediately actionable starting point, while acknowledging that dispositional factors beyond toxicity will also influence the response. Formal qualitative work to develop a multi-item direct tolerability measure is encouraged. The third problem is the GP5 baseline assessment challenge. Between 11.8% and 23.9% of participants report non-zero GP5 bother scores before treatment begins, whether due to prior treatment effects or negative expectations. The authors propose the development and patient-validated instructional text to accompany the GP5 at baseline, with specific suggested language included in the paper. For sponsors and CROs planning PRO tolerability strategy for upcoming oncology programs, the full paper provides the methodological foundation for understanding what current measures can and cannot support in regulatory submissions.   "Patient-reported outcomes have become indispensable in oncology clinical trials, yet critical methodological gaps remain that limit the value we can derive from the data patients provide. Through collaborative, systematic methodology research we can ensure that patient-reported outcome measures collected in cancer clinical trials fulfil their promise of truly patient-centred drug development." - Bill Byrom, PhD, eCOA Science, Signant Health; Cancer Control, 2026 (doi: 10.1177/10732748261432279)  What is the difference between toxicity and tolerability in cancer clinical trials? Toxicity measures the frequency, severity, and functional interference of treatment side effects, as captured by instruments such as PRO-CTCAE. Tolerability, as defined by ICH E9, is the degree to which overt adverse effects can be tolerated by the individual participant, a judgement that integrates toxicity with perceived treatment benefit, disease severity, and personal life circumstances. Current PRO measures characterize toxicity and its impact but do not directly assess whether a participant would be willing to continue treatment despite those effects. Why is the FACT GP5 baseline assessment problematic in cancer trials? The GP5 asks participants how bothered they are by the side effects of treatment. Before treatment has begun, this question is ambiguous. Published data from three solid tumor trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported non-zero GP5 bother scores at baseline. This creates bias in change-from-baseline analyses. The authors propose developing and validating instructional text, developed with patient input, to accompany the GP5 at pre-treatment assessments.  How should tolerability thresholds be developed for PRO measures in oncology trials? Tolerability thresholds define the point at which reported toxicity severity becomes intolerable for a specific patient population. They do not currently exist in validated form for PRO-CTCAE or GP5. The authors recommend anchoring severity scores against observable tolerability indicators such as treatment pauses, dose reductions, and early treatment discontinuation, using meta-analyses of existing trial data as a starting point. Qualitative approaches using hypothetical adverse event combination vignettes may help define multidimensional thresholds for composite measures. AUTHOR BIO  Name: Bill ByromTitle and Credentials: PhD, Vice President of Product Intelligence and Positioning at Signant HealthBio: Bill Byrom, PhD, is Vice President of Product Intelligence and Positioning at Signant Health, specializing in ePRO methodology, clinical outcome assessment strategy, and oncology PRO implementation.  He is the corresponding author of this open-access commentary, published in Cancer Control in 2026 (doi: 10.1177/10732748261432279), co-authored with John Devin Peipert, PhD, Professor of Health Outcomes Measurement at the University of Birmingham and a researcher at the Centre for Patient Reported Outcomes Research. The paper was accepted in February 2026 and published open access under a Creative Commons Attribution-Noncommercial 4.0 license.   READ THE FULL PAPER Designing PRO tolerability strategy for an oncology program and working through what current measures can and cannot support in a regulatory submission? Speak to Bill Byrom, PhD, about tolerability characterization methodology, GP5 implementation, and PRO-CTCAE strategy for cancer clinical trials. The paper identifies three specific methodological gaps that limit the value of patient-reported outcomes data currently collected in oncology trials and calls for the research needed to close them. ## Key Points * PRO-CTCAE describes toxicity severity, frequency, and interference, but does not directly assess whether a participant is willing or able to continue treatment given the side effects experienced, which is the definition of tolerability under ICH E9. * An analysis of three solid tumor clinical trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported FACT GP5 bother scores of 2 to 4 at baseline, before receiving study treatment, creating a measurement problem for change-from-baseline analyses. * Tolerability thresholds, defining the point at which reported toxicity severity becomes intolerable for a specific cancer population, do not yet exist in validated form. The authors propose anchoring PRO-CTCAE and GP5 scores against observable tolerability indicators such as treatment pauses, dose reductions, and early discontinuation. * A simple direct tolerability measure, asking participants whether they would be willing to continue treatment as it is, may offer immediate utility as a complement to existing toxicity measures and is proposed as a starting point for formal qualitative instrument development. * The FACT GP5 item has appeared in FDA labelling for selpercatinib in thyroid cancer, where participant-reported overall side effect impact results were supported by lower rates of treatment discontinuation due to adverse reactions, demonstrating its regulatory utility while also illustrating its indirect relationship to tolerability itself. * The paper, published open access in Cancer Control (doi: 10.1177/10732748261432279), contains the complete methodological analysis, exposure-response analysis discussion, and the authors' specific calls for research that this page does not reproduce in full. Why It Matters Now: The Methodological Gap Has Regulatory Consequences For a participant in an oncology trial managing fatigue, nausea, and neuropathy simultaneously, the question of whether they can tolerate continued treatment is not answered by recording each symptom as low-grade. Tolerability is a personal, contextual judgement that weighs experienced toxicity against perceived or anticipated benefit, shaped by disease severity, treatment expectations, and individual life circumstances. One participant in a non-small cell lung cancer qualitative study captured this directly: the willingness to endure symptoms in exchange for the possibility that the cancer will recede is a calculation that varies between individuals and cannot be read from a severity score alone. The FDA's October 2024 final guidance on core patient-reported outcomes in cancer clinical trials, alongside Project Optimus and the August 2024 final guidance on optimizing dosage for oncological treatments, has raised the evidentiary bar for tolerability characterization in oncology submissions. The field has responded by collecting more PRO data, more frequently, and across more domains. What this commentary argues is that collecting more of the same data does not resolve the underlying methodological problem. Sponsors designing oncology programs today are investing in patient-reported outcome strategies that may not yield the tolerability evidence they need for regulatory and health technology assessment decision-making unless the gaps this paper identifies are addressed. That challenge is active now, for trials in design and execution in 2026. What the Paper Argues: Three Problems That Require Research Attention The oncology PRO research community has treated PRO-CTCAE and the FACT GP5 as tolerability measures. This paper, co-authored across Signant Health and the University of Birmingham's Centre for Patient Reported Outcomes Research, argues they are not, and that the distinction matters practically. The first problem is that existing toxicity measures cannot be translated into tolerability inferences without threshold values that define when reported severity becomes intolerable for a specific cancer population. Existing analytical approaches, including time-to-event analyses using severity thresholds, area under the curve calculations, and exposure-response analysis, are more valuable when anchored against observable tolerability indicators. The authors call for studies that establish those anchors using meta-analyses of existing trial data and qualitative vignette methodologies. The second problem is the absence of direct tolerability measures. The authors propose that a single willingness-to-continue question, administered alongside GP5 and PRO-CTCAE, could provide an immediately actionable starting point, while acknowledging that dispositional factors beyond toxicity will also influence the response. Formal qualitative work to develop a multi-item direct tolerability measure is encouraged. The third problem is the GP5 baseline assessment challenge. Between 11.8% and 23.9% of participants report non-zero GP5 bother scores before treatment begins, whether due to prior treatment effects or negative expectations. The authors propose the development and patient-validated instructional text to accompany the GP5 at baseline, with specific suggested language included in the paper. For sponsors and CROs planning PRO tolerability strategy for upcoming oncology programs, the full paper provides the methodological foundation for understanding what current measures can and cannot support in regulatory submissions. "Patient-reported outcomes have become indispensable in oncology clinical trials, yet critical methodological gaps remain that limit the value we can derive from the data patients provide. Through collaborative, systematic methodology research we can ensure that patient-reported outcome measures collected in cancer clinical trials fulfil their promise of truly patient-centred drug development." - Bill Byrom, PhD, eCOA Science, Signant Health; Cancer Control, 2026 (doi: 10.1177/10732748261432279) ### What is the difference between toxicity and tolerability in cancer clinical trials? Toxicity measures the frequency, severity, and functional interference of treatment side effects, as captured by instruments such as PRO-CTCAE. Tolerability, as defined by ICH E9, is the degree to which overt adverse effects can be tolerated by the individual participant, a judgement that integrates toxicity with perceived treatment benefit, disease severity, and personal life circumstances. Current PRO measures characterize toxicity and its impact but do not directly assess whether a participant would be willing to continue treatment despite those effects. ### Why is the FACT GP5 baseline assessment problematic in cancer trials? The GP5 asks participants how bothered they are by the side effects of treatment. Before treatment has begun, this question is ambiguous. Published data from three solid tumor trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported non-zero GP5 bother scores at baseline. This creates bias in change-from-baseline analyses. The authors propose developing and validating instructional text, developed with patient input, to accompany the GP5 at pre-treatment assessments. ### How should tolerability thresholds be developed for PRO measures in oncology trials? Tolerability thresholds define the point at which reported toxicity severity becomes intolerable for a specific patient population. They do not currently exist in validated form for PRO-CTCAE or GP5. The authors recommend anchoring severity scores against observable tolerability indicators such as treatment pauses, dose reductions, and early treatment discontinuation, using meta-analyses of existing trial data as a starting point. Qualitative approaches using hypothetical adverse event combination vignettes may help define multidimensional thresholds for composite measures. ## AUTHOR BIO Name: Bill ByromTitle and Credentials: PhD, Vice President of Product Intelligence and Positioning at Signant HealthBio: Bill Byrom, PhD, is Vice President of Product Intelligence and Positioning at Signant Health, specializing in ePRO methodology, clinical outcome assessment strategy, and oncology PRO implementation.  He is the corresponding author of this open-access commentary, published in Cancer Control in 2026 (doi: 10.1177/10732748261432279), co-authored with John Devin Peipert, PhD, Professor of Health Outcomes Measurement at the University of Birmingham and a researcher at the Centre for Patient Reported Outcomes Research. The paper was accepted in February 2026 and published open access under a Creative Commons Attribution-Noncommercial 4.0 license. READ THE FULL PAPER Designing PRO tolerability strategy for an oncology program and working through what current measures can and cannot support in a regulatory submission? Speak to Bill Byrom, PhD, about tolerability characterization methodology, GP5 implementation, and PRO-CTCAE strategy for cancer clinical trials. ### Get notified on new marketing insights Here mention the benefits of subscribing ------------------------------------------------------------ FREQUENTLY ASKED QUESTIONS: Q: What is the difference between toxicity and tolerability in cancer clinical trials? A: Toxicity measures the frequency, severity, and functional interference of treatment side effects, as captured by instruments such as PRO-CTCAE. Tolerability, as defined by ICH E9, is the degree to which overt adverse effects can be tolerated by the individual participant, a judgement that integrates toxicity with perceived treatment benefit, disease severity, and personal life circumstances. Current PRO measures characterize toxicity and its impact but do not directly assess whether a participant would be willing to continue treatment despite those effects. Toxicity measures the frequency, severity, and functional interference of treatment side effects, as captured by instruments such as PRO-CTCAE. Tolerability, as defined by ICH E9, is the degree to which overt adverse effects can be tolerated by the individual participant, a judgement that integrates toxicity with perceived treatment benefit, disease severity, and personal life circumstances. Current PRO measures characterize toxicity and its impact but do not directly assess whether a participant would be willing to continue treatment despite those effects. Why is the FACT GP5 baseline assessment problematic in cancer trials? The GP5 asks participants how bothered they are by the side effects of treatment. Before treatment has begun, this question is ambiguous. Published data from three solid tumor trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported non-zero GP5 bother scores at baseline. This creates bias in change-from-baseline analyses. The authors propose developing and validating instructional text, developed with patient input, to accompany the GP5 at pre-treatment assessments. Q: Why is the FACT GP5 baseline assessment problematic in cancer trials? A: The GP5 asks participants how bothered they are by the side effects of treatment. Before treatment has begun, this question is ambiguous. Published data from three solid tumor trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported non-zero GP5 bother scores at baseline. This creates bias in change-from-baseline analyses. The authors propose developing and validating instructional text, developed with patient input, to accompany the GP5 at pre-treatment assessments. The GP5 asks participants how bothered they are by the side effects of treatment. Before treatment has begun, this question is ambiguous. Published data from three solid tumor trials found that 11.8% to 15.7% of treatment-naive participants and 23.9% of treatment-experienced participants reported non-zero GP5 bother scores at baseline. This creates bias in change-from-baseline analyses. The authors propose developing and validating instructional text, developed with patient input, to accompany the GP5 at pre-treatment assessments. How should tolerability thresholds be developed for PRO measures in oncology trials? Tolerability thresholds define the point at which reported toxicity severity becomes intolerable for a specific patient population. They do not currently exist in validated form for PRO-CTCAE or GP5. The authors recommend anchoring severity scores against observable tolerability indicators such as treatment pauses, dose reductions, and early treatment discontinuation, using meta-analyses of existing trial data as a starting point. Qualitative approaches using hypothetical adverse event combination vignettes may help define multidimensional thresholds for composite measures. Q: How should tolerability thresholds be developed for PRO measures in oncology trials? A: Tolerability thresholds define the point at which reported toxicity severity becomes intolerable for a specific patient population. They do not currently exist in validated form for PRO-CTCAE or GP5. The authors recommend anchoring severity scores against observable tolerability indicators such as treatment pauses, dose reductions, and early treatment discontinuation, using meta-analyses of existing trial data as a starting point. Qualitative approaches using hypothetical adverse event combination vignettes may help define multidimensional thresholds for composite measures. Tolerability thresholds define the point at which reported toxicity severity becomes intolerable for a specific patient population. They do not currently exist in validated form for PRO-CTCAE or GP5. The authors recommend anchoring severity scores against observable tolerability indicators such as treatment pauses, dose reductions, and early treatment discontinuation, using meta-analyses of existing trial data as a starting point. Qualitative approaches using hypothetical adverse event combination vignettes may help define multidimensional thresholds for composite measures. ------------------------------------------------------------ ABOUT THIS CONTENT ------------------------------------------------------------ Source: https://signanthealth.com/resources/are-cancer-trial-pros-actually-measuring-treatment-tolerability Author: Bill Byrom Published: June 19, 2026 This content is provided for informational purposes. Please visit the original source for the most up-to-date information.