Hello everyone,

This is Dr Bishal Gyawali from Queen’s University in Kingston, Canada. Today, I want to talk about a checklist for reporting phase three randomized controlled trials of cancer drugs.

This checklist is an output from a project we conducted at Common Sense Oncology. As you may or may not know, Common Sense Oncology is a growing movement that includes physicians, patients, the public, and major stakeholders from all over the world, all focused on outcomes that truly matter to patients.

In recent times, we have realised that we may have lost sight of what really matters to patients, often focusing on arbitrary benchmarks that may or may not lead to better outcomes. Our goal is to refocus on patient-centred results.

One of the key projects that emerged from Common Sense Oncology was the development of a checklist for properly reporting cancer drug trials. While there are existing checklists, such as the CONSORT checklist for reporting randomized trials in general publications, they are not specific to cancer trials. Cancer drug RCTs often have limitations and biases that are not well captured in these existing checklists.

As a result, we now face situations where important drug trials are published in major, highly reputable journals, with strong conclusions suggesting a change in the standard of care. However, when you read these trials in detail, you often find several problems that undermine their conclusions.

For busy practitioners or trainees who may not have the expertise or time to analyse these trials in depth, this can be problematic. Even peer reviewers and journal editors may not always be fully aware of these issues. That’s why we developed a checklist highlighting the key points that should be considered when evaluating cancer drug RCTs.

By the time this video is released, this checklist will have been published in The Lancet Oncology. I highly recommend reviewing the publication. The checklist is designed to help trainees understand how to critically read a trial and to support physicians, policymakers, and other stakeholders in assessing whether a trial has been properly reported. It is also a useful tool for peer reviewers and journal editors.

Key Components of the Checklist:

Abstract

• The abstract should explicitly define the primary endpoint of the trial.
• It should include the hazard ratio for the primary endpoint and for time-to-event endpoints, as well as for overall survival (OS), even if OS is not the primary endpoint.
• OS results should be presented with confidence intervals and a measure of absolute benefit.
• There should be an objective summary of grade 3–5 toxicity, chronic toxicities, and treatment discontinuation rates. Subjective language downplaying toxicity (e.g., stating that treatment was "tolerable," "feasible," or "acceptable") should be avoided.
• A statement about the impact on health-related quality of life should be included.
• A lay summary of the main results, written in language understandable to patients and the public, should be provided.

Methods Section

• A statement on how patients and the public were involved in study design.
• Justification for the control arm, ensuring it represents the standard of care.
• Justification for the primary endpoint—if the primary endpoint is not overall survival, evidence supporting its surrogacy for OS or quality of life should be provided.
• The statistical basis for sample size and the level of benefit the trial is powered to detect or exclude.
• Strategies used to limit dropout and censoring.
• Plans for crossover—indicating whether crossover was mandated and funded if appropriate.
• Criteria for interim analysis and early stopping—for palliative trials, early stopping should occur only if there is a definitive improvement in OS with a meaningful margin.

Results Section

• Time-to-event Kaplan-Meier curves for the primary endpoint and overall survival, with numbers at risk and numbers censored displayed.
• Absolute measures of benefit.
• Reasons for dropout and censoring, along with sensitivity analyses to assess potential effects on results (this can be in an online appendix).
• The number of patients in the control group who crossed over to receive the experimental treatment after disease progression.
• Data on post-progression treatment for all patients (can also be in an appendix).
• An objective assessment of toxicity, alongside patient-reported assessments of toxicity.
• Whether the trial meets the pre-planned ESMO-MCBS criteria for substantial benefit. If it does not, the results should be considered negative, regardless of the p-value.
• Time toxicity should be included if the expected median survival is less than 12 months.

Discussion Section

• A summary of the main results.
• An assessment of benefit in relation to physical, financial, and time toxicity.
• The drafting of the manuscript by medical writers employed by the sponsor is discouraged. The trial’s principal investigators should write the manuscript themselves.

Conclusion

This checklist provides a structured way to ensure transparent and high-quality reporting of cancer drug RCTs. There is also a separate checklist available for designing RCTs, which is included in The Lancet Oncology publication.

I strongly encourage you to use this checklist in your critical appraisal of cancer drug trials.

Thank you.