I read with interest the recently published story on the recommendations to streamline data collection in certain NCI trials presented at the Nov. 9, 2022, meeting of the Clinical and Translational Research Committee (The Cancer Letter, Nov. 11, 2022).
My reaction: “What took so long?”
As a former CTAC member and cooperative group chair I’m happy to see recommendations beginning to emerge that are “intended to define a ‘new normal’ for data collection that is less burdensome, more efficient, and more sustainable.”
Everyone engaged in cancer clinical trials recognizes that these modest recommendations are long overdue and represent just the start of what needs to be a significant overhaul of the cancer clinical research enterprise to make clinical trials more efficient, accessible, and equitable. A key recommendation relates to collection of adverse events in IND-exempt treatment trials, which constitute about 40% of NCTN trials.
CTAC now recommends that only AEs of grade 3 and higher be collected, unless assessment of treatment tolerability is a stated objective of the trial, and that collection of AE attribution and start/stop dates no longer be collected. In discussion, a CTAC member reportedly asked whether any safety signals might be missed with this new approach.
This was an important question but, it appears, based on a response by Sumithra Mandrekar [professor of biostatistics at Mayo College of Medicine and oncology group statistician at Alliance for Clinical Trials in Oncology], one that wasn’t considered in the context of preparing the recommendations.
Fortunately, the question has, in fact, already been answered and the results are clear. In 2009, the American Society of Clinical Oncology Research Committee set out to determine if adverse event data collection in clinical trials could be reduced without missing important safety signals.
We reviewed 107,884 adverse events and 136,608 concomitant medication records collected as part of 8 completed phase III clinical trials that, together, enrolled 17,184 patients. We published our analysis in 2010 (J Clin Oncol 28:5046-5053) in an article entitled “Optimizing Collection of Adverse Event Data in Cancer Clinical Trials Supporting Supplemental Indications.”
The mix of industry-sponsored and publicly funded trials included in the analysis investigated chemotherapy, biologic, and hormonal treatments in the metastatic and adjuvant treatment settings across multiple tumor types. In each case, the investigational agent had already been studied in other phase III clinical trials and had an established safety profile.
From this analysis, we concluded that no significant safety signals would be missed if safety data collection in future trials to support supplemental indications was limited to deaths, serious adverse events and grade 3-4 AEs collected in a subset of patients enrolled in the trial.
We also noted that routine collection of concomitant medications created an enormous amount of work but provided little useful information, and recommended that collection of concomitant medications be limited to specific targeted collections based on the known safety and pharmacologic profile of the investigational agent, or that are related to a specific objective of the trial.
I don’t know why it has taken NCI 12 years since this paper was published to reach the same conclusions regarding data collection for AEs and concomitant medications and to modify their processes, but I certainly hope (and believe) they will move more quickly under Monica Bertagnolli’s leadership.
To their credit, NCI moved quickly to adopt the ASCO-Friends of Cancer Research recommendations on broadening eligibility criteria for clinical trials and to amend the CTEP protocol template accordingly and it is encouraging to see many of these recommendations now being incorporated in NCI-sponsored trials. Time will tell if doing so leads to more diverse and representative clinical trial populations.
Many of the challenges of conducting cancer clinical trials are created or perpetuated by the clinical research community, often for no good reason. A fundamental tenet of the ASCO-Friends approach to broadening eligibility criteria is that all inclusion and exclusion criteria should be based on considerations of patient safety, justified by preclinical or clinical data and be related to the scientific objectives of the trial. The same is true of data collection.
We should collect what we need to achieve the study objectives and no more. Experience shows that this approach can be effectively implemented for non-IND studies. ASCO launched the Targeted Agent and Profiling Utilization Registry (TAPUR) study in 2016 with this guiding principle in mind. TAPUR is a prospective, multi-center, multi-basket, phase II clinical trial that seeks to identify signals of activity for marketed, targeted drugs used outside of their approved indications for treatment of tumors with a relevant genomic alteration.
One of ASCO’s goals in designing TAPUR was to demonstrate that cancer clinical trials can indeed be simplified, launched quickly and enroll a broadly representative patient population. The trial was launched less than a year after approval by the ASCO board of directors, and more than 2,500 patients have now been enrolled.
The study continues to accrue in a network of more than 250 clinical sites, mostly community-based. From the start, TAPUR included the broad eligibility criteria recommended by ASCO-Friends and required only the minimum data collection necessary to address the study objectives. As all drugs in the study are already marketed, TAPUR collects only grade 3-5 AEs and SAEs and does not collect concomitant medications.
No new or unexpected safety signals have been observed in any of the data reported thus far. The study has a number of other pragmatic features that were implemented long before the COVID-19 pandemic required more flexibility in trial procedures such as allowing certain patient evaluations to be performed at facilities other than the clinical trial site and shipping oral study medications directly to patients.
It was certainly energizing to read that Dr. Bertagnolli has called on us to redesign cancer clinical trials and to be faster and nimbler in doing so. We can’t wait decades for NCI to implement changes to its clinical trials, particularly when data to support changes like limiting AE collection for non-IND trials have been around for years.
No other country in the world has a publicly supported cancer clinical trials program like the NCTN, a program that, over 65 years, has brought many practice-changing, lifesaving therapies to patients. I’m excited to see what other changes the NCI has in store for its clinical trials, but I hope we will not have to wait years to see them.
