The ASCO Cancer Research Committee recently convened four disease-specific working groups—in pancreas, breast, lung, and colon cancers—to “consider the design of future clinical trials that would produce results that are clinically meaningful to patients.” An ASCO perspective statement, reported in the Journal of Clinical Oncology by Lee M. Ellis, MD, of The University of Texas MD Anderson Cancer Center, Houston, and colleagues, offers their conclusions.1
The statement furnishes the criteria for clinically meaningful improvement arrived at by the working groups for phase III trials in (1) pancreatic cancer patients who are eligible for FOLFIRINOX, gemcitabine-, or gemcitabine/nab-paclitaxel, (2) lung cancer patients with non–squamous cell or squamous cell carcinoma, (3) patients with metastatic triple-negative breast cancer with no previous treatment for metastatic disease, and (4) patients with colon cancer who have had disease progression on all prior therapies or are not candidates for standard second- or third-line options.
Each working group included clinical investigators, patient advocates, biostatisticians, U.S. Food and Drug Administration oncologists, and industry oncologists. Working Group Chairs were Emile Voest, MD, PhD, Netherlands Cancer Institute, Amsterdam, for the pancreas cancer group, Roy Herbst, MD, PhD, Yale Cancer Center, New Haven, Connecticut, for the lung cancer group, Lowell E. Schnipper, MD, Beth Israel Deaconess Medical Center, Boston, for the breast cancer group, and Alan P. Venook, MD, Helen Diller Family Comprehensive Cancer Center at University of California San Francisco, for the colon cancer group.
‘Smaller and Smarter’ Trials
The goals set forth in the perspective statement are meant to further the paradigm of designing and generating clinical trials that can demonstrate larger therapeutic gains in selected groups of patients with smaller trial populations (“smaller and smarter” trials). Furtherance of this paradigm will reflect continued improvement in the ability to identify molecular drivers of cancer and to prospectively identify patients most likely to benefit from targeted treatments.
Progress in newer modalities such as immune therapies and antibody-drug conjugates, which are providing large therapeutic gains, should also facilitate this approach. Moreover, it is expected that genomic tests used to guide cancer treatment “will not only improve in sensitivity and specificity but also decrease the amount of biologic sample necessary and lower the cost and turnaround time to enable widespread use.”
Discussion in the working groups was not limited to biomarker-driven trials, although it is acknowledged that the goals established in the treatment settings discussed, as well as in other treatment settings, will require enrichment strategies. In the settings considered by the working groups in the current perspective statement, there currently are no validated biomarkers available to select patients for specific treatments. However, as the authors note, “[W]e expect that over time, such biomarkers will be identified and that the goals set forth by these working groups will be achievable.”
Primary Goal
In arriving at goals for clinical trials, both survival and quality of life were considered as important to outcomes that are clinically meaningful for patients. There was agreement in all working groups that quality of life is difficult to measure and interpret even when validated instruments are used, with the definition of a clinically meaningful change in global quality-of-life measures being acknowledged as a particular challenge.
The working groups concluded that serial assessment of specific cancer-related symptoms using validated instruments and shorter, more cancer-specific surveys can define a clinically meaningful outcome for patients. In the absence of agreed-upon measurements of quality of life for the settings considered in the current publication, no quality-of-life elements are included in the clinical trial goals. It was also recognized that quality-of-life issues are cancer-specific and therapy-specific.
The relationship of progression-free survival to overall survival in particular clinical contexts was extensively discussed within the working groups, with each deciding to use overall survival as the primary measure of clinically meaningful outcome. All groups acknowledged the challenges of using overall survival as the primary measure, including the need for longer follow-up and the potential confounding effect of crossover and other poststudy therapies in assessing overall survival.
Due, in part, to such considerations, the use of progression-free survival as a clinically meaningful endpoint was considered to be appropriate, and it is included in the current perspective statement as a secondary endpoint. The magnitude of overall survival benefit that should be considered clinically meaningful was also extensively discussed, with complete consensus not being achieved in the Breast Cancer Working Group, although compromise was reached.
Targets for Meaningfulness
The conclusions of the working groups are summarized below. All groups except the colon cancer group focused on patients with metastatic disease receiving first-line systemic treatment. All groups selected overall survival as the primary clinical endpoint, with all stipulating a hazard ratio (HR) ≤ 0.8 corresponding to improvement in median overall survival of 2.5 to 6 months, depending on setting, as the minimum incremental improvement over standard therapy that defines a clinically meaningful outcome.
Pancreatic Cancer—FOLFIRINOX-Eligible Patients: The current baseline median overall survival was estimated at 10 to 11 months. The minimum improvement over current overall survival considered clinically meaningful is 4 to 5 months, with a target hazard ratio of 0.67 to 0.69. As secondary endpoints, 1-year overall survival would be expected to improve to a minimum of 63% from the current rate of 48%, and progression-free survival is expected to improve by at least 4 to 5 months.
Pancreatic Cancer—Gemcitabine- or Gemcitabine/Nab-paclitaxel-Eligible Patients: The current baseline median overall survival was estimated at 8 to 9 months. The minimum improvement over current overall survival considered clinically meaningful is 3 to 4 months, with a target hazard ratio of 0.6 to 0.75. As secondary endpoints, 1-year overall survival would be expected to improve to a minimum of 50% from the current rate of 35%, and progression-free survival is expected to improve by at least 3 to 4 months.
Lung Cancer—Non–Squamous Cell Carcinoma: The current baseline median overall survival was estimated at 13 months. The minimum improvement over current overall survival considered clinically meaningful is 3.25 to 4 months, with a target hazard ratio of 0.76 to 0.8. As secondary endpoints, 1-year overall survival would be expected to improve to a minimum of 61% from the current rate of 53%, and progression-free survival is expected to improve by at least 4 months.
Lung Cancer—Squamous Cell Carcinoma: The current baseline median overall survival was estimated at 10 months. The minimum improvement over current overall survival considered clinically meaningful is 2.5 to 3 months, with a target hazard ratio of 0.77 to 0.8. As secondary endpoints, 1-year overall survival would be expected to improve to a minimum of 53% from the current rate of 44%, and progression-free survival is expected to improve by at least 3 months.
Breast Cancer—Metastatic Triple-Negative Disease, Previously Untreated for Metastatic Disease: The current baseline median overall survival was estimated at 18 months. The minimum improvement over current overall survival considered clinically meaningful is 4.5 to 6 months, with a target hazard ratio of 0.75 to 0.8. As secondary endpoints, 1-year overall survival would be expected to improve to a minimum of 71% from the current rate of 63%, and progression-free survival is expected to improve by at least 4 months.
Colon Cancer—Disease Progression With All Prior Therapies (or Not a Candidate for Standard Second- or Third-Line Options): The current baseline median overall survival was estimated at 4 to 6 months. The minimum improvement over current overall survival considered clinically meaningful is 3 to 5 months, with a target hazard ratio of 0.67. As secondary endpoints, 1-year overall survival would be expected to improve to a minimum of 35% from the current rate of 25%, and progression-free survival is expected to improve by at least 3 to 5 months.
Clarifying Remarks
The authors stated that the conclusions reached by the working groups are not intended to set standards for regulatory approval or insurance coverage, “but rather to encourage patients and investigators to demand more from clinical trials.”
They also observed that symptoms from cancer progression and tolerability of treatment are of critical importance when considering whether a new treatment is associated with a clinically meaningful outcome for patients. The authors stated:
For the most part, the working groups agreed that if a therapy is less toxic than prevailing treatments, a smaller improvement in efficacy is acceptable. Conversely, a highly toxic therapy should be accompanied by an expectation of substantially greater benefit to provide a clinically meaningful outcome to patients. ■
Disclosure: For full disclosures of the authors, visit jco.ascopubs.org.
Reference
1. Ellis LM, Bernstein DS, Voest EE, et al: American Society of Clinical Oncology perspective: Raising the bar for clinical trials by defining clinically meaningful outcomes. J Clin Oncol. March 17, 2014 (early release online).
