This article is more than 5 years old. Information may no longer be current.
DFS remains viable surrogate for OS in adjuvant colon cancer trials
Click Here to Manage Email Alerts
CHICAGO — Three-year DFS remained a viable surrogate endpoint for OS in recent adjuvant colon cancer trials, according to study results presented at ASCO Annual Meeting.
Trial-level correlation between the two outcomes increased with longer OS follow-up.
“These results assure us that 3-year DFS is still an appropriate endpoint in the modern treatment era,” Mohamed E. Salem, MD, medical oncologist at Levine Cancer Institute at Atrium Health and a HemOnc Today Next Gen Innovator, said in an interview. “They also show us that the optimal duration of follow-up for survival should be longer than we previously used.”
Prior research from the ACCENT collaborative group — published more than a decade ago in Journal of Clinical Oncology — validated 3-year DFS as a surrogate endpoint for 5-year OS.
Since then, 3-year DFS has been used as a primary endpoint in many adjuvant colon cancer trials — including the practice-changing IDEA study, which showed 3 months of adjuvant chemotherapy conferred similar efficacy with less toxicity than the standard 6-month regimen for patients with low-risk stage III colon cancer.
However, many new therapeutic approaches — including biologics and targeted therapy — have been adopted since 3-year DFS became a standard endpoint. In addition, more recent data have shown improved survival after recurrence and longer OS among patients who received adjuvant FOLFOX, which consists of folinic acid, fluorouracil and oxaliplatin.
“With so many more treatment options leading to improved survival after recurrence and longer OS, we needed to answer the question of whether 3-year DFS still a good surrogate endpoint for OS in modern-era adjuvant trials,” Salem told HemOnc Today. “We also wanted to identify the ideal follow-up for OS to accurately predict outcomes.”
Salem and colleagues used the ACCENT database to evaluate individual patient data from eight randomized trials of adjuvant colon cancer therapy — consisting of chemotherapy with or without biologics — conducted from 1998 to 2009.
The trials — which had median follow-up for survival ranging from 5 to 10 years — included a combined 18,886 patients (median age, 60 years; 54% male; 83% stage III disease).
Three of those trials evaluated anti-VEGF or anti-epidermal growth factor receptor agents.
Investigators used linear regression (R2WLS) and Copula bivariate (R2Copula) models to examine the correlation of treatment effect estimates of 3-year DFS and 5-year to 8-year OS. R2 values closer to 1 indicated a stronger correlation.
Prespecified criteria for surrogacy required either R2WLS or R2Copula to be at least 0.8, with neither being less than 0.7. Researchers set a lower-bound 95% CI of greater than 0.6.
Results revealed strong trial-level correlation between DFS (median 3-year follow-up) and OS. Correlations strengthened as median follow-up for OS increased:
5-year OS follow-up: R2WLS 0.74 (95% CI, 0.54-0.94) and R2Copula 0.89 (95% CI, 0.77-1).
6-year OS follow-up: R2WLS 0.86 (95% CI, 0.74-0.98) and R2Copula 0.93 (95% CI, 0.85-1).
6.5-year OS follow-up: R2WLS 0.93 (95% CI, 0.86-1) and R2Copula 0.94 (95% CI, 0.88-1).
7-year OS follow-up: R2WLS 0.92 (95% CI, 0.76-1) and R2Copula 0.95 (95% CI, 0.89-1).
8-year OS follow-up: R2WLS 0.87 (95% CI, 0.62-1) and R2Copula 0.97 (95% CI, 0.92-1).
“This shows you in the current treatment era, it is better to wait longer than 5 years, perhaps 6 or 7 years, to see the true effect,” Salem told HemOnc Today.
“Patients who undergo adjuvant treatment for colon cancer are living longer than ever before,” he added. “It is very important to have something that provides an early indication of treatment benefit, and we believe these results can help guide the next generation of adjuvant clinical trial design.” – by Mark Leiser
Reference s :
Grothey A, et al. N Engl J Med. 2018;doi:10.1056/NEJMoa1713709.
Sargent DJ, et al. J Clin Oncol. 2005;23:8664-8770.
Sargent DJ, et al. J Clin Oncol. 2007;25:4569-4574.
Salem ME, et al. Abstract 724. Presented at: Gastrointestinal Cancers Symposium; Jan. 18-20, 2018; San Francisco.
Shi Q, et al. Abstract 3502. Presented at: ASCO Annual Meeting; May 31-June 4, 2019; Chicago.
Disclosure s : Salem reports consultant/advisory roles and speakers bureau roles with Genentech/Roche and Taiho Pharmaceutical. Please see the abstract for all other authors’ relevant financial disclosures.