Abstract 872P
Background
Improvement in OS is the gold standard for assessing treatment efficacy. To avoid delays in novel treatment approval, early treatment decisions may be based on validated surrogate endpoints which predict OS. A prior meta-analysis of clinical trials in LA HNSCC pts by Michiels 2009 showed a strong correlation of EFS with OS, supporting EFS as a surrogate endpoint in LA HNSCC. This study aimed to assess whether the correlation between EFS and OS is maintained with updated literature in pts with LA HNSCC ineligible for surgery and receiving chemoradiation therapy (CRT).
Methods
A systematic literature review (SLR) conducted on May 20, 2020 identified randomized controlled trials assessing CRT regimens in the surgery ineligible LA HNSCC population. The primary endpoints of interest were OS and EFS, defined as time from randomization to progression, surgery, or death. Trials with progression-free survival, disease- and failure-free survival endpoints were included if definitions were sufficiently similar to EFS. Trials reporting hazard ratios (HRs) or Kaplan-Meier curves for OS and EFS were included in the analysis. Correlation was assessed with regression models across all CRT trials, as well as in subgroups defined by type of CRT (concurrent CRT, sequential CRT and cisplatin + RT). The strength of the relationship between EFS log HR and OS log HR was measured with Pearson’s correlation coefficient (R).
Results
The SLR identified 27 trials with OS and EFS results. A strong correlation between EFS and OS was observed in the overall CRT analysis (R=0.85, 95% confidence interval [CI]: 0.70-0.93). Similar results were found by type of CRT, and in studies with alternative EFS definitions (see table). Table: 872P
| Analyses a | R (95% CI) | Slope | Intercept |
| All CRT trials (n=27) | 0.85 (0.70, 0.93) | 0.825 | 0.033 |
| Concurrent CRT vs concurrent CRT trials (n=12) | 0.88 (0.62, 0.97) | 0.707 | 0.12 |
| Sequential CRT vs any CRT b trials (n=15) | 0.83 (0.37, 0.96) | 1.155 | 0.107 |
| Cisplatin + RT vs any CRT b trials (n=15) | 0.79 (0.47, 0.93) | 0.694 | 0.031 |
| Cisplatin + RT vs concurrent CRT trials (n=9) | 0.82 (0.33, 0.96) | 0.653 | 0.028 |
| All trials where EFS was defined as time from randomization to disease progression or death (n=19) | 0.87 (0.69, 0.95) | 0.737 | 0.006 |
a Conducted for all trials and subgroups of trials based on the use of any concurrent CRT, adjuvant/neoadjuvant (‘sequential’) CRT, or concurrent cisplatin + RT in their armsb Concurrent or sequential CRT
Conclusions
Associations between OS and EFS were strong (R=0.79-0.88), suggesting EFS is a valid surrogate for OS in surgery ineligible LA HNSCC pts receiving CRT.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Merck & Co., Inc.
Funding
Merck & Co., Inc.
Disclosure
C.M. Black: Financial Interests, Personal and Institutional, Full or part-time Employment: Merck & Co Inc; Financial Interests, Institutional, Stocks/Shares: Merck & Co Inc. S. Keeping: Financial Interests, Institutional, Funding, Employee of PRECISIONheor, which received funding from Merck & Co, Inc. to conduct the study.: PRECISIONheor. D. Chirovsky: Financial Interests, Institutional, Full or part-time Employment: Merck and Co., Inc.; Financial Interests, Institutional, Stocks/Shares: Merck and Co, Inc. K. Ramakrishnan: Financial Interests, Institutional, Full or part-time Employment: Merck & Co., Inc.; Financial Interests, Institutional, Ownership Interest: Merck & Co., Inc. A. Mojebi: Financial Interests, Institutional, Funding, Employee of PRECISIONheor, which received funding from Merck & Co, Inc. to conduct the study.: PRECISIONheor. N. Upadhyay: Financial Interests, Institutional, Full or part-time Employment: Merck & Co., Inc. D. Ayers: Financial Interests, Institutional, Funding, Employee of PRECISIONheor, which received funding from Merck & Co, Inc. to conduct the study.: PRECISIONheor.