Abstract 951P
Background
With the successes of immunotherapy development across cancer indications, including anti–PD-(L)1 antibodies, immunotherapy is now a prevailing treatment in HNSCC trials. OS is the gold standard endpoint for evaluating clinical efficacy, especially in pivotal oncology trials, but establishing OS benefit in randomised studies requires large investments. In early phase, single-arm trials, ORR and PFS have been widely used as surrogate endpoints for OS; establishing such surrogate endpoints could improve efficiency of clinical development in terms of patient numbers, time, and cost. The relationship among ORR, PFS, and OS may vary by cancer type, treatment mechanism of action, and disease setting, and has not yet been published for immune-oncology (I-O)–based regimens investigated for recurrent/metastatic (R/M) HNSCC. A systematic review and meta-analysis were conducted to investigate such association among OS, PFS, and ORR in R/M HNSCC.
Methods
A systematic literature review was performed to identify relevant Phase II and III trials published since 2015 that investigated I-O treatments and I-O/chemotherapy combinations for R/M HNSCC. In total, 40 subpopulations in five trials were included in the meta-analysis. The meta-analysis investigated associations between ORR and log-transformed (LT) median OS (mOS), between ORR and LT median PFS (mPFS), and between LT mPFS and LT mOS. The statistical model was conducted using a weighted linear regression model with weights proportional to sample size of each subpopulation and included therapy line as covariate.
Results
A statistically significant linear relationship was identified between ORR and LT mOS (R2=0.78, p<0.0001), and between LT mPFS and LT mOS (R2=0.79, p<0.0001). A weak but statistically significant linear relationship was identified between ORR and LT mPFS (R2=0.26, p=0.018).
Conclusions
The meta-analysis showed significant correlation between ORR and OS, and between PFS and OS, supporting ORR and PFS as surrogate endpoints to predict OS effect in early phases of clinical trials evaluating I-O treatments in R/M HNSCC.
Clinical trial identification
Editorial acknowledgement
Editorial assistance in the form of copy editing and collating author comments was provided by Victoria Hunter, MSc, at Fishawack Indicia Ltd, UK, and was funded by GlaxoSmithKline.
Legal entity responsible for the study
GlaxoSmithKline.
Funding
GlaxoSmithKline.
Disclosure
X. Wang, H. Chen, A. Slowley, C. Ellis, D.C. Turner, H. Zhou: Shareholder/Stockholder/Stock options, Full/Part-time employment: GlaxoSmithKline. M. Ballas: Leadership role, Travel/Accommodation/Expenses, Shareholder/Stockholder/Stock options, Licensing/Royalties, Full/Part-time employment: GlaxoSmithKline; Shareholder/Stockholder/Stock options: Bristol-Myers Squibb; Licensing/Royalties: AstraZeneca. K.F. Bell: Full/Part-time employment: GlaxoSmithKline; Spouse/Financial dependant: Humana Pharmacy (Immediate Family member); Spouse/Financial dependant: CVS. C. Baccan: Shareholder/Stockholder/Stock options, Full/Part-time employment: GlaxoSmithKline; Shareholder/Stockholder/Stock options: Bristo-Myers Squibb.