The KRAS-Variant and Cetuximab in HPV-Positive Oropharyngeal Cancer in NRG/RTOG 1016

PURPOSE/OBJECTIVE(S)

NRG/RTOG 1016 was a non-inferiority phase III trial comparing radiation with cisplatin versus cetuximab monotherapy for patients with HPV-positive oropharyngeal squamous cell carcinoma (SCC). The trial did not meet the non-inferiority criteria for overall survival (OS) and had significantly worse progression-free survival (PFS) and locoregional failure (LRF) in patients treated with cetuximab. Based on prior evidence that HNSCC patients with a germ-line mutation in KRAS (the KRAS-variant) had a positive response to radiation with cisplatin plus cetuximab without increased toxicity, samples from RTOG 1016 were used to test the protocol-specified hypothesis that KRAS-variant patients will have better outcomes when receiving IMRT + cetuximab monotherapy compared to IMRT + cisplatin.

MATERIALS/METHODS

The KRAS-variant was tested in 562 samples at MiraDx, a CLIA-certified laboratory. OS, PFS, LRF, and distant metastases (DM) were as defined per the RTOG 1016 protocol, and hazard ratios (HRs) were estimated by (cause-specific) Cox models. Negative binomial regression was used to model the number of treatment-related acute and late (≤ and > 180 days from end of treatment, respectively) grade 3-5 adverse events. To assess the predictive role of the KRAS-variant, all models included KRAS, assigned treatment, and their interaction, with the interaction tested at two-sided 0.05. HRs and toxicity ratios are expressed as IMRT + cetuximab / IMRT + cisplatin.

RESULTS

The prevalence of the KRAS-variant was 16% with similar patient and tumor characteristics and well-balanced treatment arms for variant and non-variant patients. Median follow-up was 8.6 years. There was no significant interaction between KRAS and treatment for OS (p = 0.99), PFS (p = 0.56), LRF (p = 0.09), or DM (p = 0.19) (Table 1). In KRAS-variant patients the mean acute and late toxicity ratios were 0.53 (95% CI 0.36, 0.80) and 1.62 (95% CI 0.57, 4.62). In non-variant patients, the mean acute and late toxicity ratios were 0.80 (95% CI 0.67, 0.95) and 0.55 (95% CI 0.35, 0.87), respectively. The interaction of KRAS and treatment was not significant for acute (p = 0.07) or late toxicity (p = 0.07).

CONCLUSION

While this study does not directly refute prior evidence that KRAS-variant patients benefit from radiation + cisplatin and cetuximab, this study does not support the hypothesis that the KRAS-variant is a predictive biomarker of improved outcome in HPV+ oropharyngeal SCC patients treated with IMRT + cetuximab alone, and suggests that for KRAS-variant patients, potential benefits in LRF and acute toxicity with cetuximab may be offset by worse DM and worse late toxicity.