ctDNA-Guided Selection Optimizes Anti-EGFR Rechallenge in Metastatic Colorectal Cancer

Key Clinical Takeaways: 

•  In the phase 2 CAVE-2 GOIM trial, adding avelumab to cetuximab did not improve progression-free or overall survival compared with cetuximab rechallenge alone in RAS/BRAF wild-type metastatic colorectal cancer.
• ctDNA profiling identified a “negative hyperselection” subgroup without resistance-associated genomic alterations that experienced significantly improved progression-free and overall survival.
• These findings support the use of liquid biopsy-based comprehensive genomic profiling to guide anti-EGFR rechallenge decisions in the continuum of care for metastatic colorectal cancer.

Stefania Napolitano, MD, PhD, University of Campania Luigi Vanvitelli, Naples, Italy, discusses results from the phase 2 CAVE-2 GOIM trial evaluating anti-EGFR rechallenge strategies in patients with RAS/BRAF wild-type metastatic colorectal cancer selected using ctDNA profiling. 

For part 1, please click here. 

Transcript: 

My name is Stefania Napolitano, I work in medical oncology and precision medicine at the University of Campania Luigi Vanvitelli in Naples. 

At the data cutoff of the study, September 2025, we had 8 patients still on treatment in the combination arm and only 2 patients still on treatment in the cetuximab alone arm. There were no significant differences between the 2 arms. Of note, in the CAVE-2 GOIM study, 27% of the patients in the 2 groups received 3 or more lines of therapy for metastatic disease, indicating the inclusion of a heavily pretreated population.

Also, a preplanned exploratory analysis was done to evaluate the impact of additional anti-EGFR drug–resistant pathogenic biomarkers. In particular the molecular hyperselection, was based on plasma ctDNA analysis for the presence of the genes KRAS, NRAS, BRAF, EGFR extracellular domain, PI3K exon 20, MAPK, AKT, MET, and PTEN mutations, and HER2 amplification. Patients with the presence of these alterations at plasma ctDNA are defined as positive hyperselected. Patients with no presence of these genetic alterations are defined as negative hyperselected. 

The results in the intention-to-treat population showed that the progression-free survival of cetuximab plus avelumab versus cetuximab alone was 5.3 months versus 4.3 months. The overall survival for the combination arm was 14.8 months, and the median overall survival for the cetuximab arm was 12.8 months. The response rate for patients treated with cetuximab plus avelumab was 12%, versus 8% for patients treated with only cetuximab. 

Since the presence of high tumor mutational burden (TMB) and the absence of liver metastases are considered putative biomarkers that might predict the antitumor activity of immunotherapy in microsatellite-stable metastatic colorectal cancer, we also explored the impact of these factors on antitumoral efficacy. The presence of high TMB was a negative prognostic factor and was associated with limited response, so with very similar median progression-free survival and overall survival for both arms, cetuximab plus avelumab or cetuximab alone. 

While in the subgroup of patients without the presence of liver metastases, the combination cetuximab plus avelumab had improved clinical activity compared with cetuximab monotherapy, in this subgroup of patients, those who received the combination had partial responses, whereas no patients in the cetuximab monotherapy arm obtained a partial response.

In addition, we also conducted preplanned exploratory analyses on the presence of additional resistance genes to anti-EGFR treatment, the so-called positive and negative hyperselected patients. We performed this analysis on the overall population and for the 2 treatment arms. In particular, we had 124 patients who were so-called negative hyperselected because they did not have any alterations in genes correlated with anti-EGFR drug resistance. 

The median progression-free survival and median overall survival for negative compared with positive hyperselected patients were 5.35 months versus 3.65 months. The overall survival was 15 months for the negative hyperselected patients and 11.1 months for the positive hyperselected patients. The response rate was 12% for negative hyperselected tumors and 3% for positive hyperselected tumors. 

For patients treated with cetuximab monotherapy, the median progression-free survival was 5.1 months for negative hyperselected patients versus 2.8 months for positive hyperselected patients, and the overall survival was also higher for negative hyperselected patients compared with positive hyperselected patients. A similar difference in median progression-free survival and overall survival was also observed for patients treated with cetuximab plus avelumab, according to negative versus positive hyperselection. 

Regarding safety, there were no unexpected safety signals, with no treatment-related grade 3 or higher adverse events. There were no treatment discontinuations, and as expected, the most frequent grade 3 adverse event was skin rash for patients receiving either cetuximab plus avelumab or cetuximab alone. The percentage was 8% in the combination arm and 4% in the cetuximab arm.

There is also a translational study design for the CAVE-2 GOIM study. In fact, thanks to the comprehensive clinical data collection from all centers participating in the CAVE-2 study, several analyses are still ongoing, including updates of the study primary end point, which is overall survival, performing tailored subgroup analyses, and defining molecular characterization for patients enrolled in the study at the time of disease progression. Liquid biopsy by FoundationOne was also performed at disease progression. These analyses will be very helpful to further elucidate the clinical and molecular determinants of response and resistance, supporting a deeper understanding of patient heterogeneity within the CAVE-2 population.

In conclusion, the CAVE-2 GOIM study did not demonstrate a survival benefit from the addition of avelumab to cetuximab rechallenge. However, within the limitations of a randomized phase 2 trial, these findings underscore the critical role of comprehensive and accurate molecular stratification in predicting the clinical efficacy of anti-EGFR rechallenge strategies. Importantly, a subset of metastatic colorectal cancer remains dependent on EGFR signaling, supporting the continued use of biomarker-guided targeted therapy even in the refractory setting.

Source: 

Ciardiello D, Martini G, Bielo LB, et al. Cetuximab rechallenge in molecularly selected metastatic colorectal cancer: the randomized CAVE-2 GOIM trial. Ann Oncol. Published online: December 22, 2025. doi: 10.1016/j.annonc.2025.12.014