The ESMO GI 2026 delivered several important advances in colorectal cancer, highlighting the continued evolution of precision oncology across both early-stage and metastatic disease.
Among the most discussed presentations were KRYSTAL-10, CAPRI-2 GOIM, ATOMIC, and BREAKWATER. Together, these studies addressed some of the most relevant questions in colorectal cancer today: the role of chemotherapy-free targeted therapy, ctDNA-guided treatment sequencing, optimization of adjuvant immunotherapy, and the molecular mechanisms underlying response and resistance to targeted combinations.
To better understand the clinical implications of these findings, OncoDaily spoke with Davide Ciardiello, MD, PhD, medical oncologist at the Division of Gastrointestinal Medical Oncology and Neuroendocrine Tumours, European Institute of Oncology, IEO, IRCCS, Milan, Italy.
KRYSTAL-10: Adagrasib Plus Cetuximab vs Chemotherapy in KRAS G12C-Mutated mCRC
KRYSTAL-10 (NCT04793958) was a global, open-label, randomized phase 3 trial evaluating adagrasib plus cetuximab as second-line treatment for patients with KRAS G12C-mutated metastatic colorectal cancer who had progressed after first-line fluoropyrimidine-based doublet chemotherapy.
A total of 461 patients were randomized to receive adagrasib 600 mg twice daily plus cetuximab 500 mg/m² every two weeks (n=231) or investigator’s choice of FOLFIRI or mFOLFOX6 with or without VEGF/VEGFR inhibitors (n=230). The dual primary endpoints were progression-free survival by blinded independent central review and overall survival.
Median progression-free survival was 7.5 months with adagrasib plus cetuximab versus 8.1 months with chemotherapy (HR 0.89; 95% CI 0.71–1.13; p=0.32). At a minimum follow-up of 26.4 months, median overall survival was 21.6 versus 21.7 months, respectively (HR 0.83; 95% CI 0.67–1.03; p=0.09). The study did not meet either of its primary endpoints.
Despite this, the targeted combination demonstrated substantial antitumor activity, achieving an objective response rate of 47% compared with 16% for chemotherapy, with complete responses in 7% versus fewer than 1% of patients. Grade 3–5 treatment-related adverse events occurred in 46% and 55% of patients, respectively, with no new safety signals.
Commenting on how clinicians should interpret this apparent disconnect between tumour response and longer-term outcomes, Davide Ciardiello stated:
“Findings of the KRYSTAL-10 study should be interpreted with caution. In particular, the standard arm performed much better than expected (PFS of 8 months respected historical data of 4–5 months). Moreover, higher proportion of patients in the chemotherapy arm received subsequent KRASG12C inhibitors (4% vs 30%), and this may have influenced OS results.
In post hoc accelerated failure time modeling analyses, which adjusted for the potential effects on OS of any confounders including subsequent anti-KRASG12C inhibitor use in patients included in the control arm, adagrasib plus cetuximab was associated with a clinically and statistically significant survival benefit.
Overall, findings of the KRYSTAL-10 study support the role of target therapy for patients with KRASG12C mutant mCRC, however further evidence are required to identify the best combinatory strategy. In this scenario, different trials are evaluating the combination of KRASG12C inhibitor with anti-EGFR monoclonal antibodies and chemotherapy.”
CAPRI-2 GOIM: Cetuximab Continuation in Molecularly Selected mCRC
CAPRI-2 GOIM (NCT05312398) was a biomarker-driven phase 2 study evaluating cetuximab-based treatment across sequential therapy lines in patients with RAS and BRAF V600E wild-type, microsatellite-stable metastatic colorectal cancer.
Patients received first-line FOLFIRI plus cetuximab followed by FOLFOX plus cetuximab after disease progression. Plasma circulating tumour DNA comprehensive genomic profiling was performed at baseline and at each progression to identify patients without resistance-associated alterations across a 14-gene panel (KRAS, NRAS, BRAF, PTEN, MAP2K1, NF1, FGFR, ROS1, MET, HER2, EGFR-ECD, PIK3CA, RET, and NTRK) — defined as “negative hyperselected” — and those with one or more variants, defined as “positive hyperselected.”
Among 182 patients treated in the first-line setting, 73% were negative hyperselected. These patients achieved a significantly higher objective response rate (78% vs 54%; OR 3.00; 95% CI 1.42–6.38; p=0.003) and longer median progression-free survival (13.8 vs 9.6 months; HR 0.55; 95% CI 0.39–0.78; p=0.001).
Seventy-five patients subsequently received second-line FOLFOX plus cetuximab. Median progression-free survival remained longer in the negative hyperselected group (8.3 vs 4.65 months; HR 0.43; 95% CI 0.25–0.75; p=0.003), while objective response rate was 28% vs 20% (not significant). Overall survival from baseline was not reached compared with 22.4 months in positive hyperselected patients (HR 0.47; 95% CI 0.30–0.73; p=0.001). Overall survival from the start of second-line therapy was 27.4 versus 15.9 months, a difference that was numerically meaningful but did not reach statistical significance (HR 0.60; p=0.170).
These findings raised the possibility that cetuximab may not need to be automatically discontinued at first progression in patients whose tumours remain negative hyperselected by plasma ctDNA. Discussing whether this strategy is sufficiently supported for clinical use, Davide Ciardiello stated:
“Within the limitation of a “proof of concept” single arm phase II trial, the CAPRI-2 GOIM study supports the potential role of continuing anti-EGFR blockade after progression to first-line therapy in patients with “molecularly” selected tumors by changing the chemotherapy backbone. However, despite the promising data further evidence is required to confirm this treatment strategy before integrating into clinical practice.
In this scenario the CAPRI-3 GOIM study (NCT07389265) is currently ongoing. After progression to anti-EGFR first-line therapy a liquid biopsy will be performed. In case of RAS/BRAF, PIK3Ca, EGFR-EDC wild-type and ERBB2 non-amplified tumors patients will be randomized 1:1 to receive a standard chemotherapy doublet (FOLFOX or FOLFIRI, depending on the regimen used in the first line) combined either with cetuximab (experimental arm) or with bevacizumab (control arm).”
ATOMIC: Does Treatment Duration Matter?
ATOMIC (A021502/AIO-KRK-0317) presented a retrospective exploratory secondary analysis of the phase 3 trial in patients with stage III deficient mismatch repair colon cancer. Patients received six months of adjuvant mFOLFOX6 with or without twelve months of atezolizumab.
This secondary analysis used a six-month landmark and included patients who received at least one treatment cycle and were alive and disease-free at six months. It explored whether disease-free survival differed according to the number of mFOLFOX6 cycles received or the duration of atezolizumab.
Among patients who received six or fewer cycles of mFOLFOX6, three-year disease-free survival was nearly identical between treatment arms (80.0% vs 79.9%; adjusted HR 0.84; 95% CI 0.35–2.01). Among patients who received more than six cycles, three-year disease-free survival favored atezolizumab plus chemotherapy (88.2% vs 79.1%; adjusted HR 0.45; 95% CI 0.29–0.71). Importantly, the rate of atezolizumab continuation after FOLFOX discontinuation differed substantially between these groups: 48% in the six-or-fewer-cycle group versus 87% in the more-than-six-cycle group, which may help explain the differential benefit observed.
When immunotherapy duration was analyzed, three-year disease-free survival was 84.0% among patients receiving fewer than twelve cycles of atezolizumab and 88.0% among those receiving at least twelve cycles (adjusted HR 0.81; 95% CI 0.38–1.71; p=0.58), a difference that was not statistically significant.
It remained uncertain whether the differential benefit reflected a true interaction between chemotherapy duration and immunotherapy or was influenced by patient selection and confounding factors. Commenting on this question, Davide Ciardiello stated:
“An excellent question that is truly difficult to answer, considering the various confounding factors. In the treatment of localized colon cancer, results from previous studies have demonstrated that 6-month adjuvant chemotherapy was necessary for high-risk patients. In case of early discontinuation of oxaliplatin and continuation of fluoropyrimidine therapy, completing at least 3 months of treatment did not demonstrate a detrimental impact on survival. Similarly, the results of the ATOMIC study demonstrated that, when combining atezolizumab with FOLFOX, at least 6 cycles of oxaliplatin-based therapy are required.
In the metastatic setting, the COMMIT trial proved that combination of FOLFOX/bevacizumab with atezolizumab compared with atezolizumab single agent/chemotherapy provided longer PFS. Therefore, with all the limitation of cross-trial comparison, we might hypothesize that for this combinatory strategy chemotherapy retains a significant role and should not be early discontinued.”
The exploratory findings also raised a practical question regarding patients who cannot complete the planned duration of chemotherapy or atezolizumab because of toxicity. Addressing whether the data are sufficiently mature to guide individualized treatment duration, Ciardiello added:
“I believe the study results are very interesting and could help optimize treatment efficacy while reducing toxicity. However, these results should be interpreted with caution, given the subgroup analyses involving small numbers of patients. In this context, real-world clinical data could provide important insights.”
BREAKWATER: Molecular Correlates of Response and Resistance
The phase 3 BREAKWATER study (NCT04607421) evaluated first-line encorafenib plus cetuximab with mFOLFOX6 compared with chemotherapy with or without bevacizumab in patients with BRAF V600E-mutant metastatic colorectal cancer. The main trial previously demonstrated statistically significant improvements in objective response rate, progression-free survival, and overall survival with the triplet combination.
The translational analysis presented at ESMO GI 2026 included whole-exome sequencing and whole-transcriptome sequencing of baseline tumour samples from 498 and 497 patients respectively, together with end-of-treatment circulating tumour DNA profiling in 310 patients.
Overall survival benefit with encorafenib plus cetuximab and mFOLFOX6 was maintained across major genomic alterations, including APC, PIK3CA, RNF43, and TP53, as well as across consensus molecular subtypes and BRAF-mutant transcriptional subtypes.
Patients with elevated UV response pathway activity derived particularly pronounced benefit from the triplet combination versus control (HR 0.39; 95% CI 0.25–0.60), compared with encorafenib plus cetuximab alone versus control (HR 0.71; 95% CI 0.46–1.11), suggesting a meaningful contribution of mFOLFOX6 in this molecular context.
Commenting on the broad applicability of the regimen across the molecular heterogeneity of BRAF V600E-mutant metastatic colorectal cancer, Davide Ciardiello stated:
“The combination treatment of encorafenib and cetuximab with chemotherapy (FOLFOX/FOLFIRI) has represented a “Copernican revolution” for the treatment of BRAF V600E-mutated, MSS metastatic colorectal cancer by reverting the negative prognostic impact of the mutation. Translational data from the study demonstrate that targeted therapy combined with chemotherapy is effective regardless of the factors analyzed. Previously, RNF43 alteration had been proposed as a biomarker of response to therapy with encorafenib/cetuximab +/- binimetinib. Results from the BREAKWATER study analysis demonstrated that the combination of encorafenib and cetuximab with chemotherapy is effective regardless of RNF43 status.”
The analysis also showed that adding mFOLFOX6 reduced the emergence of acquired MAPK-pathway resistance alterations compared with encorafenib plus cetuximab alone. Discussing the implications for treatment maintenance and sequencing, Ciardiello added:
“As with anti-EGFR treatment in RAS/BRAF wild-type patients, combining targeted therapy with chemotherapy has been shown to reduce the emergence of resistance-associated alterations. For this reason, I believe that following induction chemotherapy with two agents, the best strategy is to continue maintenance treatment with fluorouracil combined with encorafenib/cetuximab, rather than using targeted therapy alone. It will also be of interest to evaluate potential liquid-biopsy-guided reintroduction or rechallenge strategies after progression to first-line therapy in patients without resistance alterations.”
Key Takeaway
Taken together, KRYSTAL-10, CAPRI-2 GOIM, ATOMIC, and BREAKWATER illustrated how biomarker-driven treatment is reshaping colorectal cancer management while also reaffirming the continuing importance of chemotherapy. Reflecting on the overall message from these studies, Davide Ciardiello concluded:
“Data from the four studies demonstrate how biomarker-based personalized therapy can transform the treatment landscape for colorectal cancer in both early-stage and advanced disease settings. In this context, colorectal cancer proves to be a malignancy where chemotherapy maintains a fundamental role, including when combined with targeted therapies in molecularly selected tumors (e.g in “EGFR dependent tumors”, BRAFV600E or KRASG12C mutant mCRC).
For patients with MSI-H/dMMR tumors, further evidence is needed to understand the role of adjuvant chemotherapy combined with immunotherapy compared to neoadjuvant or adjuvant treatments involving ICIs combinations.”
Read more about ctDNA Blood Tests in Colorectal Cancer Screening on OncoDaily.