Efficacy of anti-epidermal growth factor antibody rechallenge in RAS/BRAF wild-type metastatic colorectal cancer: a multi-institutional observational study

The anti-epithelial growth factor receptor (EGFR) monoclonal antibodies (mAbs) cetuximab and panitumumab improve survival in patients with RAS wild-type (WT) metastatic colorectal cancer (mCRC), but do not yield a significant survival benefit in patients with right-sided or RAS mutant (MT) mCRC (Di Nicolantonio et al. 2021). Therefore, several guidelines recommend the use of anti-EGFR mAb as a first-line treatment for patients with left-sided RAS/BRAF WT mCRC (Hashiguchi et al. 2020). However, a recent report suggested that anti-EGFR mAbs might be a treatment option for patients with mCRC without baseline ctDNA alteration, such as KRAS, NRAS, PTEN, and extracellular domain EGFR mutations, HER2, MET amplifications, ALK, RET, and NTRK1 fusions, even in right-sided tumors (Shitara et al. 2024). Moreover, according to the ESMO Clinical Practice Guideline for mCRC, right-sided tumors might benefit less in terms of progression-free survival (PFS) and overall survival (OS) from treatment with anti-EGFR mAb compared with left-sided tumors, but anti-EGFR mAb may be effective in terms of tumor shrinkage regardless of tumor sidedness (Cervantes et al. 2023). While RAS status is typically assessed prior to the initiation of systemic chemotherapy, most patients develop acquired resistance even after an initial response to chemotherapy containing anti-EGFR mAb (Misale et al. 2012; Siravegna et al. 2016), indicating changes in cancer clones and/or genetic status. Therefore, changes in the RAS status following anti-EGFR mAb administration remain to be characterized. The development of liquid biopsy technology, especially for the assessment of circulating tumor DNA (ctDNA), has enabled the monitoring of real-time tumor-derived genetic alterations. Mutant RAS clones have been reported to emerge in the blood during anti-EGFR mAb treatment and to decline treatment discontinuation, suggesting that the clonal evolution relates to clinical progression (Morelli et al. 2015; Siravegna et al. 2015; Siena et al. 2018; Parseghian et al. 2019). Anti-EGFR mAb rechallenge is defined as the re-administration of anti-EGFR mAb after an anti-EGFR mAb-free period in patients showing resistance to prior chemotherapy.

The concept of anti-EGFR mAb rechallenge was first reported by Santini et al. They assessed the efficacy of cetuximab rechallenge in patients with mCRC and reported promising clinical outcomes (Santini et al. 2012). However, they did not appropriately distinguish between anti-EGFR mAb reintroduction and rechallenge. The CRICKET trial was the first multicenter phase II study to evaluate the efficacy of cetuximab rechallenge. The ad-hoc analysis of this study revealed that patients with mCRC with ctDNA RAS WT prior to the anti-EGFR mAb rechallenge had longer PFS than those with ctDNA RAS MT (Cremolini et al. 2019). An ad-hoc analysis of several clinical trials on anti-EGFR mAb rechallenge also showed an association between survival and RAS status in ctDNA (JACCRO CC-08 and 09, E-rechallenge) (Osawa et al. 2018; Sunakawa et al. 2020). In terms of combination therapy as anti-EGFR mAb rechallenge and other treatments besides irinotecan, the CAVE trial investigated rechallenge with cetuximab plus avelumab in refractory RAS WT mCRC; this combined therapy was well tolerated, and patients with RAS/BRAF WT ctDNA had significantly longer median PFS and OS than did those with RAS/BRAF MT ctDNA (Martinelli et al. 2021, Ciardiello D et al. 2022). The CHRONOS study, an open-label, single-arm, phase II trial, was the first trial to prospectively evaluate the efficacy of anti-EGFR mAb rechallenge based on the mutational status of ctDNA and showed that their efficacy was resumed (Sartore-Bianchi et al. 2022). These results suggest that evaluation of the RAS mutational status of ctDNA may help to select candidates for anti-EGFR mAb rechallenge.

Several previous studies have reported the incidence of ctDNA RAS MT before salvage-line treatment and a post-hoc pooled analysis of the CAVE and VELO trial investigated that 75.2% of patients retained RAS/BRAF WT ctDNA before rechallenge with anti-EGFR mAb (Ciardiello et al. 2023, Germani et al. 2024), as well as the clinicopathological features and predictors of the efficacy of anti-EGFR mAb rechallenge, except for pretreatment ctDNA RAS status (Ciardiello et al. 2024). However, it remains unclear whether these clinicopathological features would be significant factors in daily clinical practice. It is also necessary to identify other promising predictors of response and survival for anti-EGFR mAb rechallenge.

Therefore, this study aimed to investigate the incidence of ctDNA RAS MT before the initiation of salvage-line treatment and explore the clinicopathological features and molecular biological factors associated with the efficacy of anti-EGFR mAb rechallenge for tissue RAS/BRAF WT mCRC.

In this study, the frequency of ctDNA RAS MT was 40.5%, which was associated with high tumor burden such as liver metastasis and high tumor marker levels. On the contrary, NeoRAS, which is defined as conversion from initially diagnosed RAS MT to WT following treatment, was reportedly associated with the absence of liver metastasis, small tumor diameter, and low tumor markers (Osumi et al. 2023). A previous study reported that ctDNA RAS MT tended to be found in patients with liver metastases and significantly larger tumor sizes, and the false negative rate of ctDNA was low (Lim et al. 2021). The study revealed that anti-EGFR mAb rechallenge was less effective in patients with liver metastasis compared with those without (Ciardiello et al. 2024). These patients also have synchronous metastases and significantly higher tumor markers (Lim et al. 2021). Similarly, in our analysis, the rechallenge treatment was ineffective for patients with baseline high CA19-9. Our univariate and multivariate analyses suggested that patients with ctDNA RAS MT tended to have a short PFS. These results suggest that ctDNA RAS MT detectability may depend on tumor volumes, and ctDNA RAS status may be associated with the efficacy of anti-EGFR mAb rechallenge.

Minor RAS MT, other than KRAS exon2, was detected more frequently than major RAS MT. These results are consistent with previous findings (Morelli et al. 2015). Although the mechanism for this finding remains to be determined, the proportion of cancer cells with minor RAS MT was extremely small to be detected at the initial diagnosis. RAS codon 61 and 146 MT, which have weaker RAS-GTPase activity in the transforming assay, might have a lower growth advantage compared with RAS exon2 MT. Thereafter, these cancer cells with minor RAS MT increase under the stress of anti-EGFR mAb. These minor RAS MT may be associated with acquired resistance (Morelli et al. 2015).

In this study, the efficacy of first-line anti-EGFR mAb significantly influenced the clinical outcomes of anti-EGFR mAb rechallenge in clinical practice. Previous studies that enrolled patients who responded to prior chemotherapy containing anti-EGFR mAb as the inclusion criteria showed favorable clinical outcomes of anti-EGFR mAb rechallenge (Cremolini et al. 2019; Sartore-Bianchi et al. 2022). Thus, response to prior chemotherapy containing anti-EGFR mAb may be a clinical marker for predicting the response to anti-EGFR mAb rechallenge as a positive selection. However, considering that responders to prior chemotherapy containing anti-EGFR mAb showed disease progression, some of the acquired resistance should disappear before the anti-EGFR mAb rechallenge. It is not clear what mechanism contributed to the resumed sensitivity for this positive selection.

Ad-hoc analysis of the PURSUIT study, another prospective study of anti-EGFR mAb rechallenge, showed a significantly higher ORR in patients with a longer aEFI than in those with a shorter aEFI (≥ 365 vs. < 365 days; 44.4% vs. 7.3%; P = 0.0037) (Kagawa et al. 2022). It is speculated that initially, the major clone with RAS WT re-expands during the anti-EGFR mAb-free period. Thus, sufficient aEFI between the end of the last anti-EGFR mAb administration and the first day of anti-EGFR mAb rechallenge is considered a requirement for resuming sensitivity. However, our study showed neither a correlation between PFS and aEFI (r = -0.0078, P = 0.96) nor significant differences in PFS between the two groups based on a cut-off of 12 months (≥ 12 months [n = 24] vs. < 12 months [n = 13]; P_Log−rank = 0.76; HR, 0.88; 95% CI, 0.38–2.03). Similarly, in the ad-hoc analysis of the E-rechallenge trial, no significant differences were observed in the ORR, PFS, or OS according to aEFI (Osawa et al. 2018). Furthermore, the CHRONOS study, the first trial to prospectively evaluate the efficacy of anti-EGFR mAb rechallenge based on ctDNA mutational status, showed no correlation between the aEFI and probability of clinical response. As the results for the relationship between rechallenge with anti-EGFR mAb and the aEFI are inconsistent, further studies are warranted.

Conversely, as negative selection, the exclusion of patients with predictive factors for poor response to anti-EGFR mAb, such as RAS/BRAF MT, may increase the efficacy of chemotherapy containing anti-EGFR mAb (Cremolini et al. 2017; Manca et al. 2021). Considering that responses to anti-EGFR mAb rechallenge were not observed in the non-responders to the prior chemotherapy containing anti-EGFR mAb and the lack of a clear relationship between aEFI and the efficacy of anti-EGFR mAb rechallenge in this study, some mechanisms of primary resistance to the prior anti-EGFR mAb therapy may have persisted in the non-responders before anti-EGFR mAb rechallenge. Regarding the possible mechanisms of primary resistance to anti-EGFR mAb, constitutive activation of tyrosine kinase receptors other than EGFR through uncommon genomic alterations, such as MAPK pathway mutations, HER2 mutations and amplification, MET amplification, and rearrangements of NTRK, ROS, ALK, and RET, negatively affects susceptibility to EGFR inhibition (Cremolini et al. 2017, Shitara et al. 2024). Therefore, detecting these resistant mechanisms for negative selection for the anti-EGFR mAb rechallenge is important.

In this study, the proportion of patients with ctDNA RAS WT who achieved PR was 15.6%, compared to 0% in those with ctDNA RAS MT, and patients with ctDNA RAS WT had a significantly better prognosis than those with ctDNA RAS MT. According to the results of the CRICKET trial, patients with ctDNA RAS WT had a significantly longer PFS (median PFS 4.0 vs. 1.9 months; HR, 0.44; 95% CI, 0.18–0.98; P = 0.03) and tended to have longer OS than those with ctDNA RAS MT (Cremolini et al. 2019). In Japanese clinical trials of anti-EGFR mAb rechallenge, a post-hoc biomarker analysis (JACCRO CC-08/09AR) showed that patients with ctDNA RAS MT had significantly shorter PFS and OS than those with ctDNA RAS WT prior to the anti-EGFR mAb rechallenge (mPFS: 2.3 vs. 4.7 months; HR, 6.2; P = 0.013; mOS: 3.8 vs. 16.0 months; HR, 12.4; P = 0.0028) (Sunakawa et al. 2020). Similarly, in E-rechallenge, another Japanese clinical trial on anti-EGFR mAb rechallenge, the post-hoc analysis showed that the RR of patients with ctDNA all WT (50%) was higher than that of patients with any MT (KRAS G12/G13/A59/Q61, BRAF V600E, and EGFR S492R) (Osawa et al. 2018) (Supplemental Table 5). These results suggest that anti-EGFR mAb rechallenge is less effective in patients with any gene alterations related to the EGFR pathway, including RAS detected in ctDNA prior to anti-EGFR mAb, as these gene alterations may be associated with acquired resistance to anti-EGFR mAb. ctDNA test can be used to monitor real-time mutational status compared with tissue biopsy, facilitating the selection of appropriate chemotherapy, including anti-EGFR mAb rechallenge. In addition, this ctDNA test might be beneficial for patients with lesions in areas that are difficult to biopsy.

Regarding treatment lines of anti-EGFR mAb rechallenge in clinical practice, several studies have been reported. The VELO trial is a randomized phase II trial that compared anti-EGFR mAb rechallenge with a standard of care (trifluridine/tipiracil: FTD/TPI), which showed a significant improvement in PFS (Napolitano S et al. 2023). The study did not reveal significant efficacy in OS because of crossover, but a subgroup analysis was performed for patients treated with standard of care who received active anticancer treatment in the fourth-line after progression from FTD/TPI, and the median OS from the start of fourth line therapy was significantly longer in patients treated with anti-EGFR mAb rechallenge than the OS in patients with other therapies (Napolitano S et al. 2023). Therefore, anti-EGFR mAb rechallenge is considered a promising later-line treatment for patients with RAS/BRAF WT mCRC.

Compared with FTD/TPI plus bevacizumab, which is considered the standard third-line treatment for patients with mCRC (Prager et al. 2023), the ORR of anti-EGFR mAb rechallenge tends to be high, especially for patients with mCRC with ctDNA RAS WT. The ongoing PULSE trial is a randomized, phase II, open-label trial comparing the OS of panitumumab rechallenge with standard of care (FTD/TPI or regorafenib) in patients with mCRC without any alterations confirmed with liquid biopsies. The CITRIC trial is a multicenter, randomized, open-labeled, parallel-group, phase II study that evaluated the efficacy and safety of cetuximab plus irinotecan rechallenge versus investigators’ choice in the third-line setting. In addition, the FIRE-4 study is a randomized phase III study that tests the efficacy of early switch maintenance during first line therapy and also investigates rechallenge with cetuximab in later-line treatment using biopsy tissues (Holch et al. 2016). On the contrary, the PARERE study is a randomized phase II study of panitumumab rechallenge, followed by regorafenib versus the reverse sequence in patients with RAS/BRAF WT refractory mCRC (Moretto et al. 2021), and the CAVE-2 trial is a non-profit phase II, randomized study of the combination of avelumab plus cetuximab as a rechallenge strategy, compared with cetuximab alone in pre-treated patients with RAS/BRAF WT mCRC (Napolitano et al. 2022); the RAS status was examined with ctDNA in the two studies. These ongoing head-to-head clinical trials will optimize treatment strategies in the salvage-line setting for patients with tissue RAS/BRAF WT mCRC.

This study has several limitations. First, it was a retrospective study with a small sample size, which might affect the generalizability of the findings; the retrospective nature may also introduce potential selection and information biases. Besides, the potential risk of bias needs to be considered because of the very small number of patients in our subgroup analysis. Some results of our analysis had a P-value close to the statistical significance threshold, and they need to be interpreted cautiously. The relatively high frequency of false-negative rates in the BEAMing analysis for patients with mCRC needs to be considered (Bando et al. 2019). Previous studies have highlighted the cutoff for patients with peritoneal metastases alone with a lesion diameter < 20 mm, lung metastases alone with a lesion diameter < 20 mm, or < 10 lesions in total (Vidal et al. 2017; Bando et al. 2019; Kagawa et al. 2021). Therefore, an accurate interpretation of the results of this BEAMing analysis is crucial.

In conclusion, the incidence of ctDNA RAS MT mCRC prior to salvage-line treatment was 40.5%, and liver metastases and high tumor volumes (non-resected primary tumor and high tumor markers) were associated with the appearance of ctDNA RAS MT mCRC. Rechallenge with anti-EGFR mAb may be effective for patients without RAS MT detected in ctDNA and those who respond to first-line anti-EGFR mAb, while no patients with RAS MT in ctDNA responded to anti-EGFR mAb rechallenge. This anti-EGFR mAb rechallenge strategy may be a feasible option for patients with mCRC as a late-line treatment, and we need to validate this result prospectively in the future.