Optimising the use of cetuximab in the continuum of care for patients with metastatic colorectal cancer

Abstract

The anti-epidermal growth factor receptor (EGFR) monoclonal antibody cetuximab in combination with chemotherapy is a standard of care in the first-line treatment of RAS wild-type (wt) metastatic colorectal cancer (mCRC) and has demonstrated efficacy in later lines. Progressive disease (PD) occurs when tumours develop resistance to a therapy, although controversy remains about whether PD on a combination of chemotherapy and targeted agents implies resistance to both components. Here, we propose that some patients may gain additional clinical benefit from the reuse of cetuximab after having PD on regimens including cetuximab in an earlier treatment line. We conducted a non-systematic literature search in PubMed and reviewed published and ongoing clinical trials, focusing on later-line cetuximab reuse in patients with mCRC. Evidence from multiple studies suggests that cetuximab can be an efficacious and tolerable treatment when continued or when fit patients with mCRC are retreated with it after a break from anti-EGFR therapy. Furthermore, on the basis of available preclinical and clinical evidence, we propose that longitudinal monitoring of RAS status may identify patients suitable for such a strategy. Patients who experience progression on cetuximab plus chemotherapy but have maintained RAS wt tumour status may benefit from continuation of cetuximab with a chemotherapy backbone switch because they have probably developed resistance to the chemotherapeutic agents rather than the biologic component of the regimen. Conversely, patients whose disease progresses on cetuximab-based therapy due to drug-selected clonal expansion of RAS-mutant tumour cells may regain sensitivity to cetuximab following a defined break from anti-EGFR therapy. Looking to the future, we propose that RAS status determination at disease progression by liquid, needle or excisional biopsy may identify patients eligible for cetuximab continuation and rechallenge. With this approach, treatment benefit can be extended, adding to established continuum-of-care strategies in patients with mCRC.

Introduction

Colorectal cancer is the fourth most prevalent cancer type in the world and causes nearly 700 000 deaths per year worldwide.1 Patients diagnosed with metastatic colorectal cancer (mCRC) have a limited number of systemic therapeutic options available as well as local therapies, such as resection, ablation or transhepatic irradiation via the injection of yttrium 90. Encouragingly, the median overall survival (OS) in recent phase III first-line trials in RAS wild-type (wt) mCRC now exceeds 30 months.2–5

The available biologic agents usually planned for first-line therapy include the epidermal growth factor receptor (EGFR) monoclonal antibodies (mAbs) cetuximab or panitumumab and the anti–vascular endothelial growth factor (VEGF) agent bevacizumab. All three agents, plus additional targeted drugs such as aflibercept (VEGF inhibitor), regorafenib (receptor tyrosine kinase inhibitor) and ramucirumab (anti-VEGF receptor 2), have also shown efficacy in the second and later lines of therapy.6–12 Mounting evidence from randomised, phase III trials and meta-analyses suggests that many patients with RAS wt tumours may experience improved survival outcomes when chemotherapy is combined with cetuximab compared with combination with bevacizumab in the first-line setting.3 5 13–15 Retrospective analyses and meta-analyses have shown that the survival benefit is especially pronounced in patients with RAS wt, left-sided primary tumours.16–19 Although patients with right-sided tumours appear to derive less significant benefits from all available therapies for mCRC in terms of overall response rate (ORR), progression-free survival (PFS) and overall survival (OS) than do those with tumours originating in the left side of the colon,18–20 some patients may benefit from cetuximab-based therapy if the goal of treatment is response and subsequent cytoreduction.18 19 Indeed, cetuximab has been shown to yield a very good ORR, early tumour shrinkage (ETS) and depth of response in patients with RAS wt tumours. Such alternative metrics of response have been shown to be associated with improvements in survival5 compared with bevacizumab in the randomised, phase III, first-line FIRE-3 trial (5-FU, Folinic Acid and Irinotecan (FOLFIRI) Plus Cetuximab Versus FOLFIRI Plus Bevacizumab in First Line Treatment Colorectal Cancer), which evaluated patients with RAS wt mCRC.5 21

Currently, amplifications and mutations in several genes other than RAS, detected pretherapy, are also being investigated as potential predictive biomarkers of response to anti-EGFR therapy. The BRAF V600E variant22 23 appears to be a negative prognostic marker, but due to the relatively low number of colorectal cancers harbouring BRAF mutations (and as not all BRAF mutations confer resistance), controversy remains regarding the impact of this finding.24 Other markers of interest include alterations in the EGFR pathway, mutations in the phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit α gene (PIK3CA) and amplification of the human epidermal growth factor receptor 2 gene, HER2 (mostly in patients with anti-EGFR-refractory, RAS wt disease).25–33 Furthermore, hypermethylation of tumour DNA appears to be potentially predictive of poorer clinical outcomes in patients receiving anti-EGFR therapy.34 35 The individual frequencies of all of these mutations and amplifications are low,25 27 33 36 37 and conflicting data exist regarding whether several of these genomic alterations are true biomarkers for cetuximab resistance. Additional biomarkers that are currently being explored as predictive of cetuximab response include VEGF2, MET (mesenchymal-epithelial transition factor mutations), fibroblast growth factor receptor mutations, platelet-derived growth factor, epiregulin, amphiregulin, mIR 31–3 p and hepatocyte growth factor.29 30 38–46 Presently, however, RAS mutations and tumour sidedness remain the only robust factors when selecting patients for anti-EGFR therapy in accordance with the drug labels for both cetuximab and panitumumab and the established international guidelines10 47 and are the most informative for treatment decisions in mCRC.

Colorectal tumours that are RAS wt at baseline can evolve resistance to anti-EGFR therapy via a RAS status ‘shift’ to mutated status to escape the drug’s effects.39 It is now known that this change occurs when the RAS wt tumour cell population is diminished during anti-EGFR therapy, while pre-existing or newly evolving RAS-mutant subclones can proliferate and become detectable by DNA testing.27 29 48 Other gene mutations, including in BRAF, may be subject to the same tumour heterogeneity principles and drug selection dynamics. Another key mechanism of acquired resistance that occurs in approximately 25% of patients treated with anti-EGFR therapy are mutations in the extracellular domain (ECD) of EGFR that prevent further binding of anti-EGFR mAbs to EGFR. Importantly, EGFR ECD alterations emerge as a means for cancer cells to circumvent EGFR blockade—acquired resistance—and do not apparently exist as a mechanism of primary resistance in anti-EGFR-naïve patients. Interestingly, the frequency and type of EGFR ECD mutation varies, depending on previous treatment with cetuximab or panitumumab, and each different EGFR ECD mutation may potentially confer resistance to cetuximab only, panitumumab only or both mAbs.28 49–51 Finally, these mutations frequently coexist in the same patient at the moment of disease progression, as multiple RAS, BRAF and EGFR clones are detectable in cell-free circulating tumour DNA (ctDNA) by liquid biopsy.29 52

Extending the continuum of care through the delivery of as many lines of efficacious therapy as possible is desirable so that patients with progressive disease (PD) have multiple potentially beneficial treatment options to explore before transitioning into palliative care. For mCRC, however, patients with RAS wt tumours are generally assigned two lines of intensive therapy. They receive a first-line anti-EGFR mAb (cetuximab or panitumumab) or bevacizumab plus doublet/triplet chemotherapy (5-fluorouracil, leucovorin and oxaliplatin (FOLFOX), 5-fluorouracil, leucovorin and irinotecan (FOLFIRI), 5-fluorouracil, leucovorin, oxaliplatin and irinotecan (FOLFOXIRI)). This is followed by second-line ‘switch’ of either the biologic, the chemotherapy backbone or both on PD. For patients who respond to a therapy and whose disease does not progress but require an interruption due to their preference or recovery from drug toxicity or surgery, a pause-and-resume (commonly referred to as stop-and-go or reintroduction/re-exposition) schedule of a biologic with or without chemotherapy can help them to complete the full planned treatment and maintenance period.53–55

Multiple studies with dedicated second-line, third-line or mixed second-line and further-line patient populations have indicated improved survival and ORR with cetuximab monotherapy or cetuximab in combination with chemotherapy56–62 and with panitumumab monotherapy or panitumumab plus chemotherapy.57 63–67 Overall, all current guidelines conclude that patients with RAS wt mCRC who do not receive a biologic in the first line (but are fit enough to receive it in later lines) should be considered candidates for an anti-EGFR therapy in the next available line. Then, in the third line and beyond, patients are usually offered other targeted or chemotherapy-based interventions, often with limited expectations of clinical benefit. This is because, currently, it is widely accepted that after patients with left-sided, RAS wt mCRC have received an anti-EGFR antibody and bevacizumab sequentially (in either order) and experience progression, their disease is permanently resistant to both the biologic and cytotoxic agents that have been administered. However, this conclusion may not be accurate for all patients as some tumours may retain sensitivity to the targeted agent (although becoming resistant to the cytotoxic agents in the chemotherapy programme) or regain sensitivity after a ‘treatment break’ from those specific agents.68–73 Thus, many patients exhaust standard treatment options while maintaining a good performance status and are therefore not ready to transition to a solely palliative-care approach. Here, we review the evidence for continuation or reuse strategies beyond the first line with cetuximab, a well-established agent for RAS wt mCRC, as a method of expanding the therapeutic options available to patients and maximising the number of (potentially curative) therapeutic lines before the initiation of palliative care (figure 1).

Figure 1.

A proposed treatment model for the decision-making process when choosing between cetuximab continuation vs rechallenge. ^aTypically patients with left-sided, RAS wt mCRC or those with right-sided, RAS wt mCRC in need of rapid tumour shrinkage. ^bBy liquid biopsy. ^cOther evidence-based biomarkers can also be included in the panel of tests when feasible. EGFR, epidermal growth factor receptor; FOLFIRI, 5-fluorouracil, leucovorin and irinotecan; FOLFOX, 5-fluorouracil, leucovorin and oxaliplatin; mut, mutant; PD, progressive disease; wt, wild-type.

Notably, the bulk of available evidence for anti-EGFR retreatment stems from trials that use cetuximab rather than panitumumab. Furthermore, the two mAbs are known to behave somewhat differently in the context of treatment sequencing.74–81

Selecting cetuximab-eligible patients by liquid biopsy testing

Because the predominance of specific tumour cell subclones is dictated by selection pressures such as targeted therapy, retreatment strategies with cetuximab could be individualised with longitudinal tracking of detectable RAS mutations (and any future confirmed predictive biomarker of response) as a measure of potential sensitivity to cetuximab.27 29 30 39 51 Depending on the location of metastatic disease in some patients, traditional tumour needle or excisional biopsies may be invasive and risky. Thus, a less invasive method is desirable (eg, liver needle aspirates or plasma sampling). This ‘liquid’ biopsy can fulfil the retesting requirement by providing a non-invasive method for the detection and analysis of ctDNA. Of note, >75% of patients with advanced CRC have been shown to have detectable ctDNA (79.8% in the RASANC study (RAS Mutation Testing in the Circulating Blood of Patients With Metastatic Colorectal Cancer)),82 83 and multiple useful platforms currently exist to allow the identification of tumour mutations via PCR of ctDNA: digital PCR84; digital next-generation sequencing85; BEAMing (beads, emulsion, amplification and magnetics)86; pyrophosphorolysis-activated polymerisation87; personalised profiling by deep sequencing88 and tagged-amplicon deep sequencing.89 Liquid biopsies have proven to be clinically robust. In a cohort of 98 patients with mCRC, Schmiegel et al90 used BEAMing to test for RAS tumour mutation status and found 91.8% concordance between ctDNA and tissue-based testing. Similarly, Vidal et al91 detected a 93% concordance rate in a cohort of 115 patients with mCRC. In the prospective, multicentre RASANC study (n=425), the concordance rate was 83.7% in the overall population and was even higher in patients with their primary tumour in place, liver metastasis and synchronous disease.83

Overall, RAS is the only widely recognised predictive molecular biomarker of cetuximab response and the only biomarker with extensive evidence of a high concordance between tissue-based and blood-based testing. However, we will refer to patients under consideration in this article more generally as ‘cetuximab-eligible patients’, for whom this definition is subject to change as additional research is published in the coming years. Furthermore, no subgroup analyses by tumour location are available for the studies on treatment beyond progression discussed here. However, we use the term ‘cetuximab-eligible’ to encompass optimal candidates for cetuximab-based therapy, including those with left-sided, RAS wt mCRC, and those with right-sided, RAS wt mCRC for whom cytoreduction is a key treatment goal, or who have previously responded to cetuximab-based therapy.

Continuation treatment strategy

Patients who received and progressed on cetuximab-containing regimens in the preceding line may retain cetuximab sensitivity,68–70 92 having instead become refractory to the chemotherapy backbone (table 1). In a study conducted by Feng et al,69 patients with RAS wt tumours whose disease progressed during first-line cetuximab plus chemotherapy were randomised to receive a different chemotherapy backbone with or without cetuximab as second-line treatment. The cetuximab-continuation group demonstrated better PFS, OS and disease control rates and a potentially better ORR than did the chemotherapy-only (no cetuximab continuation) group. Extended RAS analysis for the retrospective study revealed that baseline RAS wt status correlated with response to continuation of cetuximab; ETS during first-line cetuximab-based treatment also correlated with improved efficacy outcomes during second-line cetuximab continuation. Additionally, Ciardiello et al68 found evidence of a survival benefit from continued cetuximab plus FOLFOX in patients whose disease progressed during first-line cetuximab plus FOLFIRI and had KRAS/NRAS/BRAF/PIK3CA wt tumours. Both Feng et al and Ciardiello et al switched chemotherapy backbones when going from first-line to second-line treatment and used a chemotherapy switch alone as a comparator arm. Therefore, if sensitivity to cetuximab was preserved, switching to a non-cross-resistant chemotherapy backbone after PD could reinvigorate responsiveness. No new safety findings were identified from cetuximab continuation in any of these studies.68 69 72

Table 1.

Evidence for cetuximab continuation plus chemotherapy backbone switch in patients with KRAS wt (or retrospectively evaluated RAS wt) mCRC whose disease has progressed on cetuximab plus chemotherapy treatment

Line of treatment	Study	Previous regimen	Arm A			Arm B
			PFS, mo	OS, mo	ORR, %	PFS, mo	OS, mo	ORR, %
Dedicated second line	Feng et al69 (2016; retrospective study)	Cetuximab+mFOLFOX6 or FOLFIRI	Cetuximab+CT switch (n=102)			CT switch (n=96)
			6.3*	17.3*	18.6	4.5	14.0	9.4
	Ciardiello et al68 (2015; CAPRI-GOIM†)	Cetuximab+FOLFIRI	Cetuximab+FOLFOX (n=34)			CT switch (n=32)
			6.9*	23.7	29.4	5.3	19.8	9.4
	Vladimirova et al72 (2016)	Cetuximab+FOLFOX	Cetuximab+FOLFIRI (n=20)
			6.5‡	–	20.0
Mixed third and further lines	Fora et al70 (2013)	Standard-dose cetuximab+irinotecan	High-dose cetuximab+irinotecan
			2.8‡	6.6‡	(DCR=45)

All studies included a KRAS wt population, except where indicated.

*P<0.05.

†Retrospective extended RAS/BRAF/PIK3CA wt.

‡OS2 and PFS2 shown instead of all OS and PFS.

CT, chemotherapy; DCR, disease control rate; FOLFIRI, 5-fluorouracil, leucovorin and irinotecan; FOLFOX, 5-fluorouracil, leucovorin and oxaliplatin; mCRC, metastatic colorectal cancer; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; wt, wild-type.

By contrast, the majority of panitumumab retreatment trials have followed the stop-and-go (re-exposition) model rather than continuation (ie, patients did not experience progression on panitumumab immediately before retreatment).93 94 For example, a phase II Japanese study examining the administration of panitumumab plus chemotherapy following six cycles of first-line panitumumab plus FOLFOX is currently ongoing (SAPPHIRE (Safety and Efficacy Study of mFOLFOX6 + Panitumumab Combination Therapy and 5-FU/LV + Panitumumab Combination Therapy in Patients With Chemotherapy-naïve Unresectable Advanced Recurrent Colorectal Carcinoma); NCT02337946); therapy prolongation appears to be planned in the absence of PD, and the focus of the study is the feasibility of oxaliplatin treatment extension rather than panitumumab continuation.95

As has been demonstrated, deriving maximum benefit from cetuximab before introducing a biologic switch could extend the number of potentially efficacious lines of therapy for fit patients and thereby optimise the continuum of care.69 We advance the hypothesis that using ETS during first-line cetuximab and extended RAS analysis at PD can enable effective patient selection for this therapeutic strategy (figure 1).69 For patients whose tumours transition from cetuximab sensitive to insensitive (eg, by converting to RAS mutant), a treatment break followed by cetuximab rechallenge may allow another means of extending the continuum of care. Here, again, liquid biopsy for RAS status may prove highly informative.

Rechallenge treatment strategy

Evidence of rechallenge therapy in multiple tumour types has been published.96–98 Because no conclusive randomised trials have been completed to test whether longitudinal RAS status monitoring can identify patients who regain cetuximab sensitivity, we will outline the available preliminary preclinical (biological) and clinical support for this treatment approach.

Biological evidence

Tumour cells in a patient who is receiving treatment are constantly under selection pressures from the therapy being administered. Studies have suggested that mutations conferring resistance to therapy can arise during treatment. For example, when retested for (K)RAS status, >50% of patients with acquired resistance to first-line anti-EGFR therapy demonstrate a ‘switched’ status from (K)RAS wt to mutant,27 29–31 39 51 82 and the opposite observation (tumours switching from mutant KRAS to undetectable mutant status) has been made during bevacizumab treatment.99 Importantly, from studies monitoring plasma levels of RAS-mutant ctDNA, even short ‘holidays’ off anti-EGFR therapy may restore tumours to a cetuximab-sensitive state.39 68 100 Intratumour heterogeneity and drug-selected clonal evolution can account for these ‘switches’ (figure 2). The suggested mechanism for such a switch is the constant presence of a small number of RAS-mutant cell subclones persisting in the tumours, rather than de novo RAS mutations appearing in previously RAS wt tumour cells.73 Thus, a tumour can contain predominantly RAS wt clones, test as RAS wt and respond to cetuximab treatment until the RAS wt clones are depleted. In this environment, the RAS-mutant clones have the opportunity to continue proliferating and surviving and thus come to represent the dominant tumour subclonal population.51 100 Indeed, an in vitro study of two separate KRAS wt colorectal cancer cell populations showed that, under selection pressure from cetuximab treatment, the surviving population developed KRAS amplifications; when allowed to proliferate for 160 days without additional cetuximab treatment, the population of cells achieved significantly lower levels of KRAS amplification.39

Figure 2.

A model for the biological rationale for rechallenge therapy: clonal selection in heterogeneous baseline RAS^a wt tumours during anti-EGFR therapy. ^aAdditional or secondary acquired mechanisms of resistance can also be driven by mutations in the extracellular domain of the EGFR, and other potential biomarkers continue to be investigated. EGFR, epidermal growth factor receptor; mut, mutant; wt, wild-type.

The principles of clonal selection likely also apply to other mechanisms of resistance to anti-EGFR therapy.29 Indeed, a number of genetic modifications beyond RAS have been related to acquired drug resistance to anti-EGFR therapy. Interestingly, most of these different genetic aberrations are related to or have influence on MEK-ERK pathway activity.101 The reversion of these additional resistance mechanisms during the anti-EGFR therapy break is controversial and not yet clearly demonstrated. Finally, it should be noted that even if arising on-treatment mutations have been identified, the exact threshold associated with resistance to anti-EGFR therapy has not been established, and some patients continue to experience disease control for several months after RAS-mutant clones emerge.51 102 Overall, there is a strong biological rationale for cetuximab rechallenge.

Clinical evidence

There is currently abundant clinical evidence for RAS status switch103 104 but little evidence analysing the direct relationship between a return to undetectable RAS-mutant status and a resensitisation to cetuximab.39 However, there exists clinical evidence for the feasibility of cetuximab rechallenge in patients with baseline (K)RAS wt mCRC whose disease progresses on first-line anti-EGFR therapy plus chemotherapy; such patients are usually rechallenged with cetuximab with or without chemotherapy in the later-line setting after ≥3 months of therapy that does not contain an anti-EGFR targeted agent. In a study by Santini et al73 (table 2), 39 patients whose disease had previously progressed on cetuximab plus irinotecan-based therapy were rechallenged with cetuximab plus irinotecan after a treatment break during which they received non-irinotecan-based chemotherapy alone. The median treatment break for these patients was 6 months, and patients had received a median of four lines of therapy prior to study enrolment. Nevertheless, median PFS was 6.6 months, and the ORR with cetuximab plus irinotecan rechallenge was 53.8% and included two complete responses. Stable disease was achieved by a further 35.9%, for a total disease control rate of 89.7%. Finally, in a recent prospective study by Gruppo Oncologico del Nord Ovest in Italy, the CRICKET study (Cetuximab Rechallenge in Irinotecan-pretreated mCRC, KRAS, NRAS and BRAF Wild-type Treated in 1st Line With Anti-EGFR Therapy), 27 patients with mCRC were to be treated with cetuximab plus FOLFIRI or FOLFOXIRI in the first line followed by bevacizumab plus oxaliplatin-based chemotherapy in the second line and cetuximab plus irinotecan rechallenge in the third line. The third-line ORR was 23%, with 54% of patients experiencing disease control, and the study met its primary end point.105 Rechallenge treatment was well tolerated in all studies. The prospective biological determinations of ctDNA are ongoing.

Table 2.

Evidence for cetuximab rechallenge in patients with KRAS wt mCRC whose disease has previously progressed on cetuximab plus chemotherapy treatment and received who have received at least one line of additional, non–anti-EGFR therapy

Study	Previous regimen
Liu et al71 (2015; retrospective study)	Summary: Patients with KRAS wt mCRC received anti-EGFR therapy (cetuximab or panitumumab) based treatment When PD occurred, patients received a break from anti-EGFR therapy (median duration, 4.6 months) 80 patients were retreated with cetuximab ± CT ± other targeted agents Results: 58.0% (CR/PR/SD) Patients who responded to prior anti-EGFR therapy were more likely to obtain clinical benefit upon rechallenge (PFS, 4.9 vs 2.5 months; P=0.007) The clinical benefit rate on rechallenge showed a marginally significant association with interval time between the two anti-EGFR based therapies (P=0.053)
Santini et al73 (2012)	Summary: Patients with KRAS wt mCRC received anti-EGFR therapy (cetuximab) + (FOLF)IRI When PD occurred, patients received a break from anti-EGFR therapy (median duration, 6.0 months) 39 patients were retreated with cetuximab + (FOLF)IRI Results: ORR, 53.8% (plus 35.9% SD) Median PFS, 6.6 months Interval effect not discussed
Rossini et al105 (2017; CRICKET)	Summary: Patients with RAS/BRAF wt mCRC who became resistant to first-line cetuximab + irinotecan Treated with third-line cetuximab + irinotecan N=26 Results: ORR, 23% (plus 31% SD)

CR, complete response; CT, chemotherapy; EGFR, epidermal growth factor receptor; FOLFIRI, 5-fluorouracil, leucovorin, and irinotecan; IRI, irinotecan; mCRC, metastatic colorectal cancer; ORR, overall response rate; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease; wt, wild-type.

Also noteworthy is a retrospective study by Liu et al71analysing patients with baseline KRAS wt status who received cetuximab (n=76) or cetuximab plus erlotinib (n=13) after a median of 4.57 months of treatment break from anti-EGFR therapy. Patients who had previously responded to anti-EGFR therapy achieved a median PFS of nearly 5 months.71 Rechallenge treatment was found to be well tolerated. Additionally, Liu et al71 suggested that the length of the anti-EGFR treatment interval may be related to the responsiveness of patients to rechallenge, on the basis of the observation that patients whose break from anti-EGFR therapy was longer than the median appeared more likely to respond. Overall, these studies suggest that good ORR and PFS can be achieved with cetuximab rechallenge in a patient population that has experienced progression on many prior lines of therapy, and these observations underscore the importance of optimising and extending the continuum of care.

One limitation of both the retrospective trial by Liu et al and prospective trial by Santini et al (except the CRICKET study) is that enrolment occurred prior to the standardisation of extended RAS analysis; therefore, data are not yet available for patients with RAS wt tumours. Furthermore, no retest was conducted for RAS status as liquid biopsy had not yet been validated. However, the fact that a patient population with previous progression on cetuximab-based therapy can achieve a median PFS of 5–6 months with rechallenge suggests that sensitivity to cetuximab, at least in some patients, can be meaningfully regained. However, certain selection criteria could be recommended here, such as establishing a minimum recommended treatment break length or selecting patients who had a durable response or disease stabilisation during the last cetuximab-based treatment.

Additional evidence for cetuximab rechallenge therapy can be found in case studies.103 106 107 Of note, Siravegna et al39 identified a patient with mCRC who initially achieved stable disease for 6 months with cetuximab plus irinotecan; on disease progression, the patient received XELOX (capecitabine and oxaliplatin) and experienced progression again after 3 months. The patient was then rechallenged with cetuximab plus irinotecan and achieved a partial response.

Notably, anti-EGFR rechallenge strategy data are available primarily for cetuximab but not for panitumumab. The limited data on panitumumab rechallenge come from two case reports by Siravegna et al,39 in which the patients achieved a partial response and stable disease when retreated with panitumumab following RAS mutational status switch and reversal. Additionally, Hata et al108 describe two patients with good outcomes when retreated with panitumumab after a line of non-anti-EGFR therapy, although neither patient had experienced PD during prior panitumumab-based therapy (in the earlier line, one stopped panitumumab treatment due to toxicity, and the other had completed the determined number of panitumumab cycles). Finally, some reports have described responses to panitumumab treatment after failure of cetuximab, primarily in heavily pretreated patients.109 110

Ongoing studies, including additional analyses of the CRICKET study (NCT02296203), a study in Japan (UMIN000016439), and a similar study in Israel (trial identification number not yet available), will help determine whether this treatment strategy is feasible and which patients are most suitable for this approach. Although no completed trial to date has tracked RAS mutational status longitudinally throughout treatment, progression and rechallenge, a highly relevant phase III trial (FIRE-4; NCT02934529) is currently being conducted. This study has a planned enrolment of 550 patients with RAS wt mCRC who will receive first-line cetuximab-based therapy and third-line cetuximab rechallenge. This trial specifically includes non-invasive (liquid biopsy) RAS status assessment on PD and will thus directly test our proposal that longitudinal RAS status monitoring can be a tool in selecting patients suitable for cetuximab rechallenge after a treatment break. As of this writing, the FIRE-4 study is actively recruiting patients and has an estimated completion date of March 2022.

Although the studies we have reviewed have reported good clinical efficacy with cetuximab continuation and rechallenge, and evidence of a possible connection to RAS status switch, the treatment strategies outlined in figure 1 are not considered routine practice at this time. As new evidence emerges and patient selection becomes more refined, these strategies are likely to become valuable for oncologists looking to extend the continuum of potentially curative therapy for patients with a good performance status. Finally, it is also possible that other mechanisms of acquired resistance to cetuximab (eg, EGFR,27 51 HER2/MET amplifications,104 KRAS/NRAS, BRAF, PIK3CA111 112) arise in the same way and coexist in the same tumour. Some may be subject to similar dynamics of clonal selection under pressure, and, therefore, return to predominantly wt status after the anti-EGFR treatment is halted (except for EGFR ECD mutations, which do not pre-exist in untreated tumours but emerge de novo under anti-EGFR treatment and persist in the tumour).51

Conclusion

Anti-EGFR mAb plus chemotherapy is a standard of care in the first-line setting for patients with RAS wt mCRC, with a particular benefit for patients with left-sided tumours.10 11 Importantly, cetuximab can also be a significant component in second-line and later-line treatments, thereby affording additional treatment opportunities to suitable patients and optimising and extending the continuum of care. In cetuximab-naïve patients, later-line treatment with cetuximab plus chemotherapy generally results in efficacy benefits over chemotherapy alone.56 59 Use of cetuximab monotherapy or cetuximab plus chemotherapy in second-line and later-line mCRC does not yield any new safety signals, thereby further supporting its utility.56 57 59 113 114

Notably, patients do not have to be cetuximab naïve to extract benefits from second-line and later-line use of cetuximab.68–70 72 Treatment with cetuximab beyond progression in conjunction with a different chemotherapy backbone results in efficacy benefits, although this approach still needs to be confirmed in a large-scale clinical study. Evaluation of tumour RAS status on progression on first-line cetuximab plus chemotherapy may be informative in identifying candidates for continuation of cetuximab after the RAS threshold relevant for clinical resistance has been determined. Additionally, patients whose tumours ‘switched’ to RAS-mutant status following previous cetuximab-based therapy can regain cetuximab sensitivity and be rechallenged with cetuximab after a ‘holiday’ of several months, with renewed efficacy benefits.71 73 Liquid biopsy-based RAS testing will be needed to select for patients who can receive continuation cetuximab in the second line versus a treatment break (from anti-EGFR therapy) followed by rechallenge in third or further lines. We anticipate the results of trials such as FIRE-4, which may solidify the concept of cetuximab rechallenge as a routine therapeutic strategy to optimise and extend the continuum of care in patients with mCRC. To this end, further research into additional biomarkers of response to anti-EGFR agents will ensure definition of the optimal patient populations for these strategies, thereby maximising the continuum of care.