Looking for Zebras in Metastatic Colorectal Cancer

John L. Marshall, MD

When you hear hoofbeats, think of horses not zebras.” The idea behind this medical adage is that horses are more common than zebras, so if you hear hoofbeats, they are much more likely to come from a horse than a zebra. However, colorectal cancer (CRC) requires more careful analysis. Although it’s one of the most commonly diagnosed cancers in the United States,¹ and researchers’ understanding of the biology and pathophysiology of CRC has grown, its many variations require that each case be diagnosed with care.

Importance of Next-Generation Sequencing

During a recent OncLive Peer Exchange^®, a panel of CRC experts discussed the heterogeneity of metastatic CRC (mCRC) and the need to identify patient subsets within this disease to enable a precision medicine approach. They also shared their insights on some immunotherapies that are in development.Next-generation sequencing (NGS) has revolutionized cancer care, enabling researchers to better understand the complicated molecular mechanisms of oncogenesis and identify treatment targets to improve patient care and outcomes. Yet not all patients with mCRC are receiving NGS.

“I see lots of patients with colon cancer with zero molecular testing when we start out,” John L. Marshall, MD, said. In recent years, researchers have identified several mutations in patients with mCRC that can help with prognostication and guide treatment decision making, which makes these tests essential to optimizing care.

“I think, off the bat, you should have patients’ RAS status—KRAS and NRAS. You should probably know their RAF status up front—BRAF. We should definitely know their MSI [microsatellite instability] status,” panelist Bert H. O’Neil, MD, said. The panelists suggested that large CRC centers are increasingly moving toward earlier testing with comprehensive panels that assess for these and other genomic alterations, rather than using a variety of tests throughout the treatment course to assess for defects in a single gene or small set of genes.

A comprehensive panel mentioned was the FoundationOne CDx, which evaluates all guideline-recommended genes in solid tumors and provides information on MSI status and tumor mutational burden.² A 2017 study coauthored by panelist Marwan G. Fakih, MD, puts in perspective how comprehensive genomic testing can uncover not only well-characterized alterations, which were mentioned by O’Neil, but also rarer alterations that might open the door to new treatment approaches for smaller subsets of patients with mCRC, some of whom might have few treatment options.³ For example, although ERBB2 (HER2)-amplified tumors, which do not respond to EGFR antagonists, were identified in only 5.1% of the 138 patients included in Fakih’s single-institution study, those patients could now be considered for HER2-directed treatments, the viability of which was demonstrated by the HERACLES trial results.^3,4 In HERACLES, 10 of 33 patients (30.3%) with HER2-amplified, treatment-refractory mCRC achieved an objective response with dual HER2 blockade (trastuzumab [Herceptin] plus lapatinib [Tykerb]), with 2 patients achieving complete responses (CRs).⁴ At the 2018 American Society of Clinical Oncology (ASCO) Annual Meeting, the HERACLES investigators recommended a cutoff of 3 copies of ERBB2 in plasma to select patients who are most likely to benefit from HER2 blockade.⁵

Of all known actionable mutations in mCRC, RAS is the most common. In Fakih’s study, it was present in 51.4% of the patients.³ In contrast, RAF mutations were significantly less common, observed in 7.2% of patients. The frequencies of both mutations were reported to be concordant with those historically reported in mCRC.³ During the Peer Exchange, the panel noted that RAS and RAF mutations are associated with resistance to anti-EGFR therapies; thus, these treatments should be avoided in these patients. However, because it can take 10 to 14 days to receive NGS results, the panel also discussed how to proceed with treatment before a patient’s RAS and RAF statuses are known.

Considering Right- Vs Left-Sided Tumors

“I think starting with FOLFOX [folinic acid, fluorouracil, oxaliplatin], FOLFIRI [folinic acid, fluorouracil, irinotecan], if you prefer, or for some, FOLFOXIRI, may not be an unreasonable approach while waiting for the next-generation sequencing data. And that’s what we do in our practice,” Fakih said. More intense regimens, such as FOLFOXIRI [folinic acid, fluorouracil, oxaliplatin, irinotecan] (FOLFIRINOX), were deemed most appropriate for patients with more extensive and problematic metastases, where achieving a quick response is crucial. “If the NGS test reveals RAS/RAF wild-type, then addition of an EGFR-targeted agent could be considered,” panelist Gabriela Chiorean, MD, suggested.Left-sided (hindgut; splenic flexure, descending colon, sigmoid, and rectum) and right-sided (midgut; cecum, ascending colon, hepatic flexure, and transverse colon) mCRCs were recently discovered to harbor different clinical and biological characteristics, affecting prognosis and response to certain treatments.⁶ “I think that clearly the rectals and the sigmoids have better prognosis overall. If they are KRAS wild-type, they benefit even more. I think that nowadays we also know that a tumor in the transverse colon should not be considered right-sided; it should be considered left-sided. So, we learn more and more things, and I think these things are extremely valid about the location of the cancer, both prognostically as well as predictively when we are to use EGFR-targeted therapies,” Chiorean said.

The panel also discussed new data regarding right- versus left-sided tumors, including findings from a reanalysis of the phase III CALGB/SWOG 80405 trial data.⁷ This trial assessed chemotherapy (FOLFIRI or FOLFOX) combined with the anti-VEGF monoclonal antibody bevacizumab (Avastin) or the anti-EGFR monoclonal antibody cetuximab (Erbitux) in patients with KRAS wildtype mCRC.⁸ Both antibodies showed equivalent overall survival (OS) in this population, but the reanalysis revealed some distinctions.

Considerations in BRAF-Mutated mCRC

“[With analysis of sidedness], we found that patients with right-sided tumors who were RAS wild-type appeared not to do very well with the EGFR antibody. Whereas on the left side, it appeared that patients did better with an EGFR antibody than they did if they started out with bevacizumab,” O’Neil said. Based on these findings and data from other trials, including FIRE-3, CRYSTAL, and PRIME, the panel largely agreed that FOLFOX or FOLFIRI should be paired with bevacizumab in the first-line setting for patients with right-sided tumors that are KRAS and BRAF wild-type. Anti-EGFR should be strongly considered for patients with left-sided KRAS and BRAF wild-type tumors. However, panelist Wells A. Messersmith, MD, FACP, said he feels comfortable proceeding with bevacizumab for both patient subsets. “I think in the 80405 trial, there’s a fairly strong trend in benefit that I’m not sure quite reached statistical significance, and the difference was somewhat marginal,” he said. “We save the EGFR, because of rash and other considerations, for second-line therapy,” he added, noting that he discusses toxicities with patients to determine what they are willing to tolerate.BRAF mutations have been found to be more prevalent in right-sided tumors. “About a third of right-sided tumors, or 25% to 30%, are going to be BRAF mutated,” Fakih said. Overall, however, only 6% to 7% of all mCRC will be BRAF-mutant, making this another zebra.³ Many BRAF-mutated tumors are also going to be RAS wild-type and MSI, the panel noted, making it tempting to give an antiEGFR. However, according to Fakih, “There’s no convincing data that anti-EGFR [alone] would work.” However, the panel discussed a few studies that are simultaneously targeting EGFR and BRAF that are showing promise in this small subset of patients with BRAF-mutant mCRC.

SWOG 1406 assessed irinotecan and cetuximab with or without vemurafenib (Zelboraf) in subsequent lines in patients with BRAF V600—mutated, extended RAS wild-type mCRC.⁹ All patients were anti—EGFR-naïve but previous treatment with irinotecan was allowed. Patients who received the triplet had a longer progression-free survival (PFS; 4.4 vs 2.0 months) and a higher disease control rate (67% vs 22%) than those receiving the doublet. “This is a unique patient set that has no good options. If there is no clinical trial for them in the second-line setting, it is fair to consider vemurafenib, cetuximab, and irinotecan,” Fakih said.

Sequencing and Dosing Considerations

BEACON CRC assessed the safety and efficacy of combining the BRAF inhibitor encorafenib (Braftovi) with the MEK inhibitor binimetinib (Mektovi) and the anti-EGFR antibody cetuximab in patients with BRAF V600E—mutant mCRC and who received 1 or 2 prior regimens. The study results showed this triplet to be well tolerated and to have encouraging clinical activity, with a confirmed overall response rate (ORR) of 41%.¹⁰ Based on BEACON CRC data, a few months after the Peer Exchange, the triplet combination received a breakthrough therapy designation from the FDA for the treatment of patients with BRAF V600E—mutant mCRC after failure of 1 or 2 prior lines of therapy.¹¹A standard of care for patients with refractory mCRC has been cetuximab followed by regorafenib (Stivarga), but it was unclear whether this was the best sequence.¹² The panel discussed REVERCE, a phase II trial that compared therapeutic sequencing with regorafenib followed by cetuximab with or without irinotecan (RC arm) with that of cetuximab with or without irinotecan followed by regorafenib (CR arm) after failure of f luoropyrimidine, oxaliplatin, and irinotecan in patients with KRAS exon 2 wild-type mCRC.¹²

“[The study] demonstrated a survival advantage for giving regorafenib first followed by cetuximab over the opposite sequence,” O’Neil said. “Again, it’s small [17.4 months for RC vs 11.6 months for CR]. I don’t weight to put on that, but they had a particularly long progression-free survival on cetuximab in patients who had been treated with regorafenib first,” he said. In a separate biomarker analysis, the OS benefit was maintained in the RC arm regardless of the biomarker subgroup assessed.¹³ It also showed that emerging oncogenic alterations were more common following cetuximab than regorafenib.

A key challenge with regorafenib, however, has been its toxicity profile, including palmar-plantar erythrodysesthesia syndrome, fatigue, and hypertension, which has limited its use. The phase II ReDOS study sought to compare a regorafenib dosing strategy that started at 80 mg daily and escalated by 40 mg weekly to reach 160 mg daily by week 3 versus starting at 160 mg daily.¹⁴

“The progression-free survival with the low dose was 2.5 months versus 2 months with the standard dose. But the overall survival was about 9 months with starting low versus just 6 months with a standard dose,” Chiorean said. “I think the study taught us that using judicious care for dosing our patients and not really pushing the maximum dose is very important to really allow them to tolerate treatment and ensure their quality of life,” she added. Patients in the dose-escalation arm had lower overall rates of grade 3 or 4 toxicity and improvements in multiple quality of life parameters, especially at week 2 of the first cycle.

Immunotherapy Combos in MSI and Microsatellite Stability mCRC

O’Neil said that the ReDOS study would change his practice, even if the outcomes are similar, because of the reduction in toxicity. “I think when you’re having a limited benefit, which is important for some patients, it’s important not to tax them with quality-of-life issues,” he said.The panel discussion of mCRC ended with a review of several immunotherapy drugs and combinations under investigation; results are promising for MSI tumors and largely underwhelming for microsatellite stability (MSS) tumors, they noted. “[For MSS], the most common subtype, I’m seeing a lot of combination studies coming, Messersmith said. “I’m seeing a lot of novel agents by specific antibodies, antibody—drug conjugates, etc. However, at least at ASCO [American Society of Clinical Oncology] 2018, we’re not seeing anything presented that’s sort of groundbreaking quite yet,” He said that a trial evaluating a STAT3 inhibitor ended up being negative.

Another negative trial was the phase III IMblaze370 study, which assessed combination therapy with the PD-L1 inhibitor atezolizumab (Tecentriq) and the MEK inhibitor cobimetinib (Cotellic) in heavily pretreated patients with locally advanced or mCRC, 95% of whom had MSS tumors.¹⁵ In the study, the combination did not show superior OS compared with the multikinase inhibitor regorafenib (8.9 vs 8.5 months, respectively).¹⁶ Marshall said it was particularly disappointing because a few responses had been seen during phase II, and he asked the panel whether combination immunotherapy approaches should still be pursued in patients with MSS tumors. O’Neil responded that he believes there is still promise in immunotherapy approaches for MSS tumors and that identification of the right combinations might enable these cold tumors to be turned hot. “We didn’t predict the PD-1 antibodies until just a few years ago. So, I certainly think we should continue to study that; we shouldn’t give up on that,” he said.

A success story for MSI-high (MSI-H) mCRC has been nivolumab (Opdivo) in combination with ipilimumab (Yervoy), which was assessed in the phase II CheckMate142 study.¹⁷ The combination showed an ORR of 49%, with 5 CRs and 53 partial responses (PRs) among the 119 study patients. The majority of patients (n = 82) had previous treatment with fluoropyrimidine, oxaliplatin, and irinotecan. In these patients, the ORR was 46%, with 3 CRs and 35 PRs. Based on CheckMate-142 data, the FDA granted an accelerated approval to the combination for the treatment of adult and pediatric patients 12 years and older with MSI-H or mismatch repair—deficient mCRC following progression on fluoropyrimidine, oxaliplatin, and irinotecan.¹⁸

Despite these findings, Chiorean said she prefers pembrolizumab (Keytruda) or nivolumab for her MSI-H patients. “I haven’t really gone to nivolumab/ipilimumab—to adding a CTLA-4 inhibitor. I still feel that the differences are small. Again, some may argue that maybe the durability of responses is better with CTLA-4 inhibition, but I also feel that the toxicities are significantly higher,” she said.

Messersmith said he uses nivolumab with ipilimumab when he needs to obtain a response quickly. “From my standpoint, given the extra 15% to 20% percentage points of response that you get when you add the CTLA-4, if a patient’s symptomatic or having other issues, then I might add it on, knowing I’m probably going to get more adverse effects,” he said. For him, one of the greatest challenges has been figuring out when to stop these agents.

Concluding Thoughts

Messersmith added that when the results for PD-1 and PD-L1 inhibitors were first reported in CRC, response rates were higher, so these treatments looked very promising. With each subsequent medical meeting, following those early results, the reported response rates have been steadily decreasing, indicating these treatments are not the miracle they first appeared. Nevertheless, Messersmith said he is still seeing some amazing responses with these treatments in his clinic—responses he’s never seen with other treatments—which means there is definitely a role for these treatments, but they are not the panacea they may have appeared to be. Better patient selection and new combinations may help improve response rates.A common theme in the panelists’ concluding remarks was the importance of recognizing CRC as a heterogenous disease with significant differences in disease biology, tumor location, response to treatment, and prognosis. “Continuing biomarker work is going to be extremely important…look for the zebras,” Fakih said.

Chiorean added, “We need to keep in mind left versus right, which is probably the most important thing we’ve learned in the last few years about colon cancer.”

The panel lamented the lack of major therapeutic developments in the field in recent years, but they hope that the ever-increasing understanding of the mechanisms that drive mCRC will eventually lead to major breakthroughs. “I’d like to see us keep moving forward with new drug development and really, hopefully, the companies will focus on colorectal cancer as much as possible,” O’Neil said.

Messersmith said that although he saw 2018 as “a year of incremental benefit,” he was “encouraged by the fact that the whole seems to be greater than the sum of the parts,” explaining that all the little discoveries do add up. “I’m hoping that circulating tumor DNA—cheap, noninvasive blood testing—will put us all out of business on colorectal cancer and we can focus on pancreatic cancer and other things,” he said.

References

1. American Cancer Society. Colorectal cancer facts & figures. American Cancer Society website. cancer.org/research/cancer-facts-statistics/colorectal-cancer-facts-figures.html. Accessed November 27, 2018.
2. Foundation Medicine. FoundationOne CDx. Foundation Medicine website. foundationmedicine.com/genomic-testing/foundation-one-cdx. Accessed November 27, 2018.
3. Gong J, Cho M, Sy M, Salgia R, Fakih M. Molecular profiling of metastatic colorectal tumors using next-generation sequencing: a single-institution experience. Oncotarget. 2017;8(26):42198-42213. doi: 10.18632/oncotarget.15030.
4. Siena S, Sartore-Bianchi A, Trusolino L, et al. Final results of the HERACLES trial in HER2-amplified colorectal cancer. Cancer Res. 2017;77(13):CT005. cancerres.aacrjournals.org/content/77/13_Supplement/CT005.
5. Bardelli A, Siravegna G, Sartore-Bianchi A, et al. Plsma HER2 (ERBB2) copy number to predict response to HER2-targeted therapy in metastatic colorectal cancer. J Clin Oncol. 2018;36(suppl 15):3506. ascopubs.org/doi/abs/10.1200/JCO.2018.36.15_suppl.3506.
6. Stintzing S, Heinemann V. Left- and right-sided CRC: what is the impact on clinical decisions? 2018 Gastrointestinal Cancers Symposium website. gicasym.org/daily-news/left-and-right-sided-crc-what-impact-clinical-decisions. Published January 12, 2017. Accessed November 26, 2018.
7. Das RK, Furchtgott L, Meyerhardt JA, et al. Causal modeling of CALGB 80405 (Alliance) to identify network drivers of metastatic colorectal cancer (CRC). J Clin Oncol. 2018;36(suppl 15):3570. ascopubs.org/doi/abs/10.1200/JCO.2018.36.15_suppl.3570.
8. Venook AP, Niedzwiecki D, Lenz HJ, et al. CALGB/SWOG 80405: Phase III trial of irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/leucovorin (mFOLFOX6) with bevacizumab (BV) or cetuximab (CET) for patients (pts) with KRAS wild-type (wt) untreated metastatic adenocarcinoma of the colon or rectum (MCRC). J Clin Oncol. 2014;32(suppl 15):LBA3. ascopubs.org/doi/abs/10.1200/jco.2014.32.15_suppl.lba3.
9. Kopetz S, McDonough SL, Morris VK, et al. Randomized trial of irinotecan and cetuximab with or without vemurafenib in BRAF-mutant metastatic colorectal cancer (SWOG 1406). J Clin Oncol. 2017;35(suppl 4):520. ascopubs.org/doi/abs/10.1200/JCO.2017.35.4_suppl.520.
10. Van Cutsem E, Cuyle P-J, Huijberts S, et al. BEACON CRC study safety lead-in (SLI) in patients with BRAF^V600E metastatic colorectal cancer (mCRC): efficacy and tumor markers. J Clin Oncol. 2018;36(suppl 4):627. ascopubs.org/doi/abs/10.1200/JCO.2018.36.4_suppl.627.
11. Array BioPharma receives FDA breakthrough therapy designation for Braftovi in combination with Mektovi and cetuximab for BRAFV600E-mutant metastatic colorectal cancer [press release]. Boulder, CO: Array BioPharma; August 7, 2018. investor.arraybiopharma.com/news-releases/news-release-details/array-biopharma-receives-fda-breakthrough-therapy-designation. Accessed November 27, 2018.
12. Shitara K, Yamanaka T, Denda T, et al. Reverce: randomized phase II study of regorafenib followed by cetuximab versus the reverse sequence for metastatic colorectal cancer patients previously treated with fluoropyrimidine, oxaliplatin, and irinotecan. J Clin Oncol. 2018;36(suppl 4):557. ascopubs.org/doi/abs/10.1200/JCO.2018.36.4_suppl.557.
13. Tsuji Y, Shitara K, Yamanaka T, et al. REVERCE: randomized phase II study of regorafenib followed by cetuximab versus the reverse sequence for metastatic colorectal cancer patients previously treated with fluoropyrimidine, oxaliplatin, and irinotecan—biomarker analysis. J Clin Oncol. 2018;36(suppl):3510. ascopubs.org/doi/abs/10.1200/JCO.2018.36.15_suppl.3510.
14. Bekaii-Saab TS, Ou FS, Anderson DM, et al. Regorafenib dose optimization study (ReDOS): randomized phase II trial to evaluate dosing strategies for regorafenib in refractory metastatic colorectal cancer (mCRC): An ACCRU Network study. J Clin Oncol. 2018;36(suppl 4):611. ascopubs.org/doi/abs/10.1200/JCO.2018.36.4_suppl.611.
15. Roche provides update on Phase III study of Tecetriq (atezolizumab) and Cotellic (cobimetinib) in people with heavily pre-treated locally advanced or metastatic colorectal cancer [press release]. Basel, Switzerland: Roche; May 10, 2018. roche.com/media/releases/med-cor-2018-05-10.htm. Accessed November 27, 2018.
16. Bendell J, Ciardiello F, Tabernero J, et al. Efficacy and safety results from IMblaze370, a randomised Phase III study comparing atezolizumab + cobimetinib and atezolizumab monotherapy vs regorafenib in chemotherapy-refractory metastatic colorectal cancer. Ann Oncol. 2018;29(suppl 5):mdy208.002. doi: 10.1093/annonc/mdy208.003.
17. Overman MJ, Lonardi S, Wong KYM, et al. Durable clinical benefit with nivolumab plus ipilimumab in DNA mismatch repair-deficient/microsatellite instability-high metastatic colorectal cancer. J Clin Oncol. 2018;36(8):773-779. doi: 10.1200/JCO.2017.76.9901.
18. FDA. FDA grants accelerated approval to ipilimumab for MSI-H or dMMR metastatic colorectal cancer. FDA website. fda.gov/drugs/informationondrugs/approveddrugs/ucm613227.htm. Published July 10, 2018. Updated July 11, 2018. Accessed November 28, 2018.