Encouraging Data Pave the Way for New Standards in CRC

Kristen Ciombor, MD, MSCI

As the development of systemic treatments for metastatic colorectal cancer (CRC) move toward precision therapy, guidelines have adjusted to integrate decisions based on molecular drivers. For example, immunotherapeutic approaches can be based on mismatch repair deficient (dMMR) and microsatellite instability (MSI) status, whereas targeted treatments such as antibody-drug conjugates (ADCs) and tyrosine kinase inhibitors (TKIs) are preferred for patients with RAS or BRAF mutations, or HER2 amplifications.¹

“When we’re thinking about molecular testing in CRC, in particular for patients with metastatic disease, there are a number of things that we think about now,” Kristen Ciombor, MD, MSCI, said in a recent OncLive Peer Exchange^®. “Unfortunately, CRC is very prevalent with approximately 150,000 cases in the United States projected in 2023, with a 2:1 prevalence of colon vs rectal cancer. We’re also seeing it in our younger patients that we see in the clinic, so it can really span many decades in terms of age and stages as well. We tend to see more localized colon cancer and rectal cancer but unfortunately there is still a good percentage of presentations in the metastatic setting or recurs after definitive treatment for localized disease. So, we need to think about how best to treat these patients and hopefully cure more patients with our research and our advancements in drug development.”

Ciombor moderated a discussion with a panel of experts in gastrointestinal cancers following the 2023 American Society of Gastrointestinal Cancers Symposium (ASCO GI). The discussion centered on updated data in the HER2 space, but also included conversations on ongoing improvements with other approaches for metastatic disease.

Testing: When, Where, For What?

Testing standards kicked off the discussion as the panel reviewed the practices that are guiding their decision-making from the time of diagnosis. “We need to test every GI cancer for dMMR indicating that they might have MSI, which would make patients a candidate for immunotherapy, but it also identifies potential for the inherited Lynch syndrome—that’s first and foremost the molecular marker testing that I am doing,” Joleen M. Hubbard, MD, said. “Then I am [using] next-generation sequencing [NGS] at first diagnosis of metastatic disease. The main drivers I am looking for up front are any RAS mutations, BRAF mutations, and HER2 amplifications as those have the most predictive markers that we have in terms of response to our current treatments. But I also like to identify any other mutation that may identify a clinical trial opportunity in the future.”

Joel R. Hecht, MD, said that reflex testing using immunohistochemistry (IHC) was once standard practice for all patients regardless of disease status; however, that may no longer be the only path forward. “The question that has come up is about testing for individual markers vs NGS. And in fact, together with my colleagues, we’ve been having multiple discussions with our pathologist at UCLA to determine the way forward.

Older panels we had were ones that probably have the most use [identifying] RAS, BRAF, and HER2. But then the question comes up about NGS and there’s some problems with that, one of which is workflow. Not all patients have had their tissue [taken] at UCLA. It’s easier if the biopsy was originally done at your institution, but you may be getting it from a surgical center where they sent it from the pathology [team] or from a small hospital where the individual had their original tissue [taken]. The other problem really that comes up with NGS, and we’re trying to bring it back in house to a certain extent, has been time, especially for first-line therapies where that may make a difference depending on whether the patient has a RAS mutation or not. How long does it take to get [results]? The third thing we haven’t quite figured out, is how it all goes together—tissue-based assays vs ctDNA[circulating tumor DNA]—which may come back sooner. It’s really in flux. We do tend to do NGS for all patients. The question is exactly when and how and I think that’s up in the air.”

Kanwal P.S. Raghav, MBBS, MD, noted that NGS may be more helpful in identifying a wider panel of markers with 1 test and that it also provides the opportunity to have markers that may clinically actionable as well as those that may be useful for future trials. He also said that the feasibility of this approach is somewhat limited. “NGS [use] in large academic centers is clearly established…whether this can be adopted on a wide scale across community or in countries outside the United States is yet to be determined because the cost can sometime be prohibitive.”

Testing standards of the past still hold a role in the identification of valuable markers; however, the growing arguments for blood-based testing as standard practice to identify tumor shedding with ctDNA and minimal residual disease (MRD) have been reflected in data presented over the past year. “It depends on the context,” Pashtoon M. Kasi, MD, MS, said. “In advanced metastatic settings where it’s mainly NGS platforms looking for MRD or leftover DNA after curative intent therapy, those are different platforms looking at different methods…there are so many variables that go into play in terms of not just the assay. In addition to looking at the assay you’re also looking at the timing of collection, the biology, the shedding, the platform, etc. I think we’ve gone beyond concordance. It’s not a question of is the assay positive; it’s more so what to do with the results of those assays.”

ctDNA

Monitoring of ctDNA throughout systemic treatment has been supported by data confirming its role as an early indicator of relapse. “There are multiple applications of ctDNA whether it is [detecting] MRD, or monitoring metastatic disease, or any kind of treatment [efficacy] vs actually looking at acquired genomic alterations or resistance mechanisms or clinical trials,” Raghav said. “There are major barriers [and] part of it is education, and these are getting complex. What platform to use? How do you interpret it? What does variant allele frequency really mean? What is a clonal mutation? What is a subclonal mutation? These are all a part of education and out in the community because the literature is growing so rapidly, it will be hard to keep up with that.”

Optimizing the timeline for collecting samples is ongoing, but encouraging data from studies such as GALAXY, the observational arm of the Japanese CIRCULATE-Japan study (UMIN000039205), have demonstrated that risk of relapse can be identified through monitoring of this marker. The study analyzed pre- and postsurgical ctDNA data from 1039 patients with stage II to IV CRC.² Patients who had ctDNA present at 4 weeks following surgery had a higher risk of recurrence than those who did not (HR, 10.0; P < .0001).²

Kasi noted that although this study did identify risk, the timing of the analysis may need to be pushed forward. “Clinically, that’s the time where we want to start out doing chemotherapy. You want to get information on their MRD status earlier on,” he said. “Looking at real-world workflow, patients are being seen within a week or 2 of their operation,” Kasi noted regarding the timeline of tacking on the first analysis of ctDNA. In a retrospective analysis of kinetics identified using the polymerase chain reaction (PCR)-based NGS platform Signatera to compare cell-free DNA (cfDNA) with ctDNA, there are fluctuations in the correlation between the 2 markers at various time points after surgery.3 Investigators hypothesized that immediately after surgery levels of cfDNA from notmal tissue could interfere with detection of MRD via ctDNA analysis. “The question is: Will the background noise of cfDNA muffle up the signal where would you have patients that are falsely negative identified but may still have disease?” Kasi said.

In the study investigators observed that cfDNA levels were increased in the immediate postoperative period of 0 to 2 weeks and there was a variability in week 1 analyses of ctDNA positivity, but at week 2 ctDNA positivity was more consistent with later time points. Despite levels of cfDNA remaining high during weeks 2 through 4, ctDNA was consistently detected and positivity outcomes were similar to those captured in weeks 4 through 12. In a cohort of patients with the detection of ctDNA at 0 to 2 weeks, cfDNA was detected at rates of 18.4% (n = 365) compared with 17.3% (n = 110) at weeks 3 through 4 and 18.0% (n = 206) at weeks 5 through 6.³

Investigators concluded that ctDNA may be captured at a minimum of 2 weeks following surgery in patients with CRC; however, they note that a larger cohort of patients will be required to validate the results.³

Staying on Top of Trends

The panel concurred that the plethora of companies offering assays is rapidly changing and that cross-platform comparisons are tricky, especially as they become more sensitive and improve over time. “It’s important as we are digesting this information to look at the methods more than ever before with this rapidly evolving technology,” Kasi said, adding that the crowding in this space is a good problem.

Raghav noted that in addition to the prognostic capabilities of ctDNA, detecting MRD can be used for EGFR rechallenge in patients with CRC. “One of the applications, which I think is ready for clinical use is rechallenge with anti-EGFR, which is based on the precept that whenever you treat patients who have RAS wild-type disease with anti-EGFR therapies—the mainstay of treatment for CRC—they subsequently will develop some sort of resistance. This acquired resistance is because of acquisition of KRAS or NRAS mutations and others, usually they present with low clonality and with passage of time, if you remove that treatment pressure, they are going to go away. That’s why EGFR rechallenge works unlike many of the other therapies, such as chemotherapy rechallenge.”

Raghav added that although most clinical judgments can be made after 6 or 8 months, which is when the acquired resistance may have disappeared, studies evaluating ctDNA have afforded the opportunity to make a “subjective assessment in an objective category.” The phase 2 CHRONOS study (NCT03227926) examined the efficacy of rechallenge with the EGFR inhibitor panitumumab following acquired RAS/ BRAF/EGFR resistance mechanisms.⁴ A droplet PCR-based assay was used to screen 52 patients following prior treatment with an anti-EGFR agent with or without chemotherapy. Among 27 patients eligible for EGFR rechallenge, the overall response rate was 30% (95% CI, 12%-47%) with a median duration of response of 17 weeks. For the 40% of patients who achieved stable disease, this lasted for at least 4 months in 82%. The disease control rate was 63% (95% CI, 41%-78%).⁴

“[This response] compares very favorably with what you would have done for standard of care. Of course, it was a smaller study, so you need larger validations for it. But I think that was a good proof of concept of how this technology can be used in the clinical practice,” Raghav said.

HER2 Takes the Spotlight

Approximately 3% to 5% of patients with CRC are positive for HER2 with enrichment seen among those who do not have a RAS or BRAF mutation, with an incidence upward of 10%.1 “When you think about 150,000 new cases per year, it is a significant number of patients that do have HER2-positive disease,” Hubbard said.

The gold standard for HER2 is IHC, Ciombor said, but in situ hybridization as well as amplification via an NGS panel are also becoming standard. “When we use ctDNA, sometimes that can underestimate the prevalence, we have to be a little bit cautious,” Ciombor said.

Testing for HER2

At ASCO GU, Andrea Cercek, MD, presented on testing for HER2 positivity trends in CRC, which is established very well for gastric and breast cancer.⁵ Cercek’s presentation was an evaluation of patients treated in the phase 2 MOUNTAINEER trial (NCT03043313) using breast or gastric cancer algorithms to determine HER positivity. Among 105 patients from the trial, the concordance rate between central laboratory confirmed HER2 status by breast and gastric cancer algorithms was 100%. For the central laboratory–confirmed HER2 IHC score by breast and gastric algorithms, there was a 99% concordance rate between algorithms.⁵

“And interestingly, it was 100% concordance between the 2 assays in the IHC score that they achieved,” Ciombor said. “Either way was 99% concordance. So, I think that tells us we can probably establish a standard IHC test for CRC either based on the HER2 breast or gastric algorithm. That’s important to get that established.” Regarding efficacy of HER2-directed agents, MOUNTAINEER evaluated tucatinib (Tukysa) in combination with trastuzumab (Herceptin) for the treatment of previously treated HER2-positive metastatic CRC.⁶ Patients had already received at least 2 lines of systemic therapy, one of which had to be an anti-VEGF monoclonal antibody. Patients in cohorts A and B (n = 86) received both tucatinib and trastuzumab and patients in cohort C received tucatinib along (n = 31).⁶

At a median follow-up of 20.7 months, the combination elicited an objective response rate (ORR) of 38.1% among 86 patients, with a median duration of response (DOR) of 12.4 months. The disease control rate (DCR) was 71.4%. Median progression-free survival was 8.2 months with an overall survival of 24.1 months.⁶

The combination was approved by the FDA in January 2023.⁷ Updated findings from cohort C of the trial showed that patients who received tucatinib monotherapy had an ORR of 3.3% (95% CI, 0.1%17.2%), with 1 patient experiencing a partial response. Stable disease was reported in 76.7% of patients for a DCR of 80%. Protocol allowed for crossover and among the 28 patients who went on to receive the combination, the ORR was 17.9% (95% CI, 6.1%-36.9%), with 5 partial responses and 18 patients with stable disease for a DCR of 82.1%.⁶

In terms of safety, Hubbard noted that the combination is well tolerated. “I saw a couple patients have some transaminase elevation. We just dose reduced a little bit, and the patients did fine on treatment,” she said. “So, not only does it have great activity, but it’s also actually well tolerated. I’m most excited that this is an option for individuals who are continuing to live their lives 3 or 4 years after a diagnosis of metastatic disease.”

Trastuzumab Deruxtecan

The panel turned to fam-trastuzumab deruxtecan-nxki (Enhertu), an ADC with activity in several HER2-positive malignancies. The primary analysis of the phase 2 DESTINYCRC01 (NCT03384940), presented at ASCO GI, showed promising activity among patients with HER2-expressing mCRC. A total of 86 patients with a median of 4 prior therapies (range, 2-11) were treated with the ADC.⁸

At a median follow-up of 64.2 weeks, the confirmed ORR was 45.3% (95% CI, 31.6%-59.6%) among 53 evaluable patients, with a median DOR was 7.0 months. The DCR was 83.0% with a median PFS of 6.9 months (95% CI, 4.1-8.7) and median OS of 15.5 months (95% CI, 8.8-20.8).8 The agent showed similar efficacy in those without prior anti-HER2 treatment whereas the confirmed ORR was 43.8% and among those with prior anti-HER2 therapy it was 57.5%.⁸

“One of the things that we know from upper GI is that it’s been very difficult, other than with this ADC, to treat because often individuals will actually lose the HER2 positivity,” Hecht said. “Like ctDNA, we have sort of an embarrassment of riches. We have several active drugs. None of them are curative so I don’t want to get carried away. But when you have several active drugs, you start talking about sequencing.”

Hecht added that toxicity with these regimens is also important when it comes to sequencing. “The payload is cytotoxic [with ADCs]. So, we have toxicity with standard cytotoxic drugs and we have toxicity with this, with interstitial lung disease [ILD] of course being the fear.” In the study ILD was reported among 8 patients. “One thing I will say as this diffuses into the community is that our community physicians may be more aware of how to use this drug because they’re using it in breast cancer as opposed to the GI oncologist. But I think the point is that it is important to sort of stay on top of that from a toxicity standpoint,” Hecht said.

“In patients who are RAS mutation positive, this is where I think the strategy of an ADC trumps the strategy of dual anti-HER2 therapy,” Raghav said. “It is very clear to us that dual anti-HER2 therapy, whether it is trastuzumab/ lapatinib [Tykerb], trastuzumab/pertuzumab [Perjeta], it does not work in the RAS-mutant population for obvious reasons. ADC strategy, on the other hand, I think can circumnavigate that limitation. So, DESTINY-CRC02 [NCT04744831] study allows a RAS-mutant population. There was an explorational analysis of the DESTINYCRC study, which was a biomarker analysis and there were some patients who had ctDNA evidence off RAS mutations and the response rates were still seen in those patients.”

References

1. Ahcene Djaballah S, Daniel F, Milani A, Ricagno G, Lonardi S. HER2 in colorectal cancer: the long and winding road from negative predictive factor to positive actionable target. Am Soc Clin Oncol Educ Book. 2022;42:1-14. doi:10.1200/EDBK_351354
2. Kotani D, Oki E, Nakamura Y, et al. Molecular residual disease and efficacy of adjuvant chemotherapy in patients with colorectal cancer. Nat Med. 2023;29(1):127-134. doi:10.1038/s41591-022-02115-4
3. Cohen SA, Kasi PM, Aushev VN, et al. Kinetics of postoperative circulating cell-free DNA and impact on minimal residual disease detection rates in patients with resected stage I-III colorectal cancer. J Clin Oncol. 2023;41(suppl 6):5. doi:10.1200/JCO.2023.41.4_ suppl.5
4. Sartore-Bianchi A, Pietrantonio F, Londari S, et al. Circulating tumor DNA to guide rechallenge with panitumumab in metastatic colorectal cancer: the phase 2 CHRONOS trial. Nat Med. 2022;28(8):16121618. doi:10.1038/s41591-022-01886-0
5. Cercek A, Ng K, Strickler JH, et al. HER2 testing in colorectal cancer: concordance analysis between breast and gastric scoring algorithms from the MOUNTAINEER trial. J Clin Oncol. 2023;41(suppl 6):198. doi:10.1200/JCO.2023.41.4_suppl.198
6. Strickler JH, Cercek A, Siena S, et al. Additional analyses of MOUNTAINEER: a phase II study of tucatinib and trastuzumab for HER2-positive mCRC. Ann Oncol. 2022;33(suppl 7):1394. doi:10.1016/j.annonc.2022.08.023
7. FDA grants accelerated approval to tucatinib with trastuzumab for colorectal cancer. FDA. January 19, 2023. Accessed March 19, 2023. bit.ly/3Zc7jWW
8. Yoshino T, Di Bartolomeo M, Raghav KPS, et al. Trastuzumab deruxtecan (T-DXd; DS-8201) in patients (pts) with HER2- expressing metastatic colorectal cancer (mCRC): final results from a phase 2, multicenter, open-label study (DESTINY-CRC01). J Clin Oncol. 2022;40(suppl 4):119. doi:10.1200/JCO.2022.40.4_ suppl.119