Genomic Findings Reshaping CRC Field

Cathy Eng, MD, professor of gastrointestinal medical oncology at MD Anderson Cancer Center

Cathy Eng, MD

Although molecular profiling has established colorectal cancer (CRC) as a complex and heterogeneous tumor type, recent research findings are breaking fresh ground in clarifying the extent of genetic mutations among patients diagnosed with the malignancy and the potential for translating those insights into clinically useful therapies.

The next significant development in the field likely will be the FDA’s much-anticipated decision on whether the PD-1 inhibitor pembrolizumab (Keytruda) gains an indication as a therapy for previously treated patients with microsatellite instability-high (MSI-H) cancers, which is believed to account for approximately 15% of early CRCs and about 4% of metastatic tumors.¹ The agency is scheduled to make a decision on the application by June 9; if the ruling is positive, pembrolizumab would be the first immunotherapy approved for CRC.

Immunotherapy Developments

The potential for checkpoint immunotherapy agents to improve outcomes for patients with gastrointestinal malignancies has been moving forward during the past 2 years, primarily due to indications of efficacy in MSI-H tumors.

In updated phase II findings presented at the 2016 American Society of Clinical Oncology (ASCO) Annual Meeting, the overall response rate (ORR) was 57% with pembrolizumab in patients with MMR-deficient advanced CRC (16 of 18 patients) including an 11% complete response rate compared with 0% in patients with MMR-proficient tumors (0 of 25).³ Median progression-free survival (PFS) and overall survival (OS) were not reached after a median 9.3-month follow-up.

Results from that ongoing phase II study prompted the FDA to grant pembrolizumab a breakthrough therapy designation as a potential treatment for patients with MSI-H metastatic CRC in November 2015. The pending application includes clinical trial data from 5 multicohort studies in MSI-H cancers, according to Merck, which is developing the drug.

Positive outcomes for checkpoint blockade agents in MSI-H metastatic CRC have sparked interest not only in single-agent immunotherapy but also in combinations. “Immune checkpoint inhibitors have been found to be beneficial for patients as a single agent if they are MSI high,” said Cathy Eng, MD, a professor of gastrointestinal medical oncology at The University of Texas MD Anderson Cancer Center in Houston. “That’s really changed the paradigm of care in regard to looking at the benefits of immunotherapy in colorectal cancer.”

Eng said that combination trials pairing check- point blockade agents with targeted therapies are seeking to boost response rates for patients with CRC; she noted the combination of PD-1/PD-L1 pathway agents with MEK inhibition as an area of active research.

In a phase Ib study presented at the 2016 World Congress on Gastrointestinal Cancer, the PD-L1 inhibitor atezolizumab (Tecentriq) was paired with the MEK inhibitor cobimetinib (Cotellic) in heavily pretreated patients with microsatellite stable metastatic CRC.⁴ The combination demonstrated an investigator-assessed ORR of 17% in 23 patients overall. An ORR of 20% was observed in 22 patients with KRAS-mutant tumors.

In light of those positive results, a 3-arm phase III trial has been launched that compares the combination of atezolizumab and cobimetinib with atezolizumab monotherapy or with regorafenib (Stivarga) in patients with unresectable locally advanced or metastatic CRC who have received at least 2 prior regimens of cytotoxic chemotherapy.⁵ The trial is open to patients regardless of MSI or RAS mutation status; however, the estimated enrollment of 360 participants would be capped at 5% for MSI-H patients and at 50% for those with wild-type RAS disease.⁵

Genetic Testing

Other combinations under study include the pairing of nivolumab (Opdivo) with ipilimumab (Yervoy) in patients with MSI-H CRC. In findings from the phase II CheckMate-142 study presented at the 2016 ASCO Annual Meeting, the combination demonstrated an investigator-assessed ORR of 33.3% and median PFS at 6 months of 66.6% compared with 25.5% and 45.9%, respectively, for nivolumab monotherapy.¹ Additional studies combining the 2 immunotherapy agents are planned.MSI status is playing an increasingly important role in the treatment of patients with CRC. Testing is often performed at the time of diagnosis in order to “identify patients with Lynch syndrome, direct adjuvant chemotherapy, or determine prognosis,” according to recently updated guidelines on molecular testing in CRC that 4 medical societies jointly issued in February.⁶ Testing for MSI can be performed through immunohistochemistry to detect the loss of 4 MMR proteins (MLH1, MSH2, PMS2, and MSH6) expressed in tumor cell nuclei or through MSI DNA-based testing.⁶

Study findings show that patients with MSI-H status have a favorable prognosis for OS and disease-free survival depending upon stage of disease; some trial results also suggest that MSI-H patients with early-stage disease do not respond to 5-fluorouracil—based adjuvant chemotherapy.⁶ Conversely, MSI-H status, which is indicative of a higher mutational load, has been predictive of bene t from checkpoint immunotherapy in study findings from patients with metastatic CRC.

“Every patient with colon cancer needs their MSI tested,” said John L. Marshall, MD, director of the Otto J. Ruesch Center for the Cure of Gastrointestinal Cancer at the Georgetown Lombardi Comprehensive Cancer Center in Washington, DC, in an interview. “If a patient is MSI-high in stages I through III, we know their prognosis is good, better in fact, and therefore the role of chemotherapy is less clear.

“In metastatic disease, those patients have a worse prognosis and tend to be right-sided colon cancers but we now have checkpoint inhibitors,” Marshall said. “We are looking for patients with MSI-high colon cancers so that we can identify those who are at familial risk, those who should get adjuvant therapy or not, and those who should get checkpoint inhibitors.”

Indeed, study results published in the Journal of Clinical Oncology in April indicate that the prevalence of germline cancer susceptibility gene mutations is greater than previously believed.⁷ The study, which was conducted at the Dana-Farber Cancer Institute in Boston, included 1058 patients with CRC who were not preselected for age at diagnosis, personal/family history, MSI status, or MMR deficiency. Germline testing was conducted for 25 genes linked to inherited cancer risk.

The results showed that 9.9% of patients had at least 1 pathogenic mutation, including 3.1% with Lynch syndrome. Non-Lynch syndrome gene mutations were detected in 7% of patients, including BRCA1/2 mutations in more than 1% of patients.

Matthew B. Yurgelun, MD, lead author of the study, said researchers have historically believed that inherited factors account for approximately 3% to 4% of patients with CRC. When researchers included a wider pool of patients, however, a greater incidence of mutations was found, noted Yurgelun, a physician at Dana-Farber and an assistant professor of medicine at Harvard Medical School.

“These individuals were not specifically selected for factors we traditionally associate with inherited risks of cancer,” Yurgelun said in an interview. “For example, these individuals were not specifically selected for being young at the time of their colon cancer diagnosis, they were not specifically selected based on factors in their tumor, or based on any sort of family history of cancer. The fraction of individuals found to have inherited mutations in cancer susceptibility genes was quite a bit higher than what we would have expected in the past.”

The findings raise the question of whether current molecular testing is extensive enough to identify patients who would benefit from targeted therapies. The recently updated guidelines recommend RAS mutational testing for patients who are being considered for anti-EGFR therapy, BRAFV600 status for prognostic stratification and Lynch screening; and MMR testing for all patients with CRC.⁶

“One of the tricky things from our study is to answer the question of which individuals need germline testing,” said Yurgelun. “We’ve historically recommended germline testing for any CRC patient who has a particularly strong family history of colon cancer; any colon cancer patient with a strong family history of endometrial cancer, ovarian cancer, and other cancers linked to Lynch syndrome; and any CRC patient who has a significant number of colorectal adenomas or other polyps.

“Our study found that if you look beyond Lynch syndrome, which has historically been known as the most common inherited CRC syndrome, you end up finding a lot of inherited mutations and there doesn’t seem to be, at least from our data, a lot of specific factors that tell us which individuals need germline testing,” continued Yurgelun. “It raises the question of whether we should be performing germline testing on all individuals with CRC, which is a pretty slippery slope.

“I think the counterargument to that is that some of these inherited mutations still have a lot of questions behind them, even when we find these mutations, so we’re not quite ready to be recommending germline testing for all individuals with CRC,” he added. “I think we would come up with as many questions as we do answers if we were to do that.”

Marshall believes that broad molecular profiling has the potential to yield clinically useful information and is becoming more common in biomarker-guided trials that seek to match patients with drugs that have proved effective in different histologies.

Tumor Sidedness

“I believe this is where the field is going,” Marshall said. “We’re going to want to know this in everyone. But we’re dancing a line right now between what is standard of care and what is research. Our patients with metastatic disease, I think, should all be tested using broad pro ling because we know we’re not going to cure them [but we can] see if they’re eligible for one of these other drugs that are out there under these clinical trials.”Although differences in treatment outcomes based on tumor sidedness in CRC have been described in oncology research for more than 15 years, the molecular underpinnings of those variations have come into sharp focus as a result of research presented at the 2016 ASCO Annual Meeting.⁸

Retrospective findings from the CALGB 80405 study demonstrated that survival outcomes in patients with KRAS wild-type metastatic colorectal cancer were significantly longer among those with tumors originating on the left versus the right side of the colon.

The median OS was nearly 14 months longer in patients with left-sided tumors compared with right-sided tumors, including a nearly 20-month survival advantage in patients receiving frontline cetuximab (Erbitux) plus chemotherapy and a more than 7-month survival benefit in patients receiving frontline bevacizumab (Avastin) plus chemotherapy.

Overall, researchers found that median OS was 33.3 months for patients with left-sided primary tumors compared with 19.4 months for right-sided tumors. Among patients who received cetuximab, the median OS was 36 months in the left-sided cohort versus 16.7 months in the right-sided group; in the bevacizumab groups, the median OS was 31.4 months for left-sided tumors versus 24.2 months for right-sided malignancies.

Alan P. Venook, MD, the lead investigator on the study, and others in the field believe the tumor-sidedness findings reflect variations in molecular aberrations and that further study is needed to elucidate the differences.

“It is not a function of right versus left side; it’s molecular features that appear to be nonrandomly distributed across the colon,” said Venook, a professor of medicine at the University of California, San Francisco, in an interview. “I am hoping that we will have the opportunity in the next year or 2 to explain not only why patients do better on the left versus right with certain drugs, but also to prospectively understand why that is and how we should be treating patients. Right now, we do very little in terms of biomarker decision making; we decide what not to use instead of what to use. Hopefully, we will do the opposite in the future.”

Marshall is among those researchers seeking to advance understanding of the molecular differences at work in these cancers. At the 2017 Gastrointestinal Cancers Symposium, Marshall and colleagues presented study findings indicating that there are molecular variations between left-sided colon tumors and rectal tumors.⁹

Significantly higher rates of TP53 and APC mutations were found in rectal versus descending colon cancers, while the prevalence of PIK3CA and BRAF mutations was higher in the colon cancers. However, in another study, multivariate analysis found no differences in OS or PFS in patients with left-sided colon cancer versus rectal cancer.¹⁰

“The reason this is important is because doctors are now thinking clinically about if it is a right-sided or left-sided tumor,” Marshall said in discussing the 2 studies. “We should bundle rectal cancer in with left-sided colon cancer for the moment. These 2 studies determined that we don’t need to additionally stratify for rectal cancers.”

Looking forward, Marshall sees an ongoing need for research into tumor location and molecular features. “A next step includes continuing to conduct prospective studies looking at the molecular characterization of the right side versus left side,” he said. “That’s going to help us in terms of figuring out how to act in a metastatic setting. It’s not clear to me that even with next-generation sequencing that we’re going to find the secret gene.”

References

1. Overman MJ, Kopetz S, McDermott RS, et al. Nivolumab ± ipilimumab in treatment (tx) of patients (pts) with metastatic colorectal cancer (mCRC) with and without high microsatellite instability (MSI-H): CheckMate-142 interim results. Presented at: 2016 ASCO Annual Meeting; June 3-7, 2016; Chicago, IL. Abstract 3501. meetinglibrary. asco.org/content/166455-176.
2. Kim S-E, Paik HY, Yoon H, Lee JE, Kim N, Sung MK. Sex- and gender-specific disparities in colorectal cancer risk. World J Gastroenterol. 2015;71(17):5167-5175. doi: 10.3748/wjg.v21.i17.5167.
3. New data from phase 2 study evaluating Keytruda (pembrolizumab) for the treatment of cancers deficient in DNA mismatch repair show durable responses across a range of cancers [news release]. Kenilworth, NJ: Merck; June 5, 2016. businesswire.com/news/ home/20160605005086/en/Data-Phase-2-Study-Evaluating-KEY- TRUDA%C2%AE-pembrolizumab. Accessed May 15, 2017.
4. Bendell J, Kim TW, Chee CE, et al. Safety and efficacy of cobimetinib and atezolizumab in a phase 1b study of metastatic colorectal cancer (mCRC). Presented at: ESMO 18th World Congress on Gastrointestinal Cancer; June 29-July 2, 2016; Barcelona, Spain. Abstract LBA-01. doi: org/10.1093/annonc/mdw237.01.
5. NIH Clinical Trials Registry. A study to investigate efficacy and safety of cobimetinib plus atezolizumab and atezolizumab monotherapy versus regorafenib in participants with metastatic colorectal adenocarcinoma. ClinicalTrials.gov website. Identifier: NCT02788279. Updated April 3, 2017. Accessed May 10, 2017.
6. Sepulveda AR, Hamilton SR, Allegra CJ, et al. Molecular biomarkers for the evaluation of colorectal cancer: guideline from the American Society for Clinical Pathology, College of American Pathologists, Association for Molecular Pathology, and the American Society of Clinical Oncology. J Clin Oncol. 2017;35(13):1453-1486. doi: 10.1200/JCO.2016.71.9807.
7. Yurgelun MB, Kulke MH, Fuchs CS, et al. Cancer susceptibility gene mutations in individuals with colorectal cancer.J Clin Oncol. 2017;35(10):1086-1095. doi: 10.1200/JCO.2016.71.0012.
8. Venook AP, Niedzwiecki D, Innocenti F, et al. Impact of primary (1o) tumor location on overall survival (OS) and progression-free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance). Presented at: 2016 ASCO Annual Meeting; June 3-7, 2016; Chicago, IL. Abstract 3504. meetinglibrary.asco.org/content/161936-176.
9. Marshall J, Lenz HJ, Xiu J, et al. Molecular variances between rectal and left-sided colon cancers. Presented at: 2017 Gastrointestinal Cancers Symposium; January 19-21, 2017; San Francisco, CA. Abstract 522. meetinglibrary.asco.org/content/176344-195.
10. Salem ME, Yin J, Renfro LA, et al. Rectal versus left-sided colon cancers: Clinicopathological differences observed in a pooled analysis of 4,182 patients enrolled to 8 clinical trials from the ARCAD database. Presented at: 2017 Gastrointestinal Cancers Symposium; January 19-21, 2017; San Francisco, CA. Abstract 675. meetinglibrary.asco.org/content/176249-195.

MSI, which is caused by mismatch repair deficiency (MMR) in the DNA mechanism, is among the genomic markers that are growing increasingly relevant in treatment plan recommendations for patients with CRC. Experts in the field expect that trend to continue, particularly as more knowledge is gained about the correlation between tumor sidedness and molecular aberrations (Figure²).