ATOMIC Looks at Immunotherapy in Early-Stage dMMR Colon Cancer

Frank Sinicrope, MD

Investigators are seeking to determine whether immunotherapy that has been effective in metastatic colon cancers with deficient DNA mismatch repair (dMMR) can be applied to earlier-stage disease, specifically, nonmetastatic stage III colon cancer.

In the phase III ATOMIC trial (NCT02912559), investigators are evaluating adjuvant mFOLFOX6 (modified leucovorin calcium, fluorouracil, and oxaliplatin for 6 months) with or without atezolizumab (Tecentriq) to determine whether mFOLFOX6 with the anti—PD-L1 antibody confers a greater survival benefit than standard chemotherapy alone. The trial seeks to enroll 700 participants with nonmetastatic stage III disease and dMMR and has enrolled 152 patients as of February (Figure).¹ ATOMIC is the only adjuvant trial in the United States currently examining immunotherapy in stage III colon cancer, according to the trial’s principal investigator, Frank Sinicrope, MD.

“This trial was based on the data that were generated on patients with metastatic disease in whom immune checkpoint inhibitors were shown to be very active—patients whose tumors had dMMR,” Sinicrope said in an interview with OncologyLive^®. “Due to their remarkable results in this subset of colorectal cancers, we decided that perhaps this therapy would also be of benefit to an earlier stage of disease, namely, the stage III population that gets adjuvant chemotherapy following a potentially curative resection.”

Figure. Atezolizumab With Combination Chemotherapy in Stage III dMMR Colon Cancer (Click to Enlarge)

A Biomarker-Driven Approach

Investigators are enrolling patients whose tumors have MSI due to dMMR. In ATOMIC, dMMR status serves as the biomarker predictive of response to atezolizumab. Patients’ tumor tissue is stained for the expression of 4 proteins—MLH1, MSH2, MSH6, and PMS2—that govern MMR. The absence of 1 or more of the proteins, as detected by immunohistochemical (IHC) staining, denotes dMMR.² In contrast, a normal IHC test implies that all 4 MMR proteins are expressed and retained.⁴ Approximately 15% of CRCs have dMMR, and this rate decreases by stage.²

Recommended by the National Comprehensive Cancer Network’s (NCCN’s) clinical practice guidelines for managing CRC, IHC testing can be used to guide the treatment of patients with CRC based on dMMR status. In the metastatic setting, patients with IHC-confirmed dMMR constitute a biomarker-defined population that benefits less from conventional chemotherapy than do populations with MMR-proficient mCRC.² Further, whereas dMMR status correlates with response to anti—PD-1 therapies, single-agent checkpoint blockade typically does not elicit a response in those with MMR-proficient disease.

Although the NCCN’s recommendation for universal IHC testing “has been adopted somewhat widely, there may well be a number of patients who are not having their tumors tested, which is potentially detrimental to outcomes,” Sinicrope said.

The absence of key repair proteins detected during IHC staining may indicate germline mutations and events, such as Lynch syndrome. This is the most common form of genetically determined colon cancer predisposition and accounts for 2% to 4% of all CRCs.⁴ “There are specific guidelines for the evaluation and management of patients with Lynch syndrome who are also at risk for multiple other cancers,” Sinicrope said, “so [it’s crucial] that IHC testing be universally adopted in clinical practice.”

ATOMIC eligibility requirements allow dMMR status to be determined locally or by site-selected reference lab, but dMMR status must be established prior to participation. Patients known to have Lynch syndrome or have a sporadic form of dMMR are also eligible. This situation is more common than Lynch syndrome whereby the MLH1 mismatch repair gene is inactivated by DNA methylation of the gene, Sinicrope said.

Laying the Groundwork

To date, the use of immunotherapy in colon cancer has been limited to dMMR metastatic tumors. Multiple completed and ongoing clinical trials have been launched to determine whether non-MSI metastatic colorectal cancers benefit from immunotherapy, but most of these have been negative, according to Sinicrope.

However, previous studies have shown that CRCs with dMMR are sensitive to the immune checkpoint blockading properties of anti—PD-1 antibodies pembrolizumab (Keytruda) and nivolumab (Opdivo). For example, the phase II CheckMate-142 study (NCT02060188) compared nivolumab monotherapy with nivolumab in combination with other anticancer agents, including ipilimumab (Yervoy), cobimetinib (Cotellic), daratumumab (Darzalex), and Relatlimab (BMS-986016) in patients with recurrent and metastatic dMMR CRC.

The overall response rate for the 74 patients treated with single-agent nivolumab was 31.1%; 69% demonstrated disease control for ≥12 weeks.² Progressionfree survival was 50% and overall survival was 73% at 12 months. In July 2017, the FDA authorized nivolumab for the treatment of patients ≥12 years with dMMR CRC that has progressed following treatment with a fluoropyrimidine, oxaliplatin, and irinotecan, based on results from CheckMate-142.⁵

Data from 5 clinical trials in dMMR also led to FDA approval of pembrolizumab monotherapy for adult and pediatric patients with unresectable or metastatic MSI-high or dMMR CRC progressing following treatment with a fluoropyrimidine, oxaliplatin, and irinotecan.⁶ Patients received either pembrolizumab at 200 mg every 3 weeks or 10 mg/kg every 2 weeks for a maximum of 24 months of treatment.

There was a total of 149 patients with dMMR cancers across the 5 studies. The ORR with pembrolizumab was 39.6% (95% CI, 31.7-47.9), the complete response rate was 7.4%, and the partial response rate was 32.2% in patients with dMMR cancer. The ORR was 36% (95% CI, 26.0-46.0) among the 90 patients with CRC.⁶

The data suggested that atezolizumab is a potentially effective treatment option for patients with stage III colon cancer with dMMR. “In disabling the activity of the PD-L1 protein found on T cells, atezolizumab enables the immune system to attack and kill tumor cells,” Sinicrope said. “Data suggest that certain chemotherapeutic agents such as oxaliplatin may produce an increase immune cell infiltration in the tumor that may enhance the activity of the checkpoint inhibitor, such that giving chemotherapy in combination with atezolizumab may result in better responsiveness.”

Optimizing Anti—PD-1/PD-L1 Agents

The hope for ATOMIC is that the mFOLFOX6/atezolizumab regimen will not only improve outcomes, but also offer insight on how to optimally pair checkpoint inhibitors with chemotherapy in this setting. “We’re still learning more about how to best use checkpoint inhibitors—ie, whether alone or in combination with chemotherapy and also the timing of the administration relative to chemotherapy which is an important area of investigation,” Sinicrope said.

The current standard of care (SOC) for patients with stage III colon cancer is surgical resection followed by 3 to 6 months of chemotherapy. Adjuvant chemotherapy is of proven benefit, so trial investigators decided to offer all patents FOLFOX with or without immunotherapy. “Because we’re dealing with tumors and patients who can be cured with chemotherapy, which represents the [SOC], we need to give [atezolizumab] with chemotherapy,” he said. “We don’t believe that we could comfortably withdraw the chemotherapy completely and just give the checkpoint inhibitor, although this approach should be studied if our trial shows benefit for the addition of atezolizumab.”

Patients in both arms of the ATOMIC study will receive 6 months of adjuvant mFOLFOX6: oxaliplatin at 85 mg/m², leucovorin at 400 mg/m², and fluorouracil at 400 mg/m² + 2400 mg/m² (days 1-3) every 14 days. Those in the experimental arm also will receive atezolizumab at 840 mg every 14 days for 12 cycles, followed by atezolizumab monotherapy for 6 months.

“There is definitely a need to improve the outcome of patients with stage III disease. Given the remarkable results that we’ve seen in the metastatic population, [this] is a great opportunity to try to cure more patients with stage III disease who have dMMR tumors,” Sinicrope said.

ATOMIC’s primary endpoint is diseasefree survival. Secondary endpoints include overall survival and adverse events (AEs). Interim analyses are planned at 50% and 75% of AEs.

References

1. Sinicrope FA, Ou FS, Zemla T, et al. Randomized trial of standard chemotherapy alone or combined with atezolizumab as adjuvant therapy for patients with stage III colon cancer and deficient mismatch repair (ATOMIC, Alliance 021502). J Clin Oncol. 2019;37(suppl 15; abstr e15169). doi: 10.1200/ JCO.2019.37.15_suppl.e15169.
2. Overman MJ, Ernstoff MS, Morse MA. Where we stand with immunotherapy in colorectal cancer: deficient mismatch repair, proficient mismatch repair, and toxicity management. Am Soc Clin Oncol Educ Book. 2018;38:239-247. doi: 10.1200/EDBK_200821.
3. Passardi A, Canale M, Valgiusti M, Ulivi P. Immune checkpoints as a target for colorectal cancer treatment. Int J Mol Sci. 2017;18:(6):pii.E1324. doi: 10.3390/ijms18061324.
4. NCCN Clinical Practice Guidelines in Oncology. Colon Cancer, version 2.2019. National Comprehensive Cancer Network website. nccn.org/professionals/physician_gls/pdf/colon.pdf. Published May 15, 2019. Accessed September 17, 2019.
5. Opdivo [prescribing information]. Princeton, NJ: Bristol-Myers Squibb Company; 2017. packageinserts.bms.com/pi/pi_opdivo. pdf. Accessed September 17, 2019.
6. Keytruda [prescribing information]. Whitehouse Station, NJ: Merck Sharp & Dohme Corp; 2017. www.accessdata.fda.gov/ drugsatfda_docs/label/2017/125514s014lbl.pdf. Accessed September 17, 2019.

DNA mismatch repair (MMR) is the mechanism by which cells repair damaged DNA, including insertions, deletions, and misincorporations of nucleotides during DNA replication.² MMR deficiency is detected in approximately 15% of colorectal cancers (CRC) and leads to microsatellite instability (MSI) and a high burden of mutations which trigger the expression of neoantigens that attract lymphocytes to the tumor. Inflamed tumors have been shown to respond favorably to immune checkpoint inhibitors; therefore, anti—PD-1/PD-L1 agents constitute an attractive treatment option in patients with dMMR tumors.³ Among stage III colon cancers, approximately 11% are dMMR positive, according to Sinicrope, who is a gastroenterologist, internist, and oncologist at Mayo Clinic in Rochester, Minnesota.