High Expectations for Immunotherapy in GI Cancers

RHONDA BALL could see and feel a tumor on her stomach shrinking after she started taking Keytruda. - PHOTO BY AMY TANNENBAUM

Rhonda Ball was all too familiar with the ups and downs of cancer treatment when, in 2016, she entered a clinical trial for Keytruda (pembrolizumab), a drug that boosts the immune system. By then, she had endured surgery and chemotherapy to treat breast cancer that was diagnosed in 2009 and uterine cancer that occurred three years later. Finally, after an unrelated cancerous blockage appeared in her bowel in 2016, she was offered immunotherapy — a treatment that was possible because of a particular genetic characteristic of her tumors.

Two years earlier, three small tumors had emerged on the back of Ball’s skull, hip and stomach, two of which were removed and studied. DNA sequencing revealed an abnormal biomarker called microsatellite instabilityhigh (MSI-high), also known as mismatch repair-deficient. It meant that her cancer cells had a diminished ability to repair their own DNA. This deficiency could make them vulnerable to drugs like Keytruda, which are designed to free up the immune system to recognize and attack cancer.

From day one, Ball could tell Keytruda was working on the stomach tumor, which had started to break through her skin. “The very first day I had the treatment, the tumor was just burning, and visually I could tell it was shrinking,” says Ball, who, starting in August 2016, received infusions at the Ohio State University Comprehensive Cancer Center’s Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, in Columbus. “That made me think it was working.” Ball’s most recent scans in January 2018 showed no signs of cancer.

The Food and Drug Administration (FDA) approved Keytruda in May 2017 to treat any metastatic or nonresectable type of cancer that’s MSI-high. The first FDA approval based on a genetic characteristic of cancer rather than its location in the body, it was hailed as a breakthrough by oncologists — particularly those who treat gastrointestinal cancers. Keytruda is known as a checkpoint inhibitor because it interferes with the protein PD-1, which would otherwise keep the immune system in check, preventing it from going into overdrive to attack cancer.

“Anti—PD-1 therapy for MSI-high colorectal cancer was the first immunotherapy approved for gastrointestinal cancers, in part because this therapy works extremely well for the majority of patients who are MSI-high,” says Michael Overman, M.D., associate professor in the department of gastrointestinal medical oncology at MD Anderson Cancer Center in Houston. “And when it works, it often seems to keep working.”

An estimated 15 percent of colorectal cancers are MSIhigh, as are up to 39 percent of gastric tumors, although the mutational status is less common in advanced disease. About 3 percent of colorectal cancers with the abnormality occur in patients who have Lynch syndrome — an inherited collection of gene mutations that are associated with cancer risk — but other MSI-high tumors, like Ball’s, appear to form spontaneously. MSI-high status is sometimes also seen in glioblastoma brain tumors; lymphomas; and cancers of the urinary tract, ovaries or endometrium, according to a 2017 study published by Harvard researchers in the journal Nature Communications. This mutational status can also appear in biliary, bladder, breast, pancreatic, prostate, renal cell, sarcoma, small cell lung, small intestinal, thyroid and peritoneal cancers, according to Merck & Co., the maker of Keytruda.

Because MSI-high tumors cannot repair their DNA, they have about 1,700 mutations, on average, compared with 70 found in normal cancer cells. Scientists who helped develop the drug for MSI-high tumors believe it works because of the sheer quantity of mutations the irregularity causes. “The more mutations a tumor has, the more foreign it looks to the immune system,” explains Dung Thi Le, M.D., an associate professor of oncology at Johns Hopkins University School of Medicine in Baltimore and a clinical trial investigator who tested Keytruda in MSI-high cancers. Once the mutations flag MSI-high tumors as foreign invaders, the drug unveils the cancer to the immune system, which sets out to eliminate the disease.

Keytruda had previously been indicated for the treatment of several specific tumor types, including melanoma and non-small cell lung cancer. A few months after Keytruda was approved for all MSI-high cancers, another PD-1 inhibitor, Opdivo (nivolumab), won FDA approval in MSI-high, metastatic colorectal cancer that had progressed despite chemotherapy. Both drugs produced impressive results in late-stage clinical studies, with objective response rates of about one-third in patients with colorectal cancer. Side effects included fatigue, fever, diarrhea and itchy skin.

Ball, who expects to stay in the Keytruda trial until mid-2018, says she has experienced no side effects from the drug, which she gets by infusion every two weeks.

FINDING THE MISMATCH

She doesn’t require any other treatments to control her cancer. When she renewed her wedding vows with her husband on Aug. 10, as she does every year, she fit into her wedding dress, thanks to the shrunken tumor on her stomach. “I feel so blessed to have been given this opportunity,” says Ball, who lives near Ohio State and works as an insurance salesperson.

Two types of tests can identify MSI-high tumors. Oncologists can use tumor biopsies to try to uncover mismatch repair deficiency with a test that detects whether any of four key proteins are missing. They can also run a test that highlights DNA areas with repetitive sequences — a sign of MSI — which can be more accurate but is more complicated, requiring that tumor samples be compared to normal tissue, Le explains. Hence, she says, most cancer centers start with the protein testing. In fact, guidelines now call for testing of all newly diagnosed patients with colorectal cancer; if doctors don’t offer it, patients should ask for it.

With two approved drugs for MSI-high, metastatic colorectal cancer, the choice of therapy isn’t so clear-cut, says Overman, lead investigator in the Opdivo Checkmate-142 trial. “We have no real evidence to say which one is better,” he says, because no head-to-head trials have compared Keytruda with Opdivo. Still, he says, having two choices is good news for patients, many of whom have run out of options for slowing down the progression of their disease.

Chemotherapy and surgery were no longer helping Amy Scarbrough by the time she enrolled in an Opdivo trial at MD Anderson in January 2015. She received a diagnosis of stage 4 colorectal cancer at age 49 and later learned she had Lynch syndrome. After undergoing a hysterectomy and surgery to remove other cancerous growths, she still had about 100 metastases, she says.

AMY SCARBROUGH, patient treated with Opdivo for stage 4 colorectal cancer - PHOTO BY BETH HALL

Scarbrough traveled from her home in Fayetteville, Arkansas, to Houston every two weeks to receive infusions of Opdivo. By the time she stopped the treatments in November 2017, her scans showed just a few spots, which were determined to be scar tissue. The only side effects she experienced were autoimmune symptoms like shortness of breath, hives and joint pain, which were easily controlled with medication. “I did not mind at all,” Scarbrough says. “What a trade-off! It’s been worth it.”

A GROWING MENU OF TREATMENTS FOR MSI-HIGH TUMORS

The question of whether to ever take patients off immuneboosting drugs became complicated after the therapies were approved, Overman says, because there is scant data on long-term outcomes. “The question a lot of us have is, after two years of treatment, does the patient even have any cancer left? It appears to be eradicating the cancer.” Early results suggest that two years of treatment should be enough, he adds, but individual symptoms and disease status help determine the proper length of therapy for each patient.

Durability of positive responses to immunotherapy treatments is just one of many issues that may be clarified when multiple ongoing studies of the drugs generate results. For instance, despite rich data supporting the use of these drugs in MSI-high colorectal cancers, the benefits in other gastrointestinal cancers remain unclear. Keytruda, for example, was approved by the FDA in September 2017 to treat patients with gastric cancer whose disease had progressed after chemotherapy, based on a phase 2 study that looked at patients regardless of MSI status. In that trial, the objective response rate in patients with MSI-high tumors was 57.1 percent versus just 9 percent in those with non-MSI-high tumors. Between 15 percent and 25 percent of gastric cancers are MSI-high.

Two months later, data from a larger phase 3 study showed that, in patients with gastric cancer, the drug did not improve survival compared with the chemotherapy drug Taxol (paclitaxel). Some experts say these findings are not relevant when it comes to MSI-high gastric cancers, because the phase 3 trial was designed to measure drug response in patients whose tumors expressed high levels of PD-L1 — a different and unrelated biomarker. This study did not report whether MSI-high cases fared better with immunotherapy.

The high response rates to PD-1 inhibitors among patients with previously treated MSI-high colorectal cancer tumors has sparked interest in testing the drugs earlier in the treatment process. “There are trials now looking at the use of these immunotherapies in the first-line setting, and they will address the question of how these compare to chemotherapies,” says Neil Segal, M.D., Ph.D., a medical oncologist at Memorial Sloan Kettering Cancer Center in New York City.

One ongoing study is comparing Keytruda with standard firstline chemotherapy regimens in patients who have advanced MSI-high colorectal cancer. Several studies are also investigating the use of PD-1 inhibitors with other cancer treatments. For example, one trial is enrolling patients with MSI-high colorectal cancer who will receive Opdivo alone or a combination of Opdivo and Yervoy (ipilimumab), another immune-boosting treatment that targets a checkpoint called CTLA-4.

Initial results indicate that combining the two checkpoint inhibitors could be effective for a large proportion of patients with colorectal cancer who have MSI-high tumors.

In January 2018, investigators announced a 55 percent overall response rate among 119 patients who participated and received the combination versus a 31 percent response rate for patients who got Opdivo alone. The combination controlled the disease in 80 percent of patients, and 94 percent of the responders were still doing well when the trial ended.

Although some oncologists have expressed concerns about increased side effects with combination treatments, adding Yervoy to Opdivo just slightly raised the risk of side effects during the trial. Segal says more research needs to be done, but the potential for combinations of checkpoint inhibitors to provide hope to patients who have run out of options may offset the risk of side effects. “These are promising response rates,” he says. “We have seen increased side effects in other studies combining PD-1 and CTLA-4 inhibitors, but the decision of whether to proceed with the combination will depend on how much benefit there is likely to be.”

A separate arm of that trial is testing a treatment that starts with the targeted drug Darzalex (daratumumab), currently approved to treat multiple myeloma, followed by Opdivo. A different combination trial is testing Keytruda with the chemotherapy treatment Xeloda (capecitabine) plus radiation in patients with gastric cancer whose tumors test positive for both MSI-high and Epstein-Barr virus, a contributing factor in many cases of the disease.

Experimental PD-1 inhibitors also are being tested in patients with MSI-high tumors. One, called TSR-042, is in a trial that is recruiting patients with any type of MSIhigh solid tumor that hasn’t responded to conventional treatments.

A type of immunotherapy that inhibits a related checkpoint, PD-L1, is emerging as another potential treatment. The PD-L1 blocker Bavencio (avelumab) is being tested in patients with colorectal tumors that are MSIhigh or harbor a POLE mutation, which might make them equally receptive to checkpoint inhibition.

For patients like Scarbrough, the rapidly expanding understanding of how the genetic characteristics of tumors drive their responses to immunotherapy offers hope for cures that would have seemed impossible just a few years ago.

Scarbrough, a married mother of three, no longer requires treatment for her cancer but returns to MD Anderson for imaging scans every 12 weeks. She says she feels fortunate that she could participate in research that elucidated the link between MSI-high tumors and checkpoint inhibition. “If this diagnosis had happened four years earlier, the outcome would be very different,” Scarbrough says. “I’m told I have the potential to live a long life.”