Deep Dive

Outsmarting Colon Cancer

Promising new therapies for the second most deadly cancer are teaching our bodies how to fight off disease without the toxic side effects.

By Jeanne Erdmann

Illustration by Mike McQuaide

Jose Landa was looking for a change. He and his wife came to the U.S. from their home in Venezuela in 2001 because they wanted a safe place to raise their children. An attorney by training, Landa had spent more than 20 years practicing business law, but he and his wife, Jacqueline, decided to do something they found more rewarding. In 2003 they started a small blackberry farm outside Spartanburg, South Carolina, called Double J Farms (named for Jose and Jacqueline). They own 75 acres, and today Double J is one of the largest blackberry farms in South Carolina.


But in 2012 a persistent pain in Landa’s left side threatened his pastoral endeavor. For months he waited for the right diagnosis. Eventually, the pain grew so intolerable that Landa could hardly bear it. Doctors at a hospital near his home in Spartanburg tested Landa for heart conditions, liver problems, and “everything except for the colon,” he says. Finally Landa told his family doctor that the pain had become unbearable. The doctor saw an obstruction on an X-ray and referred him to a gastrointestinal specialist who immediately suspected colon cancer.

In January 2013, after a colonoscopy, Landa received a diagnosis of stage 4 colon cancer. He was 45 years old. He first had surgery to remove the tumor, part of his colon, and the surrounding lymph nodes. Then he started FOLFOX (folinic acid, fluorouracil, and oxaliplatin), the standard chemotherapy for colorectal cancer. The side effects were awful. His eyes became inflamed, blurring his vision. He also had nausea, a bad taste in his mouth, and neuropathy.

“You name it, I had it,” he says. The neuropathy was the worst. Eventually the pain and numbness in his hands and feet were so bad that he couldn’t walk or type on a keyboard. But Landa pressed on. “My doctor said the only chance I had was to keep going. I chose to fight the cancer,” he recalls. “It was really horrible, but I thought, I will get over it.”

One year later, in 2014, the pain returned with a vengeance to the original spot in his side. Every bowel movement was agony. Emergency surgery showed the cancer had returned near where it started. Genetic tests on Landa’s tumors revealed that he had Lynch syndrome. Also around this time, Landa’s sister was being treated for colorectal cancer; she also tested positive for Lynch mutations.

Landa faced a second round of chemotherapy that included FOLFIRI (folinic acid, fluorouracil, and irinotecan hydrochloride), but the treatment didn’t work. A local doctor recommended radiation, but Landa sought a second opinion at MD Anderson Cancer Center in Houston. Doctors there said that radiation would be painful and risky. He chose not to have the radiation, reasoning he would rather die than burden his family.

A short time later, a doctor from MD Anderson called and offered Landa the last spot in an immunotherapy clinical trial; he took it. From May to December 2015, Landa flew from South Carolina to Houston every three weeks for the combination treatment of Opdivo (nivolumab) and Yervoy (ipilimumab). The side effects weren’t too bad — just a lot of itching, which Landa addressed with creams. Eventually he asked the doctors to transfer his treatment from MD Anderson in Houston to Duke University in North Carolina, because the airfare and two-night hotel stays were taking a financial toll. By then, Landa was just taking Opdivo, a drug he continues on today.

Family Affair

Colorectal cancer (CRC) is the third most common cancer affecting both men and women in the U.S., and the second most deadly. CRC is expected to claim more than 50,000 people in the U.S. in 2017, while more than 135,000 new cases are expected this year. But thanks to better screening and more effective treatments, over the past two decades more people with CRC are living longer.

CRCs start in the epithelial cells that line the mucosa, the innermost layer of the colon. Cells can divide too much and form polyps. Although not all polyps turn malignant, some can evolve into cancer that can eventually spread if not removed. In 1990, Bert Vogelstein, co-director of the Ludwig Center at the Johns Hopkins Kimmel Cancer Center, showed that colorectal cancer starts with mutations in a “tumor suppressor gene” called APC, which helps prevent cells from growing out of control. This process can take about 10 years.

Some 60 to 80 percent of CRCs are caused by mutations people acquire throughout their lifetimes. This type of CRC is called “sporadic” because the mutations likely occur by chance and are not inherited. About 5 percent of CRCs are caused by mutations that we inherit from our parents. In hereditary colon cancers, these germ-line mutations carry a 40 to 80 percent lifetime risk of the disease. A strong family history puts someone at higher risk of CRC even if he or she doesn’t carry one of the mutations.

These cancers respond better than any other tumor type, and these are patients in late-stage colorectal cancer. This is not just immunotherapy; it’s really personalized immunotherapy.

The most common inherited types of CRC are Lynch syndrome (also known as hereditary nonpolyposis colorectal cancer), which Landa and his sister have, and familial adenomatous polyposis (FAP). People with FAP have hundreds, even thousands, of polyps in their rectum or colon that start forming in their early teens. People with Lynch syndrome don’t form as many polyps, but they do carry a higher lifetime risk of other cancers, including endometrial, brain, and pancreatic. Other hereditary CRCs, such as MYH-associated polyposis (MAP) and attenuated familial adenomatous polyposis, are rare. MAP patients can also have many hundreds of precancerous polyps.

If a doctor suspects one of these inherited syndromes, he or she may send the patient to a genetic counselor who can help manage risk. “When you have a genetic syndrome, one of the features is that you present with early onset disease,” says Sapna Syngal, director of the GI Cancer Genetics and Prevention Program at Dana-Farber Cancer Institute and a professor of medicine at Harvard Medical School.

Fine-Tuning Treatment

For those diagnosed with CRC, doctors can test colorectal tissue for biomarkers that help guide treatment. The most common are the KRAS and NRAS genes. These genes spur CRC cells to grow and divide. If a test reveals these mutations, the tumor will not respond to certain therapies. Another biomarker is a gene called BRAF, which also tells cancer cells to divide. In CRC, mutations in BRAF point to lower chances of survival.

Michael Morse, a medical oncologist at Duke University, has been trying to find ways to harness the immune system against CRC for the past 20 years. Although the five-year survival for CRC has increased from around 3 percent to 11 percent, it’s difficult to cure all forms of colorectal cancer, both sporadic and familial, that have already spread to other organs. Although the trend is heading in the right direction, Morse calls the number of people cured “still too small.”

Currently there are seven types of immunotherapies in clinical trials, ranging from monoclonal antibodies that act like smart missiles aiming treatment right at cancer cells, to adoptive cell transfer, in which cancer-killing immune cells are multiplied in the lab and transferred back to patients. Steven Rosenberg, chief of the surgery branch at the National Cancer Institute’s Center for Cancer Research, used adoptive cell therapy to treat a patient with metastatic colon cancer with a mutation in KRAS and whose cancer had spread to the lungs. The team isolated specialized immune cells (T cells) from the tumor that can infiltrate and attack the tumor cells with the mutated KRAS gene, grew them in the lab to expand their numbers, and then put them back in the patient. The lung tumors regressed for nine months.

“The challenge here is you have to get T cells from the patient’s own tumor, so it’s very personalized in the sense that you don’t have a drug off the shelf. But it can be done and is potentially applicable to many different cancer types,” says Vogelstein. “It is promising and certainly worthy of future development.”

One of the most promising areas, says Morse, is a marker called Microsatellite Instability High (MSI-H). The marker is found in about 15 percent of CRC tumors and can signal whether a person might have Lynch syndrome. “If MSI is detected in the tumor, then additional testing is done to determine if a person may have Lynch syndrome,” explains Morse.

Vogelstein’s lab discovered that people with MSI-H colorectal cancer would respond to PD-1 checkpoint inhibitors, like Opdivo, which uncloak the immune proteins that allow cancer to hide from killer immune cells. The Vogelstein team figured out that only MSI-H colorectal tumors respond to PD-1 because MSI-H tumors have so many more mutations than sporadic CRC. “These cancers respond better than any other tumor type, and these are patients in late-stage colorectal cancer,” says Vogelstein. “This is not just immunotherapy; it’s really personalized immunotherapy.”

Researchers are trying to trick the MSI-stable colorectal tumors that don’t have as many mutations into responding to checkpoint inhibitors, so that the treatment can help more people. “It’s great if you are one of the 2 to 4 percent who respond, but it’s not great if you are one of the 98 percent who are not responding,” says Vogelstein.

I’m optimistic we will find better ways to harness the immune response, and in those tumors that don’t have an immune response, to create one. I think hopefully the days of the toxins — the poisons — will be limited.

In May, the FDA approved another PD-1 inhibitor, Keytruda (pembrolizumab), for any metastatic solid tumor, including colon cancer, that has the MSI-H biomarker. This is the first time the FDA has approved a drug based on a shared biomarker, regardless of where in the body the tumor originated.

In the future, Morse hopes that doctors can sample tumor DNA in real time, determine whether a treatment is working, and extend lives even longer. The median survival used to be 12 to 14 months; now it is 36 to 40 months. So for now, these treatments are turning CRC into more of a chronic disease rather than a curable one, but the treatments themselves are also helping researchers understand more about the biology of this cancer, says Morse. “I’m optimistic we will find better ways to harness the immune response, and in those tumors that don’t have an immune response, to create one. I think hopefully the days of the toxins — the poisons — will be limited. They are still with us for now, but hopefully we will be able to eliminate them in the future.”

Path to Recovery

Landa was lucky. His tumor had the MSI-H marker and responded to a combination of the PD-1 checkpoint inhibitor Opdivo and Yervoy, a monoclonal antibody. Today Landa is still on Opdivo. Every few weeks he drives three hours to Duke to see his oncologist and have blood tests. He needs high doses of prednisone every time he has a CT scan because he’s allergic to the intravenous dye. The drugs have increased his appetite so he’s now trying to lose the weight he’s gained.

His recovery has been slow but constant. His cancer has been in remission since April 2016 and his life is gradually getting back to normal. He misses riding horses — something he’s done since childhood — but he still has limited feeling in his feet and the idea of being in a saddle makes him nervous. His illness also took a toll on the blackberry bushes — they suffered from inattention, becoming sick with bacterial infections. With his strength returning he has been able to nurse the shrubs back to health. Landa says he and the blackberries are walking the same path to recovery.

“In my opinion, [after] seeing people die from cancer, what happened with me is awesome. It is amazing, because I was able to get treatment in a clinical trial without pain and suffering.”