When Jen Morey was diagnosed with colon cancer in June 2013, her oncologist began treating her with a chemotherapy usually prescribed for that type of cancer. But after a couple of months, the malignancy was still growing, and rapidly. The therapy was failing.
So her doctor ordered a test to identify aspects of the tumor’s genetic makeup that might be fueling its growth.
Morey’s oncologist was surprised when the tumor profile test, as the technology is called, showed that the cancerous cells in her colon had a genetic mutation found almost exclusively in breast cancer. So he started her on a drug used mostly for fighting breast cancer.
Tumor profile tests are considered experimental, so they are typically reserved for patients, such as Morey, who have advanced cancer and no treatment options that seem likely to be effective.
Although some studies have found the tests promising, researchers and organizations including the American Society of Clinical Oncology believe further investigation is needed to determine their effectiveness.
In one clinical trial, tumor profiling identified a variety of abnormal genes in 31 lung cancer patients who had no evidence of genetic alterations based on standard tests. A drug to target the abnormality was available for eight of these patients; six received the therapy, and all six improved. Improvements ranged from stabilization (the cancer stopped growing) to tumor shrinkage.
In another trial, treatments prompted by tumor profiling benefited 10 of 25 metastatic cancer patients who had run out of other options.
Figuring out why and how malignancies grow is complicated, because there may be hundreds of abnormally behaving genes within a single cancer cell though only a few of them may be fueling the cancer. Furthermore, a patient’s abnormal cells are unique to that person, explained Michael Pishvaian, Morey’s oncologist at Georgetown Lombardi Comprehensive Cancer Center.
“Determining what is driving a person’s cancer can be like looking for a needle in a haystack,” he said.
Morey’s lab work found that needle.
She stabilized quickly on the drug that targeted the “breast-cancer” gene mutation. The drug was easier to tolerate than many chemotherapies because it attacks only cancer cells, sparing healthy ones. And because it might boost the effectiveness of the colon-cancer drug she was already receiving, she remained on the original treatment along with the new drug.
“I was hopeful for the first time in a long time,” said Morey, 38, a critical-care nurse who lives in the Friendship section of Anne Arundel County.
A CT scan in December showed that the new effort to fight the aggressive HER2 gene seemed to be working. Morey remains stable, meaning the cancer is no longer growing. She returned to work in late January after having gone on leave earlier in her treatment regimen.
“I am in this very small percentage, so everything we do is out of the box. But I am hopeful that this combination will continue to work,” she said. “And if not, there are still other drugs to try.”
“Individualized medicine” has been a buzzword in the world of cancer care for years, but the concept has been gaining increased currency. President Obama endorsed it in his State of the Union address in January, referring to it as “precision medicine.” In February, Inova said it planned to build a major complex in Northern Virginia devoted to genomics and personalized medicine.
With the continued push to find the right drug for the right person, increasingly sophisticated diagnostics are evolving, such as the tumor profiling tests. Tumor profiling is done by photographing DNA that has been extracted from a piece of a biopsy. The data captured in the photos is used to detect abnormalities. The technology has sometimes detected more than 300 genetic mutations in a single tumor.
Another approach is chemo sensitivity testing, in which part of a tumor is placed in a petri dish or test tubes with different drugs or drug combinations. The goal is to see what might kill every malignant cell. Like tumor profiling, this approach is ordered for advanced-cancer patients after all treatment options have failed. Both test types are used in most solid-tumor cancers, including disease of the breast, ovaries, prostate and colon. And there are versions for blood cancers such as leukemia and lymphoma.
In one trial where drugs were selected through chemo sensitivity testing, the tumors of 20 of 31 people with lung cancer stopped growing, shrank or disappeared. This was slightly more than double the response that standard drugs would be expected to produce in similar patients.
In a study involving 30 women with metastatic breast cancer, chemo sensitivity testing identified a combination of drugs that shrank tumors by at least 30 percent in half of the patients. The length of time until the disease progressed more than doubled.
Victor Priego, a Bethesda oncologist, orders tumor profiling or chemo sensitivity testing for almost 20 percent of his hardest-to-treat patients.
“Chemo sensitivity tests have not hit mainstream oncology because they have not been validated through many rigorous clinical trials. But I use a lab in California that has been rather consistently on the mark when you would not expect the suggested drugs to work. And this is with most cancer types,” he said.
Alex Spira, a Fairfax oncologist and researcher, believes there is merit to tumor profiling, but he doubts the reliability of chemo sensitivity tests. “What happens to a tumor when it’s treated in a test tube doesn’t directly correlate with what happens in patients,” he said.
He is less skeptical about the reliability of a similar tool where biopsied portions of a tumor are implanted in mice, as research shows that a human malignancy appears to grow almost identically in a mouse as in a person.
“We can put a small amount of one tumor into 100 mice, to test many drugs and drug combinations,” said Ronnie Morris, founder of Champions, the Baltimore-based lab that developed the mouse avatar test. He believes the technology will accelerate scientific learning.
“Clinical trials are expensive and take years, and the tested drugs do not make it to market 92 percent of the time because they were not proven effective,” he said. “Meanwhile, a drug may be a good one, but in other populations or other combinations.”
In one small trial using the mouse avatar, treatments were identified for 22 of 29 patients. Six of the 22 died before the testing was completed, but of the remaining 16 people, 13 responded to the identified drugs.
There are drawbacks to the mouse avatar; it costs $10,000 to $12,000 and is not covered by insurance. Also, a tissue sample is not always obtainable because it is in an inaccessible location, which is also the case with in vitro chemo sensitivity tests. And while results of in vitro chemo sensitivity and tumor profiling tests are back in one to two weeks, results using mice take four months.
“This diagnostic would be for advanced cancer patients,” Spira said. “But they don’t have time to wait.”
Charges for in vitro chemo sensitivity and tumor profiling tests range from $4,000 to $7,000.They are often not covered by insurance.
And then there is the question of whether they really work.
The American Society of Clinical Oncology believes there is insufficient evidence to support use of chemo sensitivity tests in general oncology practice — i.e., they should be restricted to clinical trials. But the organization states, “Because the . . . strategy has potential importance, participation in clinical trials evaluating these technologies remains a priority.” ASCO is working on a position on tumor profiling.
Tumor profiling and chemo sensitivity tests do not require Food and Drug Administration approval.
“So we can’t say, based on our experience, whether [these particular diagnostics] can predict treatment well,” said Elizabeth Mansfield, deputy office director for personalized medicine in the Office of In Vitro Diagnostics and Radiological Health at the FDA.
Concerned about a growing number of products that have not received rigorous scrutiny, the FDA is working to take on review and approval oversight.
Five percent of Spira’s patients who underwent testing have benefited from treatments guided by tumor profiling findings, though he said outcomes were dramatic. He cited one lung cancer patient whose tumor disappeared five days after beginning a new treatment.
Pishvaian, Morey’s oncologist, and Priego, the oncologist in Bethesda, say a small number of their patients have responded well to drugs selected using tumor profiling.
Cancer death rates in the United States have slowly dropped. In 1950, the age-adjusted death rate for all cancers was 193.9 per 100,000 people, according to the National Center for Health Statistics. In 2013, the rate was 163.2, a decline of 16 percent in 63 years.
Five-year survival rates are low, however, for metastatic cancers. Five-year survival for Stage 4 lung cancer is 1 percent, according to the American Cancer Society. Stage 4 colorectal cancer has an 11 percent survival rate, and Stage 4 breast cancer a 22 percent five-year survival rate.
Cancer researchers and oncologists such as Spira believe that one focus of the past 20 years — to study the genetic makeup and behavior of tumors — will lead to better outcomes.
Spira cites gefitinib, discovered through tumor profiling to target a mutation in lung cancer. (A similar drug, Tarceva, is now used by many people with advanced lung cancer.)
“For patients with this particular lung-cancer mutation, this discovery is profound. It is, to date, the biggest breakthrough in lung-cancer treatments,” he said. “It is a constant battle to find and target genes that are causing the problem. But the way to get closer to better outcomes is to concentrate on genetics.”
“For now, with what we are learning, at least we have options to try when there were no others,” he said. “In time, we will find more Tarceva-like drugs, and for cancers beyond lung cancer.”
Karidis is a freelance writer.