Knowing the cellular basics drives treatment decisions.
Researchers measure certain proteins and genetic factors, together called biomarkers, to better understand a tumor. Biomarkers can provide prognostic or predictive information, and sometimes both. Prognostic biomarkers provide information about the natural progression of the disease, such as the risk of recurrence or metastases, whereas predictive biomarkers offer insight into how a tumor would respond to a certain treatment.
The most well-known biomarkers in breast cancer are the estrogen receptor and HER2 protein, which dictate the type of treatment a patient receives, such as hormonal therapy or Herceptin (trastuzumab).
In addition to HER2 and hormone receptor status, physicians are increasingly looking to other gene profiles, including those tested by the Oncotype DX and Mammaprint assays, which are being used in the TAILORx and I-SPY 2 studies, respectively.
Oncotype DX, used mostly in the U.S., looks at 21 genes to help predict the risk of recurrence for early-stage, estrogen receptor-positive breast cancer, as well as if chemotherapy will help reduce that risk. Mammaprint, a 70-gene profile, looks at the risk of recurrence in node-negative breast cancer. However, because this assay requires fresh tissue, physicians must plan for the test before surgery, which has limited its use in the U.S. “Right now, we’re only using these assays to make a go or no-go decision regarding whether to recommend chemo,” says Joseph Sparano, MD, professor of medicine and women’s health at Albert Einstein Cancer Center-Montefiore Medical Center, but future research may hold the answer to many other treatment questions. “We hope to have an assay in five to 10 years that can identify which patients can be cured by just surgery and don’t need any additional therapy.”
The I-SPY 2 and TAILORx studies are first steps toward that reality by making use of the two assays to help determine treatment in a subset of breast cancer patients, and also by changing how clinical trials are designed, which may help all cancer patients.
I-SPY 2 (Investigation of Serial Studies to Predict Your Therapeutic Response with Imaging and Molecular Analysis 2) is unique because it’s testing several therapies—up to five experimental treatments—at a time. Twenty percent of patients are randomly selected to receive the standard of care, treatment they would receive outside a clinical trial. The other patients get standard care plus an experimental drug based on their tumor’s genetic profile.
Jane Perlmutter, PhD, lead advocate on the trial, says they will test eight to 12 drugs during the trial by beginning with five initial investigational drugs including veliparib (a PARP inhibitor) and neratinib (HER2 inhibitor) with more being added as the trial progresses.
“It’s not only finding cures, it’s finding out which drugs aren’t going to help, and those things go hand and hand.”
Its adaptive study design also means researchers will be able to learn more, faster, Perlmutter says. The standard protocol in treating locally advanced breast cancer is to remove the tumor surgically and then follow up with systemic therapy to kill any remaining cancer. It may take years to know whether the treatment worked, as recurrence can take up to 10 or 15 years to become apparent.
I-SPY 2 researchers, on the other hand, can identify benefit in months by giving patients drugs before surgery and then taking scans and biopsy samples throughout the trial. The biopsy samples are tested with Mammaprint and other assays to look for tumor biomarkers, such as those that indicate recurrence and those that predict a better response to specific drugs.
The thinking is that a change in the size of the tumor or a biomarker during the weeks of therapy will show researchers if the drug is working against the tumor. If it's not working, they change drugs. After the course of therapy, the patient undergoes surgery to remove any remaining cancer.
When a drug "graduates" with positive marks, information about the biomarker profile and the patients who responded to the drug—and those who didn't—may be enough evidence to test it further. If it doesn't look like the drug works at all, it's dropped and replaced with another one.
"While we will not be able to get a definitive answer that will allow drugs to be used outside of trials, we'll have a lot of evidence on certain kinds of drugs and biomarkers," Perlmutter says. That evidence can be a springboard for future trials that produce real results. "It's not only finding cures, it's finding out which drugs aren't going to help, and those things go hand and hand."
Researchers hope that I-SPY 2 signifies a less expensive, quicker and more productive way to test cancer drugs. It could also encourage collaboration among research institutions and pharmaceutical companies—as shown by the number of different drugs being tested in one trial.
"Our hope is that this trial not only introduces new treatments for breast cancer patients but will really push the envelope in how clinical trials are conducted—including the use of biomarkers—so we can start making faster progress toward personalized medicine," says Perlmutter.
Researchers hope that another trial, TAILORx (Trial Assigning Individualized Options for Treatment), will answer the question of whether chemotherapy is needed for thousands of women diagnosed with breast cancer each year.
Although Oncotype DX can help determine risk of recurrence and response to chemotherapy, not every breast cancer patient needs it. It’s only useful in certain women who have estrogen receptor-positive breast cancer with little or no lymph node involvement. And, it’s typically only used when it’s not known through clinical or pathological features whether chemotherapy is warranted after the initial surgery and endocrine therapy.
If the trial shows some benefit [from chemotherapy], it will be very modest, and the next challenge would be to identify which patients in that group are really benefiting.
TAILORx is expected to release data in about five years and may answer that question by randomizing those with an intermediate score to receive chemotherapy or not, in addition to hormonal therapy. If the trial shows chemotherapy doesn’t offer any benefit to women who have a mid-range recurrence score, it could save as many as 20,000 women each year from unnecessary treatment, says Sparano, lead investigator of TAILORx.
“If the trial shows some benefit [from chemotherapy], it will be very modest, and the next challenge would be to identify which patients in that group are really benefiting,” he says.
Research already hints at that, namely expression of a gene called topoisomerase 2 (TOP2A). A study published in Clinical Cancer Research in 2009 by Sparano and colleagues has suggested that patients with elevated TOP2A expression had double the risk of recurrence and would probably benefit from chemotherapy. “But we haven’t validated it,” Sparano cautions, “so it’s not something we can use right now.”
The second objective of TAILORx will help future molecular profiling studies. Every patient enrolled in TAILORx had tumor and blood specimens banked for future studies. This allows researchers to test other molecular profiles in the future without having to repeat a lengthy and expensive randomized trial.
In addition, another trial, MINDACT, is taking a different approach by comparing whether clinical features or molecular features more accurately predict benefit from chemotherapy. It is expected to complete enrollment in the next year.
These tests aren’t necessary for everyone. “They’re only useful in circumstances when there is uncertainty about what therapeutic decision to make,” says Sparano.
Currently, Oncotype DX and Mammaprint only focus on determining if a subset of breast cancer patients would benefit from chemotherapy. But the promise of molecular profiling could be realized for many more patients very soon.
“In the future we’ll see assays that may predict what patients benefit from Herceptin, endocrine therapy and, as other new therapies come out, who benefits from those therapies,” Sparano says.