People diagnosed with cancer are often thrown into an unfamiliar world where a complex medical language is spoken. One of the first concepts patients must understand is that of "response" to cancer treatment. Response to treatment is not generally measured in term of the elimination of the cancer, but often according to other markers such as tumor shrinkage. The data used to assess response varies depending up on the type and stage of the cancer.
An understanding of a treatment's true benefits is vital when it comes to making treatment decisions. For example, some women may be willing to do chemo for a 1 percent benefit—defined as 1 percent increased chance of living without disease or recurrence—while others feel the side effects outweigh the small benefit. Below, Dr. Ronald Blum, director of the Cancer Centers at Beth Israel and St. Luke's-Roosevelt Hospital in New York City, reviews some of the ways oncologists measure treatment response.
How do oncologists determine whether someone is responding to cancer treatment?
It's a complicated and challenging science to try and objectify data from clinical studies. We now use various qualitative and quantitative measures, such as tumor shrinkage, to ascertain treatment benefit. And response is measured differently in solid tumors than it is in hematologic, or blood, malignancies such as leukemia.
How is response measured in patients who have solid tumors such as breast cancer?
For patients with existing disease, often those with metastatic cancer, what is evaluated is so-called objective response. There are now internationally recognized criteria called the RECIST (Response Evaluation Criteria in Solid Tumors) criteria. According to the RECIST guidelines, a partial response, for example, is defined as a 50 percent decrease in the perpendicular diameter of a tumor mass. Often the tumors are at least round-like; they can be measured in one dimension on a CAT scan, for example.
We've learned that a partial response correlates with improvement in quality of life for the patient, such as improvements in pain, performance status, weight gain or improved lung function, if it happens to be a lung cancer, or improved intestinal function, if someone has intestinal metastases.
Then there are examples of complete responses, and those can either be clinical or pathological. If you do a CAT scan or MRI and you don't see cancer, that's a clinical response. A pathologic complete response is when, during surgery, no cancer is found.
Is there a relationship between tumor shrinkage and survival?
It's related, but whether or not it is causally related is not known. So responders live longer than non-responders, but that may not be because of the response. It may be something about the biology of the tumor.
Is quality of life used to measure response?
For certain cancers, and the best example of this is pancreatic cancer, it's very hard to measure the tumor objectively because the pancreas is a sort of amorphous organ that doesn't have clear boundaries.
Instead, other surrogates of benefit have been used. One of the drugs that is used for the treatment of pancreatic cancer, called gemcitabine, was approved by the US Food and Drug Administration based on quality-of-life parameters, rather than objective response. Benefit was measured in terms of increase in weight, pain, less consumption of analgesics—which treat pain—and longer survival.
How do you measure response in someone who has been treated and no longer has a tumor?
We use different terminologies for people without existing disease. The benefit parameter that we use is not whether or not the tumor shrinks, because it's gone. Instead, we use measures like disease-free survival, overall survival, and progression-free survival.
Disease-free survival means someone's alive without disease. Overall survival means they have had a recurrence, but are living with their disease. Progression-free means that their tumor isn't growing. If someone's tumor is minimally symptomatic and is not growing, they can live with that cancer for a long time.
How is response defined in blood cancers such as leukemia?
For leukemia, the word used is remission, as opposed to tumor response. For example, the drug Gleevec is effective in the treatment of chronic myelogenous leukemia (CML). Someone with CML can have a partial remission, which means that the bone marrow is cleared of a given percentage of tumor cells.
With Gleevec and CML, the landmark study showed that you can see complete response, meaning you can look at the bone marrow and not see any leukemic cells. But what's more important is that you can actually take a look at the cells and determine their genetics, or so-called cytogenetics. A complete cytogenetic response means that instead of having 10 percent of cells with the abnormal chromosome that causes CML, the Philadelphia chromosome, it's totally gone. We think a cytogenetic response is biologically very important because the cells that are responsible for this leukemia are gone. Although it's too early to say long-term, this is a benefit that could translate into cure.
Are "benefit" and "response" usually clearly explained to patients?
It's important for physicians and for patients to come to some understanding of the definition of benefit or response for the particular scenario. Benefit could be different for someone who's got very advanced disease—where the benefit might improvement in symptoms—than for someone where the benefit might be curative. The patient really has to be comfortable enough to ask, "What do you mean by 'benefit'?" I think this should be a dialogue and informed discussion.
I think that more and more physicians are comfortable discussing the specific definition of "benefit" and "response," and that patients are more comfortable asking the questions. That needs to continue, because it's important in making a medical decision. Patients have to know the benefit of a treatment, relative to the side effects, so they can make the best possible treatment decisions.
