Re-Defining Cancer: Regulations Support Biomarker-Defined Indications, But Will Science—and Advocates?

FDA is ready to re-imagine cancer defined by molecular biology rather than “body parts.” But it is going to be a “long transition” that will have important ripple effects on the structure of cancer treatment—and advocacy.

“It is going to happen,” says FDA’s top cancer drug review manager, Richard Pazdur.

“It” refers to the approval of new drugs to treat tumors where the indication is not based on the site of tumor origin (i.e., breast cancer, lung cancer, colorectal cancer, etc.) but rather on the molecular target of the therapy (i.e., ALK-positive, EGFR-negative, HER2 over-expressing, etc.).

Pazdur, head of the Hematology & Oncology Drug Products Office in the Food & Drug Administration, jumped up to respond to a question about re-defining cancer during a conference on companion diagnostics sponsored by Friends of Cancer Research and Alexandria Real Estate Equities Inc. on September 14.

“Is all of this leading us to a place where instead of defining cancer by body part, we define it by molecular type?” the questioner asked. Such a change, she suggested, would have significant “implications for labeling.”

“I don’t see a problem with this whatsoever,” Pazdur said. There is no “regulatory dilemma,” since there is nothing in regulation that says cancer is defined by site of origin.

“I could see very easily that an indication be for tumors that over-express biomarker Y irrespective of conventional histological subtype,” Pazdur said. “It is there. It is going to happen. It is obvious that it is going to happen.”

“It really doesn’t produce a regulatory problem because we are not confined by this whatsoever in the regulations,” Pazdur stressed. “The science is going that way.”

“The major issue is going to be one of developing the drugs in the context of existing therapies. There is going to be a transition period here.” (Also see "Targeted Cancer Therapies in Context: FDA’s Pazdur on Companion Diagnostics" - Pink Sheet, 26 Sep, 2012.)

Center for Drug Evaluation & Research Director Janet Woodcock expressed agreement with Pazdur but also injected a note of caution. The traditional definition of tumor types based on site of origin, she argues, probably won’t ever disappear as at least one element in the cancer diagnosis.

“I think from what has been seen so far is that histology may actually matter,” Woodcock site. The site of tumor origin, she points out, means something.

While it may no longer make sense to define cancer solely in tumors of histology, “it is just one of the multivariate factors that goes into the choice of therapy.”

So, Woodcock suggested, “there is going to be a long transition.”

Re-Thinking Advocacy

A “long transition” may be in everyone’s best interests, since the labeling implications of re-defining cancer may be one of the least difficult challenges.

In addition to the scientific issues involved in redefining cancer, though, there are important structural questions – not least of which is what happens to advocacy organizations whose identity is grounded in traditional definitions of cancer biology. Not only will science and medical practice have to realign and redefine specialties, patients, advocacy groups and political movements will need to change also.

Woodcock noted that this is hardly an issue that is confined to oncology. “We are not just facing this in cancer,” she observed. “We will approve diagnostic-directed subsets in [many] diseases…if they are clinically meaningful.”

And, she added, “it is difficult for some groups to transition from their traditional disease definition even though it has been observational.”

The issues, though, are especially pronounced in cancer, where advocacy and practice are so closely aligned around the historical, organ-specific diagnosis. The Friends of Cancer Research program, for example, included Nancy Roach of Fight Colorectal Cancer. Attendees who asked questions or commented on companion diagnostic issues included representatives of the Leukemia & Lymphoma Society and the Multiple Myeloma Foundation.

Or, consider the implications of a new publication in Nature, which categorizes breast cancer tumors into four major subtypes based on genetic analyses. The study is yet another landmark in what increasingly feels like the inevitable redefinition of cancer from historic diagnoses based on where a tumor originates into a disease defined more by the specific molecular biology of the tumor type.

Indeed, as the Nature article points out, one of the four major subtypes of breast cancer appears to have characteristics that suggest it will be susceptible to therapy with drugs currently used in ovarian cancer.

That is the kind of therapeutic insight arising from molecular biology that the cancer community seems very excited about.

But it also raises questions about public health priorities. “Breast cancer” advocacy encompasses several large, well-funded national organizations. Ovarian cancer has committed advocacy organizations but much less prominence. The reasons are many, but include the fact that breast cancer is 10 times more prevalent than ovarian cancer—and the reality that because breast cancer has better survival rates the “survivor” population is even larger still.

But what changes if, at some point in the future, “breast cancer” is no longer “breast cancer,” but instead a series of mysterious designations like “BRAC1 inactivations,” or “RB1 loss/cyclin E1 amplification” tumors?

Or consider a bill currently pending in Congress, the “Pancreatic Cancer Research & Education Act” (HR 733) which—as initially drafted—would have directed the National Cancer Institute to set up a program to encourage development of new treatments for pancreatic cancer.

However, the legislation was amended to win support for passage in the House, with a key change being to eliminate the disease-specific focus. So instead of a “Pancreatic Cancer” initiative, NCI is directed to set one up for “recalcitrant cancers,” defined as those with a five-year survival rate below 20% and a mortality rate of at least 30,000 per year. That definition encompasses only two diseases: pancreatic cancer and lung cancer, advocates say.

However, if cancers are redefined by molecular pathway, it is possible that no cancer would meet that definition – or, perhaps more likely, that a specific tumor mechanism that cuts across multiple historic diagnoses fits the bill, which may or may not mean any additional research into patients who today are diagnosed as “pancreatic cancer” patients.

The understandable desire by advocates to garner resources and attention for “their” cancer is key component of the current public policy dynamics in oncology. It is also a factor that needs to be considered as part of the “long transition” to a new way of defining disease.