I was recently reminded of the way we make progress in clinical research: with stories. We observe a clinical phenomenon and make up a hypothesis (story) as to why it happens—and we run with it, until proven wrong.
When I started as a breast surgeon, like everyone else, I believed the story we told about breast cancer starting in the breast, and the cancerous cells then breaking away from the tumor and moving one by one up the chain of lymph nodes and then out into the blood stream and other organs. This story was the basis of the recommendation for radical mastectomy. We believed removing the whole breast and all of the lymph nodes would slam the door and keep the cancer cells from going elsewhere in the body. This story was also the underpinning for screening and early detection. We believed if you caught a tumor when it was too small to get out, you might be able to prevent its spread.
We held on to this belief until studies showed that wide excision (lumpectomy) followed by radiation was as good as mastectomy and that while “early detection” works sometimes, it is not the panacea we had all hoped it would be. These findings made clear it was possible for cancer to head into the blood stream early on, skipping the nodes completely. That’s the problem with stories. Even if they sound good and make sense, we still need to prove them before we get too carried away.
Two recent articles that challenge current hypotheses or stories are yet another reminder of why we need to prove our tales. One of the stories that has taken hold is that we can test whether a new drug works by giving it to women with large tumors before surgery. Studies had shown that women whose tumors disappeared when treated before surgery with chemotherapy, a targeted therapy or hormone therapy (called neoadjuvant treatment) had a better outcome. So, researchers began testing new cancer drugs and new drug combinations by giving them to women before they had surgery. They believed that seeing the drug shrink the tumor was a surrogate for a traditional clinical trial that would follow the women for several years to see if it had improved overall survival. Unfortunately, it now appears that tumor shrinkage does not always correlate with overall survival.
We first saw this with the NeoALTTO trial which studied lapatinib (Tykerb) and trastuzumab (Herceptin). We learned that the drugs could shrink the tumors. But we also learned that how the tumor responded to the treatment had no impact on survival. This tells us that, undoubtedly, there is a difference between the cancer cells that thrive in the breast and those that are able to get out of the breast. This doesn’t mean that getting neoadjuvant treatment and having your tumor shrink is bad. It just means that your tumor’s response may not tell the whole story.
The other hypothesis that is in flux involves circulating tumor cells. New technology allows us to do a liquid biopsy (blood test) to find tumor cells or tumor DNA. The story we have been telling is that the cells and DNA were representative of the metastases and that we could use them to determine what treatment the tumor would respond to and when a treatment needed to be changed.
While this is a seductive story it turns out that there is a lot of heterogeneity in the circulating cells and it is not yet clear what they actually represent. Could the circulating tumor cells be the cells that couldn’t get into another organ and therefore remain circulating through the blood stream? Are they being shed by the metastases? Or is it the other way around? This new technology is exciting, but it is important that we test all of these hypotheses before we start relying on our findings. I am reminded of the era of when we believed in high-dose chemotherapy and stem cell transplant to treat breast cancer. We thought if we gave more chemotherapy we could cure people, and we told such a good story women fought their insurance companies to get this treatment. But when the randomized controlled study was done, it showed that it was not the amount of chemotherapy that mattered but rather that it matched the tumor.
All the advances we are making are encouraging. But we should not let our beliefs be the story. We need to continue to stick to the old slogan, “Trust but verify!”