An Embarrassment of Riches in Oncology With Wayward Decision Support: A Recipe for Disaster?

The decision support that is currently available to community oncologists has made it tremendously difficult to keep pace with what many consider to be monumental advances in the understanding and management of breast and lung cancers, according to Maurie Markman, MD. He added that the momentum driving drug discovery is unlikely to ease up, leaving providers ill-equipped to deliver optimal care.

“I wanted to discuss the critical and very difficult topic of decision support in oncology. I cannot emphasize simply how important it is, and, quite frankly, how little attention has been given to it,” Maurie Markman, MD, editor-in-chief of OncologyLive®, said.

In this special with OncLive, Markman, who is president of medicine & science at City of Hope Atlanta, Chicago, and Phoenix, discussed the importance of decision support in oncology, painting a vivid picture of the knowledge base required just 40 years ago and how it pales in comparison with what is required for optimal disease management today.

Breast and Lung Cancer Management From the Perspective of Medical Oncologists 40 Years Ago

Markman: When I started as a fellow and as a junior faculty member approximately 40 years ago, [cancer care was much different]. Many of you probably don’t remember or even think about where we have been, but the difference is so striking and so relevant. Just think about how fast we’re moving now compared with what I’m describing.

When I was a fellow at the Johns Hopkins Oncology Center, [now the Sidney Kimmel Comprehensive Cancer Center] and the National Cancer Institute, where I did my training, in both institutions if you had a patient with breast cancer, the first therapy was surgery, which was followed by radiation, and maybe adjuvant chemotherapy. If a patient developed metastatic disease, you gave chemotherapy and maybe hormone therapy, and that was it. I cannot emphasize how simple the paradigm was. For patients, this was not necessarily a good situation, but that’s where we were. If a patient was estrogen or progesterone receptor positive, we’d give them hormonal therapy that could have been a surgical manipulation, or it could have been tamoxifen at that point in time, and that’s what we had. We did give adjuvant chemotherapy so that could have been cyclophosphamide, methotrexate, and fluorouracil or a doxorubicin [Adriamycin]-based regimen. Again, that was it.

In the metastatic setting, we had the same drugs. There was very little focus on the questions surrounding quality of life, [although] toxicity was an issue. The immediate concerns [were] nausea, vomiting, and lowering the blood counts. This was before the days of growth factors. There was no such thing as molecular stratification of disease. We did not use combination therapies. [Treatment] was [given] sequentially, certainly when it came to the question of hormonal therapy and chemotherapy. There were very limited second-line options. You might try it, but you knew it didn’t work very well. Or a patient could go on a clinical trial, [but] that wasn’t of that much value. There was no such thing as neoadjuvant chemotherapy. The idea that you [would] give chemotherapy first followed by surgery didn’t happen. There was no such thing as tumor-agnostic drugs—and very importantly, the length of survival was relatively short, and the concept of thinking of this illness as more of a chronic illness would not have crossed anybody’s [mind]. If [the disease] was metastatic, you’d give hormone therapy [or] you might give chemotherapy for a short period, and that was it. [After] adjuvant therapy, you were done.

Lung cancer [management] was even simpler, but it was much worse for those patients. We divided patients [based on whether they had] small cell lung cancer [SCLC] or non–small cell lung cancer, which we still do today. But the issue there was, why did you care for SCLC? You wouldn’t operate, you’d give chemotherapy, because all patients had metastatic disease, so surgery was not an option. In NSCLC, chemotherapy didn’t work very well, so in a small percentage of patients, you might operate, you might give radiation, and then you’d potentially give chemotherapy for a very short period. That’s all you had to know.

Chemotherapy was very ineffective; it was a little more effective in SCLC [where you’d] get a high response rate lasting for a number of months, and then [proceed to] palliative care. For NSCLC, you didn’t worry about anything other than the fact that it wasn’t SCLC. That was the management paradigm. Certainly, there was no such thing as adjuvant chemotherapy that was of any value. That’s where we were. There was no such thing as molecular classification or liquid tumor biopsies––that was the management. If you were a clinical oncologist in that era, you could take care of [patients with] lung cancer and breast cancer [because] you had the same chemotherapy drugs, which were limited. You had to know how to use hormonal therapy, but that was [all that was required for disease] management. That was the management of two of the most common cancers as recently as 40 years ago. Decision support wasn’t really that critical, and that’s where we were.

Dramatic Changes in Disease Management

The key is just how much things have changed in those two common illnesses. In breast cancer today, one would not think about management without looking at molecular characterization. We have a variety of algorithms for use of hormonal therapy or chemotherapy in the adjuvant setting. Also, hormonal therapy is not just hormone therapy now. We use other drugs along with hormone therapy in multiple approaches [in the] first- and second-line settings. Molecular characterization is critical as we move forward.

Neoadjuvant chemotherapy has become a standard, [alongside] radiation and surgery. The management of patients has now become not a question of simply [how to] sequence [approaches, as in the past where] you would [perform] surgery for breast cancer, then give radiation, and perhaps chemotherapy and hormone therapy. Therapies are now considered multimodality from diagnosis, which adds an enormous amount of complexity, thought processes, and new information. From the medical oncologist’s perspective, you’re taking care of patients with breast cancer, so you have to know all this. There are now tumor-agnostic drug indications that also need to be considered.

Now, we turn to lung cancer. SCLC has not changed as dramatically as NSCLC, but again, today, for someone with NSCLC, you would not think about therapy unless you have done molecular [geno]typing to look for EGFR mutations, ALK abnormalities, and ROS1 alterations, etc. The standard of care dramatically differ[s] depending upon the profile, and [you have to] follow the patient because resistance occurs. Mechanisms of resistance are also different. EGFR mutations can be very specific, and we have very specific drugs for [different patient subsets] that are FDA approved.

If you’re taking care of someone with lung cancer optimally, you have to know all this. You have to understand it. You have to know about the most recent advances, the new drugs, and the most recent changes and indications. I haven’t even mentioned immunotherapy yet, but then we add that as well and ask, ‘What is the role of immunotherapy in the frontline and second line with chemotherapy, with targeted therapy? When do you use it? What about the different drugs?’ In the future, we’re going to be seeing the use of a variety of liquid tumor biopsies, which you will be able to [use to] monitor patients on therapy. That’s a different paradigm than breast cancer.

Check back for parts 2 and 3 in which Dr Markman will discuss the challenges behind fixing what’s broken, how clinical trials are contributing to the problem, and how the introduction of immunotherapy has added an additional layer of complexity to the paradigm.