Why Do Cancer Surgeons Cure More Patients Than Medical Oncologists?
Robert A. Nagourney, M.D.
Who does it really serve when a clinical trial involving 597 patients results in an increase in survival by 10 days?
And when the authors of the published trial results are quoted as saying, “To our knowledge, this randomized phase III trial is the first to demonstrate statistically significantly improved survival in advanced pancreatic cancer by adding any agent to gemcitabine.”
Is this the best we can do?
Surgery remains the most curative form of cancer treatment. While the reasons for this are many, the most obvious being earlier stage of disease and the better performance status of the patients, there are other factors at work. Surgeons tend to be rugged individualists, prepared to make life and death decisions at a moment’s notice. The surgeon who enters the pelvis expecting an ovarian cyst only to find disseminated ovarian cancer must be prepared to conduct a total hysterectomy and bilateral ovary removal if he/she is to save the patient’s life. It is these types of aggressive interventions that have that revolutionized the treatment of advanced ovarian cancer.
What of the medical oncologists who, with the exception of leukemiaterm and some lymphomas, confront diseases that are difficult to eradicate and for which treatments can be toxic? Trained as incrementalists, they do not expect cures so much as palliation. Their role is not to make hard decisions, but instead to rely upon precedence. Educated in the school of small advances, these physicians are not rewarded for individual successes but they are harshly criticized for any departures from community standards.
Deprived of the opportunity to make bold decisions, medical oncologists follow opinion leaders who instruct them to accrue to standardized protocols. As meaningful advances are few and far between, enormous numbers of patients must be accrued to provide sample sizes with any hope of achieving statistical significance. Among the most disturbing examples of this approach was a trial reported in patients with inoperable pancreatic cancer. The study compared single agent gemcitabine to gemcitabine plus erlotinib. The trial achieved an improvement in survival that led the FDA to approve the two-drug combination. Yet, the actual improvement in median survival was a mere 10 days. The authors beamed, “To our knowledge, this randomized phase III trial is the first to demonstrate statistically significantly improved survival in advanced pancreatic cancer by adding any agent to gemcitabine.” (Moore, MJ et al J Clin Oncol, 2007). To the average observer however, a clinical trial that required 569 patients to improve median survival from 5.91 months to 6.24 months (10 days) would hardly seem cause for celebration.
Medical oncologists have become so accustomed to these marginal advances that they are unmoved to depart from standard protocols lest they be accused of breeching guidelines. This might be acceptable if chemotherapyterm provided meaningful benefits, but the extremely modest advantages provided by even the best clinical trials scream for medical oncologists to think, well, more like surgeons.
While community oncologists think it heresy to step around a National Comprehensive Cancer Network (NCCN) guideline, investigators at the best institutions, the opinion leaders, have begun to question the merit of blind protocol accrual and come to recognize that many critical questions cannot be easily answered through the current trial process. Questions such as the role of liver resection for colon cancer patients with disease spread to the liver or the role of additional chemotherapy after that liver surgery, simply may not lend themselves to randomized trials. In a review of the topic by one of the leading investigators in the field, Dr. Nancy Kemeny from Memorial Sloan-Kettering in New York examined this dilemma, “The management plan for each patient should be decided by a multidisciplinary team, it may not be possible or ethically defensible to perform large randomized adjuvant trials comparing chemotherapy with surgery alone or comparing modern chemotherapy with older regimens. It may be reasonable to extrapolate from adjuvant trials and meta-analyses showing predominantly disease-free survival benefit. Each decision on postoperative chemotherapy should be viewed in context of prior treatment, surgical preference and individual patient characteristics.”
How refreshing. Finally a clinical investigator has recognized that patients must be managed on an “individual basis” regardless of what the clinical trial data does or does not support.
The concept of personalized medicine flies in the face of contemporary guideline driven treatment. Individualized care is on a collision course with the NCCN. It is time for medical oncologists to reclaim the high ground in doing what is right for patients, using resources that enable them to make smart decisions and to eschew standardized care. In cancer, the dictum “one size fits all” is more accurately “one size fits none.”