Although the goals of quality-care and value-based oncology initiatives are legitimate and laudable, some of the assertions should be examined more closely.
Maurie Markman, MD
It is more than a little uncomfortable to be on the other side of a “mom and apple pie” issue, but this commentary seeks to ask difficult questions regarding undoubtedly well-intended quality-of-care and value-based initiatives, including paradigm-changing activities of the American Society of Clinical Oncology (ASCO) and its participation in the Choosing Wisely campaign.1,2
The goals of these efforts within the oncology community are highly legitimate. However, there are challenges to effective implementation of such programs that may develop because the decision to include a particular item may not have been optimally considered or because there may be an insufficient understanding of the reasons for the presumed overutilization or inappropriate use of a given strategy being reviewed.
At the onset, let’s establish that there are a number of items in the ASCO quality initiatives that require no further discussion, and their inclusion is both self evident and unquestionably appropriate.
For example, it is difficult to argue against avoiding the delivery of chemotherapy within the last 2 weeks of a cancer patient’s life. Of course, there will unfortunately always be individual patients who experience a rare fatal complication of therapy or a sudden serious event related to the presence of the malignancy itself that results in a rapid downhill course.
Further, there will be patients treated aggressively for curable malignancies (eg, acute leukemia) who experience an expected serious toxicity such as severe neutropenia that tragically translates into a short-term fatal outcome like infection-associated septic shock. But, in general, oncologists should be able to define a patient population with overwhelmingly progressive disease or such a poor performance status where a 2-week or shorter survival is realistically the most likely outcome.
Similarly, demanding that patients of reproductive age have the opportunity to discuss medically acceptable options for fertility preservation with documentation in the medical record that such discussions have occurred should be non-negotiable.
Again, this statement does not mean fertility preservation must absolutely be offered simply because of a patient’s age, but rather that documentation is included in the medical record that this critical course of action has been considered if appropriate.However, the reason for inclusion of other items, or the underlying rationale for the selection of a particular metric to define appropriate utilization, can be questioned.
Perhaps the most seriously flawed oncology-related metric that has been selected by ASCO for inclusion in the Choosing Wisely Campaign is the use of colony-stimulating factors (CSFs) in routine clinical practice.2 In its most recent update,3 ASCO reaches this conclusion about the administration of these agents:
“Prophylactic use of CSFs to reduce the risk of febrile neutropenia is warranted when the risk of febrile neutropenia is approximately 20% or higher and no other equally effective and safe regimen that does not require CSFs is available.”
What is the justification for the selection of this 20% figure to distinguish appropriate from inappropriate delivery of CSFs? What possible reason can be advanced to objectively declare that somehow “quality” in this area has been achieved if CSFs are employed when the risk of a febrile neutropenic episode is predicted in 1 of 4 treated patients, but not if the risk is predicted in only 1 of 6 patients? Further, and an even more serious concern, from what database is this 20% risk of febrile neutropenia for a specific chemotherapy regimen defined? The most likely answer here are data generated from evidence-based (randomized) trials, often the specific studies employed for regulatory approval of a particular novel agent. Unfortunately, this answer is highly problematic. It is increasingly recognized that the populations of patients entered into large phase III randomized industry-sponsored or cooperative-group conducted clinical trials bear remarkably limited resemblance in their clinical features (eg, age, performance status, presence of comorbidities, prior therapeutic history) to the real world of patients with cancer treated in the community. And it is the community experience that represents by far the largest number of patients managed with malignant disease.
As noted in ASCO’s own CSF guidelines, there are patients with a heightened risk for febrile neutropenia, with factors including “age ≥65, poor performance or nutritional status, poor liver or renal dysfunction, cardiovascular disease, multiple comorbid conditions, and previous chemotherapy or radiation therapy” where prophylactic use of CSFs may be appropriate.3
In fact, in a recent review of more than 160,000 patients with cancer involving multiple tumor types and chemotherapy regimens, 74%-98% of this “real world population” was found to have one or more risk factors for febrile neutropenia.4 Further, within the population with at least one risk factor, the incidence of febrile neutropenia across multiple regimens ranged from 7.2% to 29%.
The critical point to be noted here is that highly clinically relevant risk factors for febrile neutropenia are extremely common in the community setting, and the absolute risk of an individual patient experiencing febrile neutropenia may be substantial.
It has recently been noted in one report on the topic that success to date in changing physician behavior as a result of the Choosing Wisely campaign has been, to put it kindly, underwhelming.5 Another report of a controlled trial that attempted to reduce patient interest in what the investigators characterized as “low-value” screening strategies, including prostate and colon cancer screening in particular patient populations, demonstrated that these efforts were completely unsuccessful in changing that stated interest.6
Importantly, although these experiences are not specific to ASCO, they emphasize the difficulty associated with impacting utilization patterns, both at the physician and patient/public level. In conclusion, certain existing oncology-related parameters previously selected as part of this worthy quality and value-based effort, and the specific metrics employed in their assessment, should at some point in the not-so-distant future be reexamined for their own value in influencing this important undertaking.
Maurie Markman, MD, editor-in-chief, is president of Medicine & Science at Cancer Treatment Centers of America, and clinical professor of Medicine, Drexel University College of Medicine. firstname.lastname@example.org.