Going Off-Label: It's Time to Ponder the Nature of Sufficient Evidence

OncologyLive, December 2012, Volume 13, Issue 12
Partner | Oncology Network Providers | <b>Cancer Treatment Centers of America</b>

In the absence of a phase III randomized trial, what constitutes sufficient evidence to justify use of a treatment, particularly a nontraditional treatment or one that challenges common practice?

Maurie Markman, MD

Editor-in-Chief of OncologyLive

Senior vice president for Clinical Affairs and National Director for Medical Oncology Cancer Treatment Centers of America, Eastern Regional Medical Center

Oncologists are well aware of the concept of off-label use of a commercially available antineoplastic agent. In fact, this strategy is a routine and often highly beneficial component of the management of a particular patient’s malignancy. Further, physicians recognize that care must be individualized based on factors that are unique to a specific patient, including age, relevant comorbidities, prior treatment history, and availability of various options. In addition, cost, third-party coverage, and patient choice must be considered.

However, in considering off-label use and individualization of care, another factor must be taken into account: In the absence of a phase III randomized trial—or of approval based on phase II data and the FDA’s accelerated approval program—what constitutes sufficient evidence to justify use of a treatment, particularly a nontraditional treatment or one that challenges common practice? And how do we, as healthcare providers, decide which “unproven” treatment options to present to our patients?

This commentary is meant to challenge the reader to ponder what some might consider a radical departure from the normal medical school or oncology training program view of sufficient evidence. Consider, for example, the recent report of the impressive impact of aspirin use in preventing death from colon cancer in patients whose tumors possess a mutation in PIK3CA (Table 1).1 Despite the fact that this analysis was based on sound epidemiologic and laboratory science, the reported outcome did not emanate from the results of a phase III randomized trial and involved only a limited total number of colon cancer patients whose cancers were retrospectively examined for the presence or absence of specific molecular markers. Further, disease management was not standardized, and there was no ability to control for recognized relevant prognostic factors (eg, age, stage, tumor grade, comorbidities). Finally, the specific impact of the use of aspirin pertains to less than one-fifth of individuals (17% of tumors found to have mutations in PIK3CA) with colon cancer.

Table 1. Impact of Aspirin Use on Disease- Specific Mortality in Colon Cancer Based on Presence or Absence of a PIK3CA Mutation1


of Patients


of Deaths

Hazard ratio

(Deaths for use vs

no use of aspirin)

No mutation





No aspirin



PIK3CA mutation





No aspirin



Hazard ratio adjusted for stage of disease and other variables.

Although it would be inappropriate to declare the results of this analysis definitive, the question remains whether this level of evidence should prevent an individual physician from presenting the option of aspirin administration to a select group of patients with colon cancer. Is it truly necessary to require a randomized phase III trial before including a particular intriguing strategy in routine clinical practice?

Assuming that such a trial is actually undertaken, what if it requires a decade or longer to generate relevant long-term clinical evidence? Should all individuals who might potentially benefit from the use of aspirin be denied this treatment option during this extended interval?

For completeness, it should be acknowledged that it is uncertain what entity would even be willing to conduct a phase III randomized trial examining the role of aspirin in patients with colon cancer whose tumors possess a PIK3CA mutation. While it is theoretically possible that a government research agency would elect to spend the large sums of money necessary to conduct this multiyear effort, it is virtually certain there will be precisely zero interest by any pharmaceutical company to fund this effort for the purpose of generating revenue from additional sales of generic aspirin.

It is also important to note that despite the minimal cost of aspirin—according to the study results, benefits might be obtained through the daily use of low-dose aspirin—only a fraction of the entire population would be expected to achieve any benefit from this strategy. Thus, unless one wished to make the theoretical argument that all patients with colon cancer should be given aspirin, perhaps for its favorable cardiac effects, the use of this inexpensive drug intervention would mandate a relatively expensive molecular analysis of the tumor. Of course, a counterargument could be advanced that such an analysis might be undertaken to discover other relevant information, such as KRAS mutational status and the finding of germline polymorphisms that may be relevant to the metabolism of antineoplastic chemotherapy.

It is not difficult to cite other examples of highly provocative data that have recently appeared in the peer-reviewed oncology literature that are based on far less rigorous evidence than a randomized phase III trial, and yet have the realistic potential to favorably impact survival and quality of life for patients diagnosed with serious malignant conditions (Table 22-7).

Table 2. Possible Favorable Cancer Outcomes for Nontraditional Strategies Not Based on Phase III Trial Data

✓ Statin use to reduce the risk of cancer mortality2

✓ Multivitamin use to prevent cancer in men3

✓ Surgical resection improving survival in low-grade gliomas (vs a strategy of "watchful waiting")4

✓ Beta blocker administration to prevent cancer recurrence5-7

Thus, the issue is clear: The evidence may not be unequivocal, and it is possible that the evidence may someday be refuted by a higher level of evidence, such as a large phase III randomized trial. But individual patients may not require such definitive data. Faced with a potentially life-threatening situation, patients may feel comfortable using the existing level of evidence to make a decision. In the opinion of this commentator, patients should be given the opportunity to consider treatment options that go beyond the traditional off-label use of antineoplastic therapies.

In presenting treatment options to patients, it is essential that healthcare professionals clearly and objectively explain the potential benefits and risks of clinically reasonable therapeutic options. Healthcare professionals should also explain the inherent limitations of any data analysis used to inform that decision, such as an epidemiologic or retrospective study versus a phase III randomized trial.

A discussion of treatment options should not only consider all appropriate management options (including possible participation in a clinical trial), but must also specifically encompass approaches in which gold-standard evidence from a phase III randomized trial seriously challenges the existing management paradigm. For example, many in the gynecologic oncology surgical community continue to declare that an attempt at maximal surgical cytoreduction in advanced ovarian cancer must remain the standard of care, despite evidence from an international, multicenter, phase III randomized trial that strongly supports the therapeutic equivalence and reduced treatment-related morbidity and mortality associated with delivering several cycles of platinum-based systemic chemotherapy prior to surgical resection.8 Thus, in discussing treatment options with a patient newly diagnosed with advanced ovarian cancer, the evidence supporting neoadjuvant chemotherapy should be presented along with the traditional treatment approach.

In conclusion, there is a critical distinction between the evidence required by the academic community to maximize the opportunity for certainty, by government regulatory agencies to decide if a given drug should be permitted to enter the commercial market, and by patients considering therapeutic options for their own life-threatening illness. And in this specific discussion, the opinion of your patients, based on their review of all available evidence related to reasonable therapeutic strategies, should not be ignored, even if their opinions are not in agreement with the academically defined level of evidence or seriously challenge long-standing approaches employed by the practicing oncology community.


  1. Liao X, Lochhead P, Nishihara R, et al. Aspirin use, tumor PIK3CA mutation, and colorectal-cancer survival. N Engl J Med. 2012;367:1596-1606.
  2. Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancerrelated mortality. N Engl J Med. 2012;367-1792-1802
  3. Gaziano JM, Sesso HD, Christen WG, et al. Multivitamins in the prevention of cancer in men. JAMA. 2012;308:1871-1880.
  4. Jakola AS, Myrmel KS, Kloster R, et al. Comparison of a strategy favoring early surgical resection vs a strategy favoring watchful waiting in low-grade gliomas. JAMA. 2012;308:1881-1888.
  5. Melhem-Bertradt A, Chavez-MacGregor M, Lei X, et al. Beta-blocker use is associated with improved relapse-free survival in patients with triplenegative breast cancer. J Clin Oncol. 2011;29:2645-2652.
  6. De Giorgi V, Grazzini M, Gandini S, et al. Treatment with β-blockers and reduced disease progression in patients with thick melanoma. Arch Intern Med. 2011;171:779-781.
  7. Diaz ES, Karlan BY, Li AJ. Impact of beta blockers on epithelial ovarian cancer survival. Gynecol Oncol. 2012;127:375-378.
  8. Vergote I, Trope CG, Amant F, et al. Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer. N Engl J Med. 2010;363:943-953.