The care and management of patients with oncologic disorders is highly complex, often requiring combinations of surgical, radiologic, and chemotherapeutic interventions. Supportive measures are usually needed when patients utilize these primary and adjunctive treatments. Several layers of health professionals are involved in the care and treatment of patients with cancer, which leads to the need for a strong scientific basis for the treatments prescribed.
Health professionals and managed care executives know well that cancer management is not straightforward— it is not like peptic ulcers, not like hepatitis C, and not even like the autoimmune disorders. Myriad cancer treatments, including chemotherapies, biologics, and immunotherapies, have been approved for use, some with multiple indications. Not every drug—and there are many more in the pipeline—works for every patient, and the extent to which they work varies patient by patient. This lends itself to much experimentation by individual providers to try to optimize an individual patient’s clinical outcome and quality of life. It is then unsurprising that a paper from the American Society of Clinical Oncology found that 50% or more of prescribing of cancer agents is for unapproved indications.1
A Cautionary History for Managed Care
The magnitude of off-label prescribing of oncology agents makes it difficult for payers—health plans, insurers, selfinsured employers, and governments— to address what should be covered and what should be excluded from coverage. In the 1990s and early 2000s, managed care organizations had been wary of applying traditional utilization management tools to the oncology category of medications. Part of this hesitation is related to the acute nature of the disease. Part is related to the fact that a specific medication may not work for everyone with lung cancer, for example. However, much of this reluctance can be traced back to Fox v HealthNet, in which a jury awarded Nellie Fox’s family $89 million in 1993 for the California health plan denial of coverage for an experimental treatment (high-dose chemotherapy followed by bone marrow transplant) to this patient with metastatic breast cancer.2 (Ms Fox’s family obtained the investigational treatment and she passed away soon afterward.) Since then, managed care has been wary of the medico-legal implications of not covering investigational cancer treatments.
An additional concern may now have arisen, partly owing to managed care’s caution in using its tool box to manage oncology costs. In a survey conducted for this article, one health plan medical director said that the biggest concern he faced in off-label prescribing was not the legal implications of withholding coverage, but “the medico-legal risk for the plan for allowing unproven treatments with high risk for toxicity or mortality.”
The Need for Credible, Authoritative Resources
In the 1970s and 1980s, Medicare officials faced a similar situation—in the face of highly questionable treatments with little or no evidence behind them (remember Laetrile?), how should they decide whether to reimburse a provider for use of an investigational treatment? Congress chose to let experts outside the government help them make this coverage determination. The Omnibus Budget Reconciliation Act of 1993 required Medicare to cover cancer interventions that were included in one of 3 compendia available at the time:
(1) the American Hospital Formulary Service’s Drug Information (AHFSDI),
(2) the American Medical Association’s Drug Evaluations (AMA-DE), or
(3) the US Pharmacopoeia’s Drug Information (USP DI). If the treatment was judged by one of these organizations as “medically acceptable,” it would be covered. Commercial plans that also participated in the Medicare Advantage program (then called Medicare+Choice) were required to comply, and many also adopted these criteria for their commercial populations.
If a Medicare Advantage plan opted to deny coverage for a drug for a specific off-label use, but the drug’s use was recommended by one of the Medicareapproved compendia, the plan was expected to explain why “not having a US Food and Drug Administration (FDA)-approved indication” was not deemed an acceptable reason.3
In 2008, the Centers for Medicare & Medicaid Services (CMS) updated its list of accepted compendia, partly because 2 had ceased publication over time. This review was based on several criteria, including details on the evidence available for each intervention; use of published criteria for weighing the evidence; use of a prespecified method for making recommendations; a transparent process for evaluating therapies (though not necessarily a public process); and a process for publicly identifying potential conflicts of interest of the compendia’s parent publishers and owners, reviewers, and committee members.4 Four compendia were chosen5:
(1) the American Hospital Formulary Service—Drug Information published by the American Society of Health-System Pharmacists (www.ahfsdruginformation.com)
(2) Elsevier Gold Standard’s Clinical Pharmacology (www.clinicalpharmacology.com)
(3) The National Comprehensive Cancer Network’s (NCCN’s) Drug Information & Biologics Compendium (www.nccn.org)
(4) Thomson Micromedex DrugDex (www.micromedex.com/products/drugdex)
The Clinical Pharmacology and Drug-Dex compendia are available through subscription only. The AHFS-DI and NCCN compendia can be accessed without charge. In order to keep abreast of changes in the publishing environment, CMS will review its compendia choices annually. Medicare recognizes 26 peerreviewed journals as accepted sources for clinical study evidence for off-label uses of oncology medications ().
The review process, particularly for off-label uses of oncology products, is complex and has multiple levels. Jane Miller, PharmD, senior associate editor, AHFS DI, told Evidence-Based Oncology that under the current off-label oncology review process, AHFS staff prepare an evidence table based on their review of the literature, propose a level of evidence rating, and prepare a narrative summary, which are provided, along with copies of the relevant published articles, to the AHFS Oncology Expert Committee for review. “The committee members are asked to perform an independent review, indicate their vote on the level of evidence rating, and provide a grade of recommendation and supporting comments using an evidence-based analysis of the available literature,” said Dr Miller. Executives at NCCN declined to provide comment for this article.
The Value of Compendia
For MCOs that have commercial and/or Medicaid products in addition to Medicare beneficiaries, Medicare coverage decisions are often used as a benchmark for their other patient populations. Although all Medicare Advantage plans must consider compendia in their coverage decisions, this is not required for insurers covering commercial populations.
For commercial insurers particularly, health plans and insurers will look to other resources or guideposts to help make coverage determinations. That is not to say that compendia do not play an important role in making coverage determinations for commercial plans. For example, a brief, informal survey of a small sample of regional managed care payers indicated that for most, a therapy found in the compendia still ranks as the most important determinant of coverage. Certainly, for commercial plans, FDA approval plays a role, but inclusion of the product in respected practice guidelines is important, as is evidence published in the peer-reviewed literature (not abstracts). Yet, compendia listing is often sufficient for the coverage determination. Clinical pathways, such as those from NCCN, and clinical studies from SWOG (formerly known as the Southwestern Oncology Group) or the Eastern Cooperative Oncology Group help form the foundation of oncology coverage decision making. Keeping apace with advances can be a daunting challenge for managed care executives.This raises the value of compendia—an independent body to track the latest advances in the use of oncology products and post a transparent assessment of the evidence behind their use.
Health plans do seem to value specific compendia more than others. All of the respondents in the informal survey of managed care payers indicated that NCCN’s compendium was the most highly rated. This may be a result of the fact that NCCN’s compendium is based on its own highly respected clinical guidelines. Clearly, the AHFS-DI has the longest history, having been implemented before 1993.
Comparing the Compendia
Using multiple compendia for oncology coverage decisions can be daunting, and CMS recognized this during its review, leading to its 2008 decision. For example, CMS stated that the compendia use various systems to rate and recommend agents that are not exactly the same across different compendia, which can lead to “cross-walking” challenges. “In general, a use is identified by a compendium as medically accepted if the (1) indication is a Category 1 or 2A [rating] in NCCN, or Class I, Class IIa, or Class IIb in DrugDex; or, (2) narrative text in AHFS or Clinical Pharmacology is ‘supportive.’
“A use is not medically accepted by a compendium if the (1) indication is a Category 3 in NCCN or a Class III in DrugDex; or, (2) narrative text in AHFS or Clinical Pharmacology is ‘not supportive.’ The complete absence of narrative text on a use is considered neither supportive nor non-supportive.”
There are a few distinguishing features among the compendia, other than how they express their ratings of the evidence. The compendium from NCCN is derived from its own oncology clinical practice guidelines, which are “consensus- based” (ie, a multidisciplinary panel of mostly subspecialists review the evidence and reach a consensus on practice).
Another distinguishing feature relates to how a medication is added to the compendia. Pharmaceutical manufacturers and biotech companies can apply to NCCN and AHFS to have their products considered for inclusion in the compendia through a formalized application process. DrugDex and Gold Standard do not have a formal application process. The compendia’s publishers claim to update their lists on a regular basis: AHFS is updated at least monthly.6 According to NCCN, its compendium is updated monthly; in addition, NCCN will review applications from industry or oncology community specialists to include a product at least once annually.
Final decisions on drug coverage are through a “multi-level balloting process” at AHFS, whereas NCCN’s consensus approach leans toward a decision by the majority. Neither DrugDex nor Gold Standard utilize a voting process, but rely upon the opinion of external reviewers to evaluate their recommendations.6
The compendia publishers in general do not explicitly consider the cost of an agent to treat off-label indications. “Although cost considerations may be described in AHFS DI monographs, drug prices per se are not included,” said Dr Miller, “since such prices can change frequently and because of the difficulty in establishing prices that would be meaningful across various practice settings.” Sometimes, cost considerations may be taken up by AHFS DI for nononcology reviews, but only after the clinical question has been reviewed,6 For example, drugs accepted may come with broad caveats, such as “Given the modest improvement in survival rates reported to date, any decision to use cetuximab in combination with vinorelbine and cisplatin [for stage IIIB or IV non-small cell lung cancer] must involve careful consideration of the regimen complexity (eg, weekly infusions) and associated toxicity, cost, and quality of life.” She added, “Because reviewers are asked to provide comments supporting their recommendations, cost of therapy may be discussed in the off-label determination much as it may be discussed in the overall AHFS DI database.”
Conflicts of interest were a considerable talking point when CMS made its compendia choices in 2008. Each compendium has attempted to address these issues, through various measures. At NCCN, a panel chair has the responsibility of evaluating all potential financial conflicts. This reviewer will categorize financial conflicts of interest into one of several defined levels. At AHFS DI and at DrugDex, financial conflict of interest form submission is required for all reviewers and professionals involved with its compendium. The Sidebar outlines the depth of reviewer disclosure requirements by AHFS DI.
Protection for Pharmaceutical Companies and Health Plans
In the 1980s, health plans and insurers were concerned about their medicolegal risks in covering unproven treatments and the possible harm to patients. If an investigational agent caused death, they could be subject to lawsuits from the family. The Fox v HealthNet decision and other similar suits caused payers to shy away from aggressive management of oncology drugs because of the fear of other medico-legal risks: the risk of not providing coverage. Today, however, following compendia recommendations may serve as a solid defense for payers seeking to exclude certain therapies for both clinical risk versus benefit, and cost concerns, discouraging civil suits of this type.
Interestingly, this may serve to deter lawsuits against the pharmaceutical industry marketing practices as well: Marketing of off-label uses of an approved drug may result in heavy fines in areas other than oncology. Yet, compendia listing may provide limited protection for marketing off-label oncology uses. In an article in 2008, Daniel Miller, director of Delaware’s Medicaid Fraud Control Unit, reported that “The state has nearly 150 current whistleblower cases it could investigate—most of them off-label related. If we find out that the drug in question is recommended by the government-approved compendia, then we probably wouldn’t take the case,” he said.
However, the FDA itself may have provided its own form of protection: in 2009, it implemented its Good Reprint Practices guideline.7 According to Ben Martin, member of the law firm Epstein, Becker & Greene, this may represent an “effective safe harbor.” He explained, “A manufacturer can hand out reprints of a scientific article used to support the addition to the compendium, as long as the manufacturer includes a copy of the approved labeling and does not mark up or highlight the reprint in any way.” Mr Martin pointed out that this applies to any off-label use (not just oncology products and their inclusion in compendia).
The only certainty for the near term is that oncology therapy will become even more complex. Dr Miller commented, “Revisions related to off-label uses must be balanced against other database revisions (including important safety information, newly approved drugs, revised authoritative clinical guidelines, new labeled uses, new formulations, and other developments) that affect numerous therapeutic areas. In addition, the rigorous, highly structured, evidencebased process that AHFS employs for off-label oncology determinations is a resource-intensive process with no separate, dedicated source of funding.” She indicated that possible resource limitations in the future could have an impact on the timing of updates to the compendia.
Health plans and insurers will likely continue to rely on compendia as a foundation for coverage determinations, particularly without better evidence for solid decision making. Does this mean that the cautionary approach to oncology therapies will continue into the future? The high costs of specialty pharmaceuticals and new chemotherapies and biologics will pressure managed care organizations to review their tactics (or their member premiums).
Sidebar. Disclosing Conflicts of Interest at AHFS DI
According to Jane Miller, PharmD, senior associate editor, AHFS DI, participation is solicited but voluntary, and no honorarium or other benefit is provided. These experts must provide full disclosure of interest, including any affiliation they have with or financial involvement they have in the sponsor of the drug(s) under consideration and in directly competitive products. In the case of balloted determinations made by the AHFS Oncology Expert Committee, conflict of interest disclosure policies follow the definition of a publicly transparent process for identifying potential conflicts of interest as established in Section 414.930(a) of the Code of Federal Regulations (CFR).
“All AHFS Oncology Expert Committee members are required to complete disclosure forms listing direct and indirect financial interests that are relevant to matters considered by the Committee for themselves, their spouse, and any minor children upon appointment to the Committee and at least every 3 years thereafter. In addition, prior to each off-label use review, those Committee members selected to participate in the review are required to update this disclosure information to identify any potential new conflicts of interests.
“Prior to each off-label use review, the AHFS staff screen Committee members’ current disclosure information and select individuals who do not have direct or indirect conflicts of interest involving themselves, their spouse, or minor children that relate to the specific off-label use being reviewed. When the disclosures are evaluated for potential conflicts, interests or affiliations involving the sponsor of the drug under consideration, a potential competitor, or an affected firm or organization are all considered. An interest is considered relevant if the Committee’s decision on medical acceptance of the off-label use has the potential to result in a direct and predictable impact on the interest.” She also described AHFS’s guidelines for evaluating reviewers’ financial interests:
• Committee members are not allowed to participate in the off-label determination process if they have disclosed relevant direct or indirect financial interests involving themselves, their spouse, or any minor child that have a combined value exceeding $50,000.
• Committee members who have disclosed relevant direct or indirect financial interests involving themselves, their spouse, or any minor child that have a combined value of less than $50,000 are not asked to serve as reviewers. However, if the Committee member’s expertise is deemed to be required, the AHFS staff determine the merits of using the Committee member in an advisory capacity to provide expert review of the subject matter. The AHFS staff may decide to grant a waiver permitting the Committee member to participate in an advisory capacity, but the Committee member is not permitted to participate in the review of clinical materials or in the voting process.
• Committee members who have affiliations with a sponsor or organization (ie, consultant, speakers bureau) that are relevant to the off-label use under consideration, or whose spouse or any minor child has such affiliations, regardless of whether they receive any form of monetary compensation, are not asked to serve as reviewers. If the Committee member’s expertise is deemed to be required, the AHFS staff determine the merits of using the Committee member in an advisory capacity to provide expert review of the subject matter. The AHFS staff may decide to grant a waiver permitting the Committee member to participate in an advisory capacity, but the Committee member is not permitted to participate in the review of clinical materials or in the voting process.
AHFS DI indicates American Hospital Formulary Service’s Drug Information.
Payer PerspectiveInterview With Gary M. Owens, MD
EBO: What are your greatest concerns about offlabel use for oncology products?
The concerns are 3-fold: (1) use of expensive agents for patients where there are no or minimal data on clinical effectiveness, in other words, providing futile care; (2) the possible untoward side effects and even mortality for patients who have minimal chance of responding to the drug; and (3) the cost of the new cancer agents (averaging $5000 per month or more), which is just too great to allow unrestricted use.
EBO: What tools have you found most useful when determining coverage for off-label indications of oncology medications?
Personally, I found the NCCN Compendium to be the number 1 resource for oncology, and to a lesser extent, AFHS-DI, Micromedex, and Elsevier compendia.
EBO: How often should oncology compendia be updated to keep up with developments in this space?
They should be updated at least on a quarterly basis and possibly even more frequently if new, groundbreaking data are released. Cancer research is rapid, and the compendia must keep abreast of changing knowledge.
EBO: Do you anticipate that combinations of oncology treatments will be increasingly tried over the next 5 years? If so, what may be the best tool(s) for managing their use?
I think that as we gain experience with new biologics and small molecules, such as the tyrosine kinase inhibitors and JAK inhibitors to name a few, we will see more attempts at combination therapy in an effort to improve outcomes. I think these combinations will need to be tried in carefully designed postmarketing protocols (eg, SWOG, Eastern Cooperative Oncology Group, National Cancer Institute) and only after proven in this way should they be introduced into common practice. If proven by protocol, then the current compendium process is the best way to manage the combinations. Again, the cost is just too great to allow unfettered use of combinations without the evidence to support it.
EBO: Specialty pharmaceuticals are increasingly used to treat oncologic disorders. For biologic agents, will health plans and insurers consider covering more genomic testing to ensure optimal use of these expensive agents?
Yes, they will, but only if the clinical utility of the genetic/genomic or gene expression test can be shown to have clinical utility in patient selection, drug selection, or side-effect management. These tests have the potential to help guide therapy, but not all are created equal. We can only afford to pay for those that guide therapy and have clinical utility. Some examples of these include tests for KRAS, BRAF, HER overexpression, and ALK mutations.
EBO: Do you believe that compendia publishers adequately manage the risk for conflict of interest of reviewers?
We certainly live in an era where potential conflict can be there. We don’t always know the relationship between a reviewer and a manufacturer. For instance, a reviewer could be a principal investigator in a study and have built-in bias. This is a difficult area to oversee.
Dr Owens is president of Gary Owens Associates in Glen Mills, PA, and a former health plan medical director.Funding Source: None.
Author Disclosures: Mr Mehr reports receiving payment for involvement in the preparation of this article.
Authorship Information: Concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content; statistical analysis; provision of study materials or patients; obtaining funding; administrative, technical, or logistic support; and supervision.1. American Society of Clinical Oncology. Reimbursement for cancer treatment: coverage of off-label drug indications. J Clin Oncol. 2006; 24:3206-3208.
2. Eckholm E. $89 Million awarded family who sued H.M.O. The New York Times, December 30, 1993. http://www.nytimes.com/1993/12/30/us/89-million-awarded-family-who-sued-hmo.html. Accessed September 26, 2012.
3. Medicare Benefit Policy Manual. Chapter 15, section 50.4.5-Unlabeled Use for Anti-Cancer Drugs. Rockville, MD: Centers for Medicare & Medicaid Services.
4. Tillman K, Burton B, Jacques LB, et al. Compendia and anticancer therapy under Medicare. Ann Intern Med. 2009;150(5):348-350.
5. Compendia as Authoritative Sources for Use in the Determination of a “Medically Accepted Indication” of Drugs and Biologicals Used Off- Label in an Anti-Cancer Chemotherapeutic Regimen. Centers for Medicaid & Medicare Services; October 24, 2008. http://www.cms.hhs.gov/transmittals/downloads/R96BP.pdf. Accessed September 21, 2012.
6. Cote B. Oncology compendia: greater transparency and regulatory attention at the forefront. Oncology Business Review. 2010;10-14.
7. Good Reprint Practices for the Distribution of Medical Journal Articles and Medical or Scientific Reference Publications on Unapproved New Uses of Approved Drugs and Approved or Cleared Medical Devices. Food and Drug Administration 2009. http://www.fda.gov/regulatoryinformation/guidances/ucm125126.htm. Accessed October 18, 2012.