Making Evidence-Based Decisions in Specialty Pharmacy

Formal oncology clinical practice guidelines are not as reliable as some may think, suggests a study presented at the 2013 Academy of Managed Care Pharmacy Nexus meeting.

While none of the targeted therapies in oncology that have been approved over the past few decades have “cured” patients of metastatic disease, many of them have successfully helped patients achieve remission. Some of the immune-based therapies in the pipeline may prove to be curative, however, noted David Hong, MD, of the University of Texas MD Anderson Cancer Center, at the 2013 Academy of Managed Care Pharmacy (AMCP) Nexus meeting held October 15 to 18 in San Antonio, Texas. But how will their efficacy be measured, and how will decisions regarding the use of these immunomodulators be made?

Molecular profiling not only allows health care providers to tell when a targeted therapy may work to treat a patient with cancer, it also provides evidence for when a drug is not likely to be beneficial in a certain patient. But, many factors besides molecular testing play into a provider’s decision to choose one intervention over another.

Increasingly, prescribing practices are being influenced by outside factors such as industry guidelines and compendia. The groups weighing in on these decisions include the American Society of Clinical Oncology (ASCO), the National Comprehensive Cancer Network (NCCN), and the Clinical Pharmacogenetics Implementation Consortium (CPIC). Decisions governing care for the cancer patient are compiled in such compendia as the American Hospital Formulary Service (AHFS), Thompson Micromedex DrugDex, Clinical Pharmacology, and the NCCN sources.

In a panel alongside Dr. Hong, Edward Li, MD, BCOP, associate professor at the University of New England College of Pharmacy, exposed some of the major challenges and limitations of the current clinical practice guidelines and compendia in oncology. Citing research from the Journal of Clinical Oncology, Dr. Li pointed out that no guideline fully met the Institute of Medicine’s “Clinical Practice Guidelines We Can Trust” standards. This was primarily due to problems surrounding transparency and issues of conflict of interest in relation to prescribing practices. Other knowledge gaps in compendia were introduced by problems surrounding external review, systematic review (ie, “How did the therapy make it into the compendia?”), strength of evidence, articulation of recommendations, and issues with logging and maintaining updates to practice recommendations.

Most concerning to Dr. Li was the fact that the inclusion of recommendations are based on judgments, and these judgments depend on who is on certain panels, presenting a major challenge regarding conflict of interest. Additionally, drug updates within each source also varied; some did not include revisions to reflect the addition of newer treatments. Without frequent updates, it can be hard for patients and providers to determine whether a drug is affordable, argued Dr. Li, as more affordable treatments may not be listed as soon as they become available.

To illustrate the variation across compendia, Dr. Li used erlotinib as a case study, comparing the information about the drug within the AHFS, DrugDex, Clinical Pharmacology, and NCCN databases. Based on this comparison, information across the sources was inconsistent, and there wasn’t even epidermal growth factor receptor mutation information listed in the monograph found within the AHFS database.

A more comprehensive approach to drug selection comes with the use of an alternative resource called PharmGKB, noted Dr. Li. PharmGKB is, according to its website, “a pharmacogenomics knowledge resource that encompasses clinical information including dosing guidelines and drug labels, potentially clinically actionable gene-drug associations and genotype-phenotype relationships.” The resource goes through pure pharmacogenomics studies and extracts relevant gene-drug pairs from the data, effectively teasing out the most important relationships among the pairs, asserted Dr. Li. Although it is not yet recognized by Medicare as an official source, Dr. Li maintains that PharmGKB is more comprehensive than compendia and guidelines and contains more genomic information relevant to pharmacodynamics and pharmacokinetics. As Dr. Li notes, “It gives you the information, but interpretation is left up to the provider.”

The PharmGKB Knowledge Pyramid