INSIDE THE BLACK BOX is an occasional column providing insight into the U.S. Food and Drug Administration (FDA) and its policies and procedures. In this installment, FDA supervisory toxicologist Todd Palmby, PhD, and pharmacologist Eias Zahalka, PhD, MBA, discuss the approach taken in the Office of Hematology and Oncology Products (OHOP) to assess and convey the risk of potential embryo-fetal harm when treating pregnant patients with cancer. Drs. Palmby and Zahalka are members of the Division of Hematology Oncology Toxicology (DHOT) within the OHOP. The DHOT is responsible for reviewing and evaluating nonclinical (animal) studies submitted to the OHOP.
With an increasing number of products approved to treat cancer patients in the United States, oncologists sometimes struggle with making informed decisions about treatment options for pregnant patients and patients who are thinking about becoming pregnant. Finding and interpreting information about a product’s risk to embryo-fetal development can be challenging. The ASCO Post spoke with Drs. Palmby and Zahalka about the basis for determining the risk associated with a product to cause embryo-fetal harm if used to treat a pregnant patient with cancer and how that risk is conveyed to health-care professionals and patients.
Balancing Risk and Benefit
How can a health-care professional or patient determine the level of risk of fetal harm if a product is used to treat a pregnant patient?
Dr. Palmby: The best place for a health-care professional to learn about a product’s level of risk for causing embryo-fetal harm is the drug labeling, also referred to as the package insert. What is known about the embryo-fetal risk is found in the Pregnancy section of the labeling. Depending on the severity of the risk, information may also be found in the Warnings and Precautions section, the Contraindications section, and sometimes within a Boxed Warning. Information on embryo-fetal risk may also be included in other communications associated with the product, such as a Medication Guide or other FDA-approved patient labeling.
When a patient with cancer is pregnant, discussions of treatment options between the patient and the health-care professional can be guided by information on the level of risk to the fetus and balanced against the benefits of treatment. The Patient Counseling Information section of the package insert includes advice for health-care professionals to convey to women that a product may have the potential to cause embryo-fetal harm and thus to recommend avoiding pregnancy.
Quantitative Assessment of Risk
How is a product’s risk of causing fetal harm conveyed in a label, and what information is used to determine the level of risk to the fetus if the product is administered to a pregnant patient?
Dr. Zahalka: Prior to product approval, clinical trials do not generally include pregnant patients. Therefore, when a new drug is approved, animal data are usually used to establish such risk. The source of the animal data can be from peer-reviewed published scientific data or original nonclinical studies (ie, nonclinical studies investigating effects during organogenesis or the peri-/postnatal period). The nonclinical studies can be conducted in rodent and/or other species (usually rabbits for small molecules and monkeys for biotechnology products).
A quantitative assessment of risk is presented in the Pregnancy section of the labeling in the form of calculated exposure- or dose-multiples of the human dose. These multiples are provided to inform the health-care professional how close the clinical exposure or dose is to the observed adverse effect level in animal models.
Once a drug has been studied or is on the market for some time, human data may be available and can be added to the labeling. The source of clinical data can be obtained from observational studies conducted during the postmarketing period or from published case reports.
Pregnancy and Lactation Labeling Rule
What is the Pregnancy and Lactation Labeling Rule, and how does it change the labeling of oncology products to convey the risk of use during pregnancy?
Dr. Zahalka: The wheels of change have been in motion since the late 1990s and will affect the labeling of all pharmaceuticals, including oncology products. In 2008, a proposed Pregnancy and Lactation Labeling Rule (PLLR) was issued. The PLLR issued on December 4, 2014, will change the way product labeling organizes the information about embryo-fetal risk. The letter categories will be discontinued and replaced by a narrative risk summary, including the available human or animal data.
The final PLLR, which goes into effect on June 30, 2015, incorporated comments provided by experts in the field and important stakeholders who are responsible for making the changes in their product labeling including the pharmaceutical industry. The new label format will be applied to any drug or biologic approved after implementation of the PLLR as well as to previously approved drugs that are subject to label revision by way of FDA’s 2006 Physician Labeling Rule. Revision of the labels of previously approved drugs to conform to the PLLR format is expected to be implemented over 3 to 5 years.
More Detailed Information
What was the purpose for changing the pregnancy category approach in the package insert?
Dr. Zahalka: The main concern with labeling practices using pregnancy categories is that complete risk information is not conveyed by a single letter category rating system. For instance, the spectrum of animal data for a Category C labeled drug on which the risk was based could range from minor fetal body weight changes to severe malformations, which gave the reader no indication where along that spectrum this drug fell. Additional criticisms are that the animal data descriptions in the pregnancy section cannot be understood by the average person and that the letter system utilized gave the reader an incorrect impression that reproductive risk increases from Category A to X.
The goal of the PLLR is to give health-care professionals and patients more detailed information in the labeling regarding the potential risk of drug administration during pregnancy and lactation. The labeling will replace the letter category with a risk summary paragraph, which provides an overview of clinical and nonclinical effects, along with a reader-friendly estimate of risk. It is hoped that the additional information will enable a more informed approach toward managing risk.
Changes in Labeling Section
How will the package inserts differ under the Pregnancy and Lactation Labeling Rule compared with the former pregnancy category approach?
Dr. Zahalka: There will be no major change in content as compared with the former approach; however, the labeling section will be organized somewhat differently. Under the PLLR, the labeling will contain three subsections: one for pregnancy, one for lactation, and one for females and males of reproductive age.
In the Pregnancy section, the labeling change eliminates the pregnancy categories A, B, C, D, and X. The previous Sections 8.1 “Pregnancy” and 8.2 “Labor and Delivery” are merged into the single “Pregnancy” subsection under the PLLR. Additionally, section 8.3 “Nursing Mothers” is renamed the “Lactation” subsection and is now section 8.2. A new 8.3 section is named Females and Males of Reproductive Potential. (For a comparison of the current prescription drug labeling and the new PLLR labeling requirements, visit http://www.fda.gov/drugs/developmentapprovalprocess/developmentresources/labeling/ucm093307.htm.)
The Pregnancy and Lactation subsections will have three principal components: a risk summary, a clinical considerations section, and a data section. The inclusion of a narrative risk summary in these subsections is a key change under the PLLR. It is intended to provide the reader with enough detail to inform risk but not extensive technical details. In this manner, it is hoped that the Pregnancy/Lactation sections of the label will be more informative than the Pregnancy category labeling system and will be presented in a more reader-friendly format. The subsection Females and Males of Reproductive Potential includes information on pregnancy testing, recommendations on the use of contraception, and the risks of infertility. Finally, when there is an open pregnancy registry for the product, contact information will be included in the Pregnancy section.
Oncology Products vs Other Products
Are the pregnancy sections of the package insert handled differently for oncology products than for products to treat other diseases?
Dr. Palmby: In many cases, products approved for the treatment of cancer pose a risk to embryo-fetal development if administered to a pregnant patient. Although some anticancer agents were previously a Pregnancy Category C or X, the majority were Category D. Under the PLLR, products will include a risk statement indicating the potential to cause embryo-fetal toxicity. Most oncology specialists are aware of the potential for embryo-fetal harm if an anticancer product is administered to a pregnant patient, which must be balanced against the potential benefit of treatment.
Products approved to treat diseases other than cancer that are teratogenic in animals may be contraindicated for use in pregnancy, especially if there are also data of adverse effects in pregnant patients or a mechanism of action indicating an increased risk of embryo-fetal toxicity.
In contrast, there are a limited number of anticancer products that are contraindicated for use in pregnancy. This is due to a number of factors, including the relatively controlled setting in which anticancer agents are used, the need to limit barriers to obtaining necessary treatments for cancer patients, the serious and life-threatening nature of the disease, the often limited therapeutic options available, and the desire to leave treatment choices in the hands of patients and their health-care professionals.
Intent of Pregnancy Labeling
Why do labels often state that a product can cause embryo-fetal toxicity when administered to a pregnant woman?
Dr. Palmby: For many oncology products, the risk of embryo-fetal toxicity is significant, based on a combination of data from pregnant animals and the product’s mechanism of action. In addition, many products used to treat cancer are dosed near a maximal tolerated dose in patients, which means that often there is little to no difference between the drug exposures achieved in patients compared with those in pregnant animals during embryo-fetal development studies. The toxicity to the developing fetus can be expected, since many of the drugs used are often designed to damage DNA or inhibit a pathway or process critical for cell proliferation or survival.
The intent of pregnancy labeling for oncology products is to convey the significant level of embryo-fetal risk of a product if it is administered to a pregnant patient while not contraindicating its use in pregnancy. Historically, this was accomplished by assigning a pregnancy Category D, whereas under the PLLR, a risk summary will be used. Some patients are diagnosed with cancer while pregnant, others become pregnant while being treated for their disease, and still other patients wish to become pregnant. Labeling a product as having the potential to cause embryo-fetal toxicity lets patients and their health-care professional choose the appropriate treatment plan given the seriousness of their disease and the need for treatment.
Risk Evaluation and Mitigation Strategies
What strategies are used to mitigate the potential embryo-fetal risk a product may have?
Dr. Palmby: For oncology products, these strategies range from product labeling to Risk Evaluation and Mitigation Strategies (REMS). For many products approved to treat cancer, including available information in the package insert is sufficient to convey the appropriate level of risk to health-care professionals. It is recognized that oncology specialists treating female patients with cancer are aware of the potential risk to a developing fetus that many anticancer products pose and have discussions with their patients about avoiding pregnancy while being treated.
In a few cases, oncology products have been contraindicated for use in pregnant patients and have associated REMS, such as thalidomide (Thalomid), lenalidominde (Revlimid), and pomalidomide (Pomalyst). Boxed warnings in the package insert are sometimes warranted for particular products. The labels for trastuzumab (Herceptin), pertuzumab (Perjeta), and ado-trastuzumab emtansine (Kadcyla) include a boxed warning due to postmarketing data-reporting effects in offspring of pregnant patients treated with trastuzumab and related mechanisms of action between these products. In addition, a large number of women of reproductive potential are likely to be treated with these products.
Does the indicated patient population matter when determining the level of FDA’s concern about a product that can cause embryo-fetal harm?
Dr. Palmby: Yes, the patient population is a large part of how FDA determines the level of risk management that is necessary for a product. Many components are considered, including animal data (eg, severity of effects, relative exposure) and mechanism of action, age range and reproductive potential of the population to be treated, the intended prescribers, whether patients have been previously treated, and the stage of the disease.
For example, the active ingredient in vismodegib (Erivedge) is a hedgehog pathway inhibitor, which caused severe malformations in offspring of pregnant rats at exposures significantly lower than those observed in patients treated with the recommended dose. Similar findings were first observed with cyclopamine, a plant alkaloid that also inhibits the hedgehog pathway and causes craniofacial deformities in the offspring of sheep grazing in fields with weeds containing this chemical. In addition, there is a large body of literature describing the critical role of the hedgehog pathway in embryonic development.
Following an extensive discussion among FDA reviewers and the applicant, it was decided that certain elements should not be included in a risk management plan for embryo-fetal toxicity due to the specific indicated population and the impact those elements may have on patient access to the product. The package insert and the medication guide put into place when vismodegib was approved were deemed sufficient for that indication, but a risk management plan would have to be reevaluated for any subsequent indications considered for approval.
Resources for Patients
How can a patient learn more about taking medications during pregnancy?
Dr. Zahalka: There is additional information in the following links to the FDA’s website that may be useful:
http://www.fda.gov/forconsumers/byaudience/forwomen/womenshealthtopics/ucm117976.htm
http://www.fda.gov/forconsumers/byaudience/forwomen/ucm118567.htm
http://www.fda.gov/drugs/developmentapprovalprocess/developmentresources/labeling/ucm093307.htm ■
Disclosure: Drs. Palmby and Zahalka reported no potential conflicts of interest.
Inside the Black Box is Guest Edited by Richard Pazdur, MD, Director of the FDA’s Office of
Hematology and Oncology Products.
