An Evaluation of the Impact of PD‐1 Pathway Blockade on Reproductive Safety of Therapeutic PD‐1 Inhibitors

Abstract

This report discusses the principles of reproductive toxicity risk assessment for biopharmaceuticals blocking the PD-1/programmed cell death ligand 1 (PD-L1) pathway, which have been developed for the treatment of patients with advanced malignancies. The PD-1/PD-L1 pathway is a T-cell co-inhibitory pathway that normally maintains immune tolerance to self. Its role in pregnancy is to maintain immune tolerance to the fetal allograft. In cancer patients, this signaling pathway is hijacked by some neoplasms to avoid immune destruction. PD-1/PD-L1-blocking agents enhance functional activity of the target lymphocytes to eventually cause immune rejection of the tumor. A therapeutic blockade of PD-1/PD-L1 pathway that occurs at full target engagement provides a unique challenge to address the risk to pregnancy because disruption of the same pathway may also reduce or abrogate maternal immune tolerance to the fetal alloantigens inherited through the father. Typically, nonclinical reproductive and developmental toxicity (DART) studies in animals (rats and rabbits) with clinical drug candidates are conducted to identify potential risk in humans and to determine exposure margin for the effects on reproduction as part of the risk assessment. However, for biopharmaceuticals for which the desired mechanism of action cannot be separated from potential deleterious effects to the fetus and when the only relevant toxicology species is nonhuman primate (NHP), the risk to reproduction can be predicted by a mechanism-based assessment using data generated from murine surrogate models as supportive information without conducting DART in NHPs. Such an approach has been used in the evaluation of pregnancy risk of anti-PD-1 agent, pembrolizumab, and has been demonstrated as an important alternative to performing DART studies in NHPs.

Case Report of a Pregnancy During Ipilimumab Therapy

“…Checkpoint blockers that affect the PD-1 pathway may have a higher risk of fetal loss or birth defects than ipilimumab as a result of the role of that pathway in protecting the placenta and fetus from attack by the mother’s immune system. 5 Given that pregnancy induces a relative immunosuppressed state that protects the fetus, melanoma arguably can progress more rapidly during pregnancy, although no conclusive evidence proves this theory. No known mechanism by which pregnancy directly influences melanoma growth exists, and the prognosis in the case of melanoma does not appear to be affected by pregnancy.…”

Immune checkpoint inhibitors in challenging populations

“…In animal studies, anti-PD1/PD-L1 clearly increased the risks of spontaneous abortions. 74,75 However, in surviving animals, no increased risks of birth defects were noted. At this time, anti-PD1 agents are categorized as pregnancy category D by the Food and Drug Administration, whereas ipilimumab is pregnancy category C (because of the less clear role of the CTLA4 axis in fetal immune tolerance).…”

“…PD1/PD‐L1 interactions appear to play a key role in maintaining fetal tolerance; indeed, the placenta is often used as a positive control for PD‐L1 expression because of its strong and ubiquitous expression. In animal studies, anti‐PD1/PD‐L1 clearly increased the risks of spontaneous abortions . However, in surviving animals, no increased risks of birth defects were noted.…”

Goldilocks’ Determination of What New In Vivo Data are “Just Right” for Different Common Drug Development Scenarios, Part 1

“…The remaining issue is to generate some “ADME confidence” to understand the relevance of potential risks based on clinical exposure, including an appreciation of the blood levels at which one might expect adverse effects. For an available target (e.g., one which can bind blood‐borne therapeutic molecules), this could be addressed by the conventionally thorough ADME characterization of Cmax and AUC in nonpregnant animals, together with an emerging estimation of the timing of fetal exposure to antibodies across species (Bowman et al., ; Moffat et al., ), but other approaches could be taken using appropriate scientific justification such as that used for Keytruda (Poulet, Wolf, Herzyk, & DeGeorge, ), which includes molecules that interfere with maternal immune tolerance of pregnancy (Prell, Halpern, & Rao, ). For a target with such characteristics used for a high‐risk tolerance indication, we suggest that these are the key data that will allow a confident risk assessment, and in vivo animal data are redundant and do not impact the risk assessment.…”