Is sunitinib a Narrow Therapeutic Index Drug? – A systematic review and in vitro toxicology-analysis of Sunitinib vs. Imatinib in cells from different tissues

Highlights

• The question if Sunitinib is a NTID or not might have impact on regulatory requirements for generic applications.

• Review of metadata of Sunitinib argues for the classification as NTID according to FDA definition.

• Cell viability assays from cells of different origin reveal higher toxicity of Sunitinib compared to the non-NTID Imatinib.

• Different protein activation patterns point to different mechanisms of action of both substances.

• Clinical investigation of the therapeutic index of Sunitinib is deemed necessary with regard to generic applications.

Abstract

Narrow Therapeutic Index Drugs (NTIDs) are characterized by a small range between therapeutic and toxicological effect. Missing international harmonized definition for NTIDs the EMA does not even have a definition of NTIDs in contrast to the U.S. FDA, Health Canada, and the Japanese NIHS. Sunitinib, a tyrosine kinase inhibitor (TKI), indicated for the treatment of certain cancer types, will be running off-patent soon. Falling into the category of NTID would have a major impact on regulatory requirements for generic applications. Our analyses of metadata revealed numerous arguments in favor of a NTID designation. We used in vitro experiments to also give initial experimental answers. Five cell types of different tissue origin were examined for determination of IC₅₀-values in cell viability assays. For comparison, the first-in-class TKI Imatinib was used as reference non-NTID drug. In addition, apoptotic proteins were investigated with respect to their expression and phosphorylation status. These in vitro experiments showed systematically higher toxicity of Sunitinib compared to Imatinib and a different expression and phosphorylation pattern of apoptotic proteins. In vitro data can only give preliminary results and further experiments with clinical blood samples and tumor biopsies are needed to finally clarify NTID status of Sunitinib.

Introduction

Targeted therapies on the molecular level like tyrosine kinase inhibitors (TKIs) have brought an enormous therapeutic advantage in treatment of certain types of cancer. The most prominent TKI Imatinib (Glivec^®) was invented in 1996 and revolutionized the treatment of chronic myeloid leukemia (CML) (Iqbal and Iqbal, 2014). The U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) approved this drug as first-in-class TKI in 2001. Imatinib is orally bioavailable and approved mainly in hematologic indications. The receptor tyrosine kinase (RTK) BCR-ABL protein is considered an ideal target for Imatinib, since the BCR-ABL mutation is present in almost all patients with CML (Arora and Scholar, 2005). Besides BCR-ABL, Imatinib has also been shown to inhibit other RTKs (Deininger et al., 2005). For a complete list of Imatinib's targets please refer to Section 3.1. Currently, Imatinib is the standard of care in CML as it has improved dramatically the outcome of this disease and the overall survival of patients (Buchdunger et al., 2000, Buchdunger et al., 1996, Okuda et al., 2001). However, indications of Imatinib are not limited to hematology. It also contributes to clinical management of solid tumors such as gastrointestinal stromal tumor (GIST) in advanced and adjuvant settings (Blay, 2009).

After Imatinib further TKIs were developed for treatment of other types of cancer.

Since 2006, Sutent^® is approved on the European and US-American market as capsule for oral administration in doses of 12.5 mg, 25 mg, and 50 mg and is a multi-targeted RTK inhibitor with anti-angiogenic and anti-tumor activities.

Sunitinib's main mode of action is inhibition of angiogenesis mediated via vascular endothelial growth factor receptors (VEGF-Rs) and platelet-derived growth factor receptors (PDGF-Rs). In addition, Sunitinib blocks the signaling pathways of other RTKs thereby inhibiting cell growth (Christensen, 2007, Yang et al., 2010). Furthermore, it can induce apoptosis of human umbilical vein endothelial cells in vitro (Mendel et al., 2003). For complete list of Sunitinib's targets please refer to Section 3.1. Sunitinib is indicated in advanced GIST after failure of Imatinib treatment and delayed the time of tumor progression on average from 1.5 up to 6.3 months and significantly reduced death rate (Demetri et al., 2006). Sunitinib is also indicated in metastatic renal cell carcinoma (mRCC) (Motzer et al., 2007, Motzer et al., 2009, Gore et al., 2009) and advanced pancreatic neuroendocrine tumors (pNET) providing patients with survival benefit (Raymond et al., 2011).

Narrow Therapeutic Index Drugs (NTIDs) are also referred to as drugs with a narrow therapeutic window. The therapeutic window is the difference between efficacy and toxicity of a drug. The wider this window is, the safer the drug (i.e. Penicillin). An additional measure is the therapeutic index (TI), which is the quotient of median lethal dose (LD₅₀) and effective dose (ED₅₀). The TI provides information on safety in the application of a drug. In NTIDs small dose changes can lead to undesired or even toxic effects. Therefore, the switch between two galenic formulations (i.e. tablets) with slightly different potency is not as harmless as in non-NTIDs. Whenever a given drug substance is designated as NTID, applying for a marketing authorization with the same drug substance for a generic NTID, clinical data are mandatory (i.e. a bioequivalence study). A waiver based on the biopharmaceutical classification system is not possible in NTIDs (BCS-based biowaiver).

Handling of NTIDs in a global context is highly divergent. Thus, a systematic search was conducted on regulatory requirements of acceptance criteria and designated NTIDs in regulatory agencies (Table 1).

In contrast to the European Union, the U.S., Canada, and Japan have implemented a definition of NTIDs. In the U.S. NTIDs are called narrow therapeutic ratio drugs (NTRDs; Canada: critical dose drugs, CDD; Japan, narrow therapeutic range drugs) (Table 1).

The Center of Drug Evaluation and Research (CDER) of the FDA published a list of NTIDs that among others consist of Levoxyine, Digoxin, and Warfarin in 1988.

However, in the Guidance for Industry “Bioavailability and Bioequivalence Studies for Orally Administered Drug Products — General Considerations” (FDA, 2003) the FDA gives the following definition of NTRDs: “… drug products as containing certain drug substances subject to therapeutic drug concentration or pharmacodynamic monitoring, and/or where product labeling indicates a narrow therapeutic range designation. Examples include digoxin, lithium, phenytoin, theophylline, and warfarin.”

Since concerns remained whether the bioequivalence and quality standards for NTIDs are sufficient, the Applied Clinical Pharmacology Services (ACPS) held a meeting on NTIDs and suggested that the FDA should file a list of NTIDs and develop criteria for regarding a given drug an NTRD. Furthermore, FDA decided that the standard acceptance criteria on bioequivalence are not sufficient for NTIDs.

In July 2011 the FDA held a meeting only to discuss the following two topics (FDA, 2011b):

• Bioequivalence and Quality Standards for Narrow Therapeutic Index Drug Products.

• Impact of Formulation and Quality on the Safety and Performance of Generic Drug Products.

At that moment the same bioequivalence limit acceptance criteria for “normal” generic drugs and NTIDs were valid. Namely, the confidence interval of AUC and C_max between test product and reference product falls within limits of 80–125%. At the same time Health Canada tightened their acceptance limits to 90–112%.

The result of the meeting was that the FDA formed “proposed approaches for NTIs” and that they “proposed approaches for pharmaceutical quality evaluation of NTIs” (FDA, 2011b). The FDA proposed that sponsors for Abbreviated New Drug Applications (ANDAs, the U.S. pendent to a generic application in Europe) should conduct a “replicate design study to quantify the variability of both the test and reference products and use a scaling approach for determination of bioequivalence”. Different molecular variations should be submitted as New Drug Application (NDA), such as esters or ethers of an active substance. Furthermore, FDA recommended that the bioequivalence limits should be tightened to 90–111%. FDA also added a remarkable acceptance criterion on definition of bioequivalence: the 90% confidence interval of test and reference AUC and C_max ratio must include 100% (a criterion that regularly is not fulfilled in Europe). FDA tightened the potency limit of NTIDs to 95–105%. According to the FDA, applicants of a NDA or an ANDA should “report detailed content uniformity data in the annual report” (FDA, 2011b).

TKIs are considered less toxic than classical cytotoxic drugs, because they predominantly should affect cancer cells. However, TKIs also have numerous, severe side effects eventually entailed with fatal outcome and are also diverse in their toxic potency and TI. Sunitinib led to severe toxic side effects, especially hepatotoxicity (labeled by black box warning in the US PI), renal failure, heart failure, pulmonary embolism, gastrointestinal perforation, and hemorrhages (EMA, 2015b). Severe toxic side effects are often temporarily limited, however multiple, low grade toxicities are usually of longer duration and thus are similarly challenging for physicians and patients. One such example of a commonly observed adverse event with Sunitinib (up to 60% of patients) is hand-foot skin reaction (HFSR), which can lead to dose adjustments and treatment interruptions when not appropriately managed (Manchen et al., 2011, Yeh et al., 2014). Although the receptor inhibition profiles of Sunitinib and Imatinib have some overlap there are major differences and different toxic potency can be anticipated. Sunitinib runs out of patent soon and numerous generic applications are expected. The question of whether Sunitinib may potentially be a NTID is therefore of major impact.

In order to give initial answers whether a NTID designation of Sunitinib is supported or not, we combined a systematic literature search with in vitro laboratory data comparing Sunitinib to Imatinib. Imatinib was chosen as a comparator as it is a TKI which definitely does not fulfill the criteria as NTID (Table 1) because there is a more than 2-fold difference in the minimum toxic concentrations and minimum effective concentrations in the blood (also see Section 3.1. and FDA, 2011a, Eckstein et al., 2014). We performed a systematic search in PubMed and compared pharmacodynamic and pharmacokinetic parameters of both TKIs. Second, the cytotoxic and apoptotic potency of Sunitinib in comparison to Imatinib is considered to provide relevant experimental evidence for a different toxic potency of both TKIs.

We analyzed systematically whether Sunitinib in three continuous cell lines (HepG2, HaCat, and HEK293) and two different primary cell types (human keratinocytes and dermal fibroblasts from either plastic surgery or dermal excision) is more toxic than Imatinib. The different cell lines chosen should reflect different target tissues of important TKI side effects (hepatic, renal, skin). In respect to molecular mechanism of action that may be different between the two substances in different tissues, we performed a systematic comparison of the expression pattern of 35 proteins involved in apoptotic pathways after treatment with Sunitinib and Imatinib, respectively.