A Comprehensive Monograph on Tivantinib (ARQ-197): From a Promising c-MET Inhibitor to a Discontinued Antineoplastic Agent

Executive Summary

Tivantinib, also known by its development code ARQ-197, is an orally bioavailable small molecule that entered clinical development as a highly selective, non-ATP-competitive inhibitor of the c-MET receptor tyrosine kinase.[1] The c-MET pathway, a critical mediator of cell growth, survival, and metastasis, is frequently dysregulated in human cancers, making it a high-priority target for therapeutic intervention.[2] Tivantinib's unique mechanism, which involved binding to the inactive conformation of the kinase, initially positioned it as a novel and promising agent in the field of targeted oncology.[3]

However, the trajectory of Tivantinib's development is a cautionary tale in pharmaceutical research, defined by a fundamental and consequential misinterpretation of its primary mechanism of action. As the drug progressed, a compelling body of evidence emerged demonstrating that its principal cytotoxic activity did not stem from c-MET inhibition. Instead, Tivantinib functions as a potent antimitotic agent by disrupting microtubule polymerization, a mechanism entirely independent of the c-MET pathway.[5] This critical disconnect between the drug's presumed target and its actual biological function created a foundational flaw in its clinical development strategy.

An ambitious clinical program was launched to evaluate Tivantinib across a broad spectrum of solid tumors, most notably in hepatocellular carcinoma (HCC) and non-small cell lung cancer (NSCLC).[2] Encouraging signals from a Phase 2 trial in HCC, which suggested a significant survival benefit in a subgroup of patients with high c-MET expression, prompted the initiation of large, biomarker-driven Phase 3 studies.[10] These pivotal trials, including METIV-HCC for liver cancer and MARQUEE for lung cancer, were designed with the assumption that Tivantinib was a targeted MET inhibitor.

Ultimately, these large-scale studies failed to meet their primary endpoints of improving overall survival.[12] The promising efficacy signal observed in the small MET-high subgroup did not translate to the larger, prospectively evaluated patient populations. These definitive failures, coupled with significant safety concerns, led to the global discontinuation of Tivantinib's development by its partners, including ArQule, Daiichi Sankyo, and Kyowa Hakko Kirin, around 2017.[15] The story of Tivantinib serves as a critical case study on the paramount importance of rigorous mechanism of action validation in the preclinical stage and illustrates the perils of building a clinical biomarker strategy that is misaligned with a drug's true pharmacologic activity.

Chemical Identity and Physicochemical Properties

A comprehensive understanding of an investigational drug begins with its precise chemical and physical characterization. This section provides a definitive dossier of Tivantinib's nomenclature, structural identifiers, and key physicochemical properties, which collectively define the molecule and influence its pharmacological behavior.

Nomenclature and Identifiers

Tivantinib has been designated by several names and unique identifiers across various chemical and biomedical databases, ensuring unambiguous reference in scientific literature and regulatory filings.

• Generic and Common Names: The officially recognized International Nonproprietary Name (INN) and United States Adopted Name (USAN) is Tivantinib. During its development, it was primarily referred to by the code ARQ-197 or ARQ 197.[2]
• Systematic (IUPAC) Name: The chemically precise name, which describes its structure and stereochemistry, is (3R,4R)-3-(5,6-dihydro-4H-pyrrolo[3,2,1-ij]quinolin-1-yl)-4-(1H-indol-3-yl)pyrrolidine-2,5-dione.[1] The (3R,4R) designation specifies the absolute configuration at the two chiral centers on the pyrrolidine-2,5-dione ring, which is critical for its biological activity. Other stereoisomers, such as the (3S,4S) form, exist but are not the active pharmaceutical ingredient.[18]
• Registry and Database IDs:
• CAS Number: The primary Chemical Abstracts Service registry number is 905854-02-6.[1]
• DrugBank ID: DB12200.[1]
• PubChem Compound ID (CID): 11494412.[22]
• ChEMBL ID: CHEMBL2103882.[1]
• FDA UNII (Unique Ingredient Identifier): PJ4H73IL17.[1]

Molecular Structure and Formula

The molecular formula and structural representations provide a blueprint of the molecule's atomic composition and connectivity.

• Molecular Formula: The empirical formula for Tivantinib is C23​H19​N3​O2​.[1]
• Structural Representations:
• SMILES (Simplified Molecular Input Line Entry System): C1CC2=C3C(=CC=C2)C(=CN3C1)[C@H]4[C@@H](C(=O)NC4=O)C5=CNC6=CC=CC=C65. The @ and @@ notations define the specific (3R,4R) stereochemistry.[1]
• InChI (International Chemical Identifier): InChI=1S/C23H19N3O2/c27-22-19(16-11-24-18-9-2-1-7-14(16)18)20(23(28)25-22)17-12-26-10-4-6-13-5-3-8-15(17)21(13)26/h1-3,5,7-9,11-12,19-20,24H,4,6,10H2,(H,25,27,28)/t19-,20-/m0/s1.[1]
• InChIKey: UCEQXRCJXIVODC-PMACEKPBSA-N.[1]

Physicochemical Characteristics

These properties govern the drug's behavior in biological systems, including its solubility, membrane permeability, and potential for oral absorption.

• Molecular Weight: The average molecular weight is 369.424 g/mol, and the monoisotopic mass is 369.147726864 g/mol.[2]
• Physical State: At room temperature, Tivantinib is a crystalline solid.[20]
• Solubility: Tivantinib is characterized by very poor aqueous solubility, with a predicted value of 0.00745 mg/mL.[2] It is, however, soluble in various organic solvents, including dimethyl sulfoxide (DMSO) at concentrations of 20 mg/mL or higher, dimethylformamide (DMF) at 20 mg/mL, and ethanol at approximately 5 mg/mL.[6] This low aqueous solubility is a common challenge for oral drug formulation.
• Lipophilicity: The partition coefficient (logP), a measure of lipophilicity, has predicted values ranging from 3.27 to 4.04, indicating that Tivantinib is a lipophilic molecule capable of crossing cell membranes.[2]
• Ionization Constant (pKa): The predicted strongest acidic pKa is approximately 9.72, which corresponds to the proton on the imide nitrogen of the pyrrolidine-2,5-dione ring.[2]
• Drug-likeness Properties: Tivantinib generally adheres to Lipinski's Rule of Five, a set of guidelines used to evaluate the potential for a compound to be an orally active drug. It has a molecular weight under 500 Da, a logP value under 5, two hydrogen bond donors, and three to five hydrogen bond acceptors (depending on the calculation method), with zero violations of the rule.[22] These properties suggested a favorable profile for oral bioavailability.

Table 1: Tivantinib Identifiers and Physicochemical Properties
Property	Value	Source(s)
IUPAC Name	(3R,4R)-3-(5,6-dihydro-4H-pyrrolo[3,2,1-ij]quinolin-1-yl)-4-(1H-indol-3-yl)pyrrolidine-2,5-dione	1
CAS Number	905854-02-6	1
DrugBank ID	DB12200	1
Molecular Formula	C23​H19​N3​O2​	1
Average Molecular Weight	369.424 g/mol	2
SMILES	C1CC2=C3C(=CC=C2)C(=CN3C1)[C@H]4[C@@H](C(=O)NC4=O)C5=CNC6=CC=CC=C65	1
InChIKey	UCEQXRCJXIVODC-PMACEKPBSA-N	1
Physical State	Crystalline Solid	20
Predicted Water Solubility	0.00745 mg/mL	2
Predicted logP	3.27 - 4.04	2
Hydrogen Bond Donors	2	2
Hydrogen Bond Acceptors	2 - 3	2

Preclinical Pharmacology and Dual Mechanism of Action

The scientific narrative of Tivantinib is dominated by the evolution in understanding its mechanism of action. Initially lauded as a highly specific targeted agent, subsequent investigations revealed a more complex and ultimately contradictory pharmacological profile. This discrepancy between the proposed mechanism and the actual cellular activity is central to explaining the trajectory of its clinical development and eventual failure.

The c-MET Inhibition Hypothesis

Tivantinib was first identified and advanced into clinical trials based on a compelling preclinical profile as a selective inhibitor of the c-MET receptor tyrosine kinase.[1] The hepatocyte growth factor (HGF)/c-MET signaling axis is a well-validated oncogenic pathway, and its dysregulation through overexpression, amplification, or mutation is a known driver of tumor cell proliferation, survival, angiogenesis, invasion, and metastasis.[2]

• A Unique Binding Mechanism: What distinguished Tivantinib from other kinase inhibitors was its proposed non-ATP-competitive mechanism of action.[4] Most kinase inhibitors compete directly with adenosine triphosphate (ATP) for binding in the enzyme's active site. In contrast, crystallographic studies revealed that Tivantinib binds preferentially to the inactive, dephosphorylated conformation of the c-MET kinase.[3] By binding to this inactive state, Tivantinib was thought to stabilize the kinase in a conformation that prevents its autophosphorylation and subsequent activation, even in the presence of its ligand, HGF.[3] This allosteric-like mechanism was hypothesized to confer greater selectivity compared to traditional ATP-competitive inhibitors.[29]
• Selectivity and Potency: In cell-free enzymatic assays, Tivantinib demonstrated an inhibitory constant (Ki​) of 355 nM against the c-MET kinase.[18] Profiling against a broad panel of over 229 other human kinases demonstrated high selectivity, with only a few other kinases being weakly inhibited at much higher concentrations.[3] This profile supported the narrative of Tivantinib as a precisely targeted agent against the c-MET pathway.

The Emergent Role as a Microtubule Disruptor

Despite the strong in vitro biochemical data, conflicting evidence began to emerge from cell-based assays that questioned the exclusivity of the c-MET inhibition hypothesis.

• Contradictory Cytotoxicity Data: A key early observation was Tivantinib's broad cytotoxic activity across a wide panel of human tumor cell lines, with an effective concentration (EC50​) in the range of 300-600 nmol/L.[5] Crucially, this cytotoxicity was observed in cell lines that were not dependent on c-MET signaling for their survival and, most tellingly, in cell lines that did not express the MET protein at all.[5] This indiscriminate killing of cancer cells, regardless of target expression, is a hallmark of a conventional cytotoxic agent rather than a targeted therapy and was fundamentally inconsistent with the biology of HGF/MET.[5]
• Definitive Mechanistic Studies: More rigorous follow-up studies provided a definitive alternative mechanism. It was demonstrated that Tivantinib inhibits tubulin polymerization in a concentration-dependent manner in cell-free assays.[6] In cellular models, treatment with Tivantinib led to profound disruption of microtubule dynamics, resulting in a G2/M phase cell cycle arrest and subsequent induction of apoptosis.[5] These are the classic cellular consequences of an antimitotic agent, similar to well-known chemotherapies like the vinca alkaloids or taxanes.
• Lack of In-Cell MET Inhibition: The most compelling evidence refuting the original hypothesis came from experiments that directly measured MET activity within cancer cells. In multiple cell lines, including those with constitutive MET activation or those stimulated with HGF, Tivantinib failed to inhibit MET autophosphorylation, even at concentrations well above those required for its cytotoxic effects.[5] This finding directly contradicted the central premise of its development, indicating that while Tivantinib could bind to the purified, dephosphorylated kinase in a test tube, it was ineffective at inhibiting the active kinase in a complex cellular environment.

Synthesis of Mechanistic Understanding

The clinical development of Tivantinib proceeded for years under the paradigm that it was a targeted c-MET inhibitor. This assumption directly informed the clinical trial design, which included a biomarker strategy to select patients with tumors exhibiting high levels of MET expression, as these patients were hypothesized to be most likely to respond.[10] However, the overwhelming preclinical evidence that emerged later demonstrated that Tivantinib's dominant, clinically relevant mechanism of action is antimitotic activity via microtubule disruption—an effect that is entirely independent of a tumor's MET status.

This created a fatal disconnect at the heart of the development program. The drug was being evaluated using a precision medicine approach, yet its biological activity was that of a non-targeted cytotoxic agent. The initial preclinical claim of selective c-MET inhibition led to a logical, but ultimately flawed, clinical strategy. The failure to rigorously validate the drug's mechanism of action in relevant cellular models before committing to a biomarker strategy meant that the patient selection criteria were misaligned with the drug's true biological function. The promising signal seen in the small MET-high cohort of the Phase 2 HCC trial was likely a spurious correlation or a statistical anomaly, which could not be replicated in larger, more robust Phase 3 studies. Consequently, the failure of Tivantinib in the clinic should not be interpreted as a failure of c-MET as a therapeutic target. Rather, it represents a failure of preclinical validation that led to a fundamentally misguided clinical development strategy.[5]

Pharmacokinetics and Drug Metabolism (ADME)

The absorption, distribution, metabolism, and excretion (ADME) profile of a drug dictates its exposure, efficacy, and safety in patients. The pharmacokinetic properties of Tivantinib revealed significant complexities, including non-linear exposure and impactful genetic variability, which likely contributed to the challenges observed in its clinical development.

Absorption, Distribution, and Bioavailability

Tivantinib was developed as an orally administered agent, and its physical properties suggested good potential for absorption.[1]

• Administration and Absorption: As an orally bioavailable small molecule, Tivantinib is absorbed through the gastrointestinal tract.[1] Following a single oral dose, peak plasma concentrations ( Tmax​) are typically reached between 2 and 4 hours.[31] Studies with radiolabeled Tivantinib indicated that absorption is nearly complete, particularly under fed conditions.[32]
• Pharmacokinetic Parameters: Phase 1 studies characterized Tivantinib's pharmacokinetic profile. At a dose of 360 mg twice daily (BID), the maximum plasma concentration (Cmax​) was approximately 1459 ng/mL, and the area under the concentration-time curve (AUC) was 8257 ng·h/mL.[33] However, a key finding was that increases in drug exposure were not proportional to increases in dose, suggesting complex absorption or clearance mechanisms.[31] Furthermore, significant interpatient variability in pharmacokinetic parameters was consistently observed across studies.[31] The drug has a relatively short terminal elimination half-life ( T1/2​) of approximately 2.7 hours, necessitating a BID dosing schedule to maintain therapeutic concentrations.[33]

Metabolism and Excretion

Tivantinib undergoes extensive biotransformation in the body before it is eliminated.

• Metabolic Pathways: The parent Tivantinib molecule accounts for no more than one-third of the total drug-related radioactivity in plasma, indicating that its metabolites are major circulating components.[32] The primary metabolic pathways identified are oxidation, particularly hydroxylation, and glucuronidation.[32]
• Key Metabolizing Enzymes: The metabolism of Tivantinib is primarily mediated by the cytochrome P450 (CYP) enzyme system. Specifically, CYP2C19 and CYP3A4 have been identified as the major enzymes responsible for its biotransformation.[3]
• Major Metabolites: In humans, the two most abundant circulating metabolites are designated M4 and M5. Both are products of hydroxylation at the benzyl position of the molecule's tricyclic ring structure.[32] Additional metabolites, such as M4845, M4846, M5991, and M6385, have also been characterized.[29] Current evidence suggests that these metabolites do not significantly contribute to the efficacy or toxicity of Tivantinib.[32]
• Excretion: The primary route of elimination for Tivantinib and its metabolites is through fecal excretion, which occurs following secretion into the bile. Only trace amounts of the unchanged parent drug are found in urine, feces, or bile, confirming that it is almost entirely metabolized before being cleared from the body.[32]

Pharmacogenomics and Drug Interaction Potential

The reliance on specific CYP enzymes for metabolism introduces significant potential for variability due to genetic factors and co-administered drugs.

• The Overlooked Impact of CYP2C19 Polymorphisms: A critical finding from a Phase 1 study was the observation of a patient who exhibited significantly higher drug exposure and more severe adverse events than the rest of the cohort.[4] Genetic analysis revealed this individual carried the CYP2C19*2 polymorphism, a well-known loss-of-function allele that results in a "poor metabolizer" phenotype.[4] The prevalence of CYP2C19 poor metabolizer alleles varies significantly among ethnic populations, being most common in individuals of Asian descent (up to 35%), followed by those of African (17%) and Caucasian (15%) descent.[4] This genetic variability was likely a major, unmanaged contributor to the high interpatient pharmacokinetic variability and inconsistent safety profile seen in clinical trials. The increased incidence of severe toxicities, such as the fatal interstitial lung disease (ILD) observed in the Japanese ATTENTION trial, may be at least partially attributable to the higher frequency of poor metabolizers in that population, leading to systemic overexposure to the drug.[35] The failure to prospectively account for this predictable pharmacogenomic factor with dose adjustments or patient stratification represents a significant oversight in the clinical development strategy.
• Drug-Drug Interaction Potential: Nonclinical studies raised concerns about Tivantinib's potential to interact with other drugs. At clinically relevant concentrations, Tivantinib was shown to inhibit multiple CYP enzymes (CYP3A4, CYP2C19, CYP2C9, CYP1A) as well as the important drug efflux transporter P-glycoprotein (P-gp).[37] This profile necessitated dedicated clinical drug-drug interaction studies to provide guidance on the safe co-administration of Tivantinib with other medications metabolized or transported by these pathways.[37]

Clinical Development and Efficacy Analysis

Tivantinib was the subject of an extensive and ambitious clinical development program that spanned multiple phases and a wide array of malignancies. Initially buoyed by a promising, albeit flawed, preclinical rationale and an encouraging Phase 2 signal, the program ultimately culminated in a series of definitive Phase 3 failures that led to its termination.

Overview of the Clinical Trial Program

Tivantinib was evaluated in numerous Phase 1, 2, and 3 clinical trials as both a monotherapy and in combination with other anticancer agents.[1] The investigations covered a broad range of advanced solid tumors, including hepatocellular carcinoma (HCC), non-small cell lung cancer (NSCLC), colorectal cancer, pancreatic cancer, metastatic breast cancer, prostate cancer, and gastric cancer.[9] Key combination partners included the EGFR inhibitor erlotinib, the multi-kinase inhibitor sorafenib, and the anti-angiogenic agent bevacizumab.[8]

Table 2: Summary of Pivotal Phase 3 Clinical Trials for Tivantinib
Trial Name / NCT ID	Phase	Indication & Line of Therapy	Treatment Arms	Primary Endpoint	Key Outcome	Status	Source(s)
METIV-HCC (NCT01755767)	3	HCC (MET-High), 2nd Line	Tivantinib vs. Placebo	Overall Survival (OS)	No improvement in OS. Median OS: 8.4m (Tiv) vs. 9.1m (Pbo). HR=0.97, p=0.81	Completed	12
JET-HCC (NCT02029157)	3	HCC (MET-High), 2nd Line (Japanese population)	Tivantinib vs. Placebo	Progression-Free Survival (PFS)	Failed to show survival benefit.	Completed	9
MARQUEE (NCT01244191)	3	NSCLC (non-squamous), 2nd/3rd Line	Tivantinib + Erlotinib vs. Placebo + Erlotinib	Overall Survival (OS)	No improvement in OS. Median OS: 8.5m (Tiv) vs. 7.8m (Pbo). HR=0.98. Stopped for futility.	Terminated	14
ATTENTION (NCT01377376)	3	NSCLC (WT-EGFR), 2nd/3rd Line (Asian population)	Tivantinib + Erlotinib vs. Placebo + Erlotinib	Overall Survival (OS)	No significant improvement in OS. Median OS: 12.7m (Tiv) vs. 11.1m (Pbo). HR=0.891. Stopped for toxicity.	Terminated	35

Hepatocellular Carcinoma (HCC): The Rise and Fall of a Biomarker Strategy

The development of Tivantinib in HCC is a paradigmatic example of how a promising signal from a small, subgroup analysis can fail to be validated in larger studies.

• Phase 2 Promise: A randomized, placebo-controlled Phase 2 study provided the critical impetus for the HCC program. In this trial of 107 patients with advanced HCC who had progressed on prior therapy, a pre-planned analysis was conducted based on the MET expression status of tumors, as determined by immunohistochemistry (IHC). While the overall population showed a modest benefit, the subgroup of patients with "MET-high" tumors (defined as ≥50% of tumor cells with moderate or strong staining) demonstrated a striking and statistically significant improvement in overall survival. In this MET-high cohort, the median OS was 7.2 months for patients treated with Tivantinib compared to just 3.8 months for those on placebo (Hazard Ratio: 0.38; p=0.01).[10] This result was interpreted as strong validation for both Tivantinib's mechanism and the MET-high biomarker, justifying a large-scale, biomarker-selected Phase 3 program.
• Phase 3 Failure (METIV-HCC & JET-HCC): The subsequent Phase 3 trials were designed to confirm the Phase 2 findings in a larger, global population of MET-high HCC patients.
• The METIV-HCC study (NCT01755767) was a randomized, double-blind, placebo-controlled trial that enrolled 340 patients with MET-high, inoperable HCC who had been previously treated with systemic therapy.[12] The study definitively failed to meet its primary endpoint of improving overall survival. The median OS was 8.4 months in the Tivantinib arm versus 9.1 months in the placebo arm (HR=0.97, p=0.81). There was likewise no benefit in the secondary endpoint of progression-free survival (PFS), with a median of 2.1 months for Tivantinib versus 2.0 months for placebo (HR=0.96).[13]
• The JET-HCC study (NCT02029157), a similar Phase 3 trial conducted in a Japanese population, also failed to demonstrate a survival benefit for Tivantinib.[9]
• Following these unambiguous negative results, the development partners announced the termination of the HCC program, and Kyowa Hakko Kirin ceased all development of Tivantinib in its Asian territories.[12]

Non-Small Cell Lung Cancer (NSCLC): A Failed Combination Strategy

In NSCLC, the clinical strategy focused on combining Tivantinib with the EGFR tyrosine kinase inhibitor (TKI) erlotinib. The rationale was based on preclinical data suggesting potential synergy and results from a Phase 2 study that hinted at improved outcomes, particularly in patients with KRAS-mutant or EGFR wild-type tumors.[9]

• Phase 3 Failure (MARQUEE & ATTENTION): Two large Phase 3 trials were initiated to confirm the efficacy of the combination.
• The MARQUEE study (NCT01244191) was a global trial that randomized 1,048 patients with previously treated, non-squamous NSCLC to receive either Tivantinib plus erlotinib or placebo plus erlotinib.[14] An interim analysis revealed that the study was unlikely to meet its primary endpoint, and it was subsequently stopped early for futility. While the combination did result in a statistically significant improvement in PFS (median 3.6 months vs. 1.9 months; HR=0.74; p < 0.001), it conferred no improvement in the primary endpoint of overall survival (median 8.5 months vs. 7.8 months; HR=0.98).[14]
• The ATTENTION study (NCT01377376) was a Phase 3 trial conducted in a similar patient population in Asia (EGFR wild-type).[35] This study was terminated prematurely due to a significant safety concern: an unacceptable and imbalanced incidence of interstitial lung disease (ILD) in the Tivantinib arm.[36] Although the study was underpowered due to early termination, the available data showed no statistically significant benefit in OS (median 12.7 months vs. 11.1 months; HR=0.891).[36]

Investigations in Other Malignancies

Beyond HCC and NSCLC, Tivantinib was evaluated in a multitude of other solid tumors in Phase 1 and 2 settings. These included studies in metastatic castration-resistant prostate cancer, metastatic colorectal cancer, pancreatic adenocarcinoma (NCT00558207), triple-negative breast cancer (NCT01575522), and malignant mesothelioma (NCT01861301).[2] Across all these indications, Tivantinib failed to demonstrate a level of clinical activity sufficient to justify advancement into later-stage trials, and many of these studies were terminated without proceeding further.[2]

Safety and Tolerability Profile

The safety profile of Tivantinib was a significant factor in its clinical development, characterized by notable hematologic toxicities that often required dose modifications and, in one pivotal trial, a severe pulmonary toxicity that led to its premature termination. The pattern of adverse events also provided important, albeit underappreciated, clues to the drug's true mechanism of action.

Common and Dose-Limiting Adverse Events

Across the extensive clinical program, a consistent pattern of adverse events (AEs) was observed.

• Most Frequent Adverse Events: The most commonly reported AEs of any grade were generally gastrointestinal and constitutional, including fatigue, nausea, diarrhea, decreased appetite (anorexia), and rash.[4]
• Hematologic Toxicity: The most clinically significant and frequent dose-limiting toxicities were hematologic in nature. Grade 3 or higher neutropenia, anemia, and leukopenia were consistently observed at higher rates in Tivantinib-treated patients compared to control arms.[3] Febrile neutropenia, a serious complication of severe neutropenia, was also reported.[4] This pronounced myelosuppression frequently necessitated dose reductions, with the initial dose of 360 mg BID often being lowered to 240 mg BID or even 120 mg BID in later trials to improve tolerability.[13] A meta-analysis confirmed that Tivantinib significantly increased the risk of Grade ≥3 anemia (Relative Risk 2.15) and neutropenia (RR 5.31).[48]

Table 3: Frequency of Key Grade ≥3 Adverse Events in Pivotal Trials
Adverse Event	Trial	Tivantinib Arm (%)	Control Arm (%)	Source(s)
Ascites	METIV-HCC	7.1	5.3	13
Anemia	METIV-HCC	4.9	3.5	13
Neutropenia	ATTENTION	24.3	Not Applicable (Erlotinib only)	36
Febrile Neutropenia	ATTENTION	13.8	Not Applicable (Erlotinib only)	36
Interstitial Lung Disease (ILD)	ATTENTION	9.2 (14 cases)	3.9 (6 cases)	35

Serious Adverse Events and Risk Management

Beyond the common toxicities, several serious adverse events emerged that posed significant risks to patients and ultimately impacted the viability of the drug.

• Interstitial Lung Disease (ILD): The most alarming safety signal was the increased incidence of ILD, a potentially fatal inflammation of the lungs. This risk was particularly pronounced in the Asian patient population of the ATTENTION trial. The study was terminated prematurely by its Safety Review Committee after observing 14 cases of ILD, including 3 deaths, in the Tivantinib arm, compared to 6 non-fatal cases in the placebo arm.[35] This finding highlighted a critical regional and likely pharmacogenomic-related risk profile.
• General Deterioration and Increased Mortality: In the METIV-HCC trial, Grade ≥3 AEs were reported at similar overall rates in both arms (55.6% in the Tivantinib arm vs. 55.3% in the placebo arm). However, concerning signals included a higher incidence of ascites and "general deterioration" in the Tivantinib group. Most notably, the rate of deaths occurring within 30 days of the last dose of study drug was higher in the Tivantinib arm (22.1%) compared to the placebo arm (15.8%), with the most common causes being general deterioration and hepatic failure.[13]

The prominent and dose-limiting hematologic toxicities observed with Tivantinib are the classic safety signature of an antimitotic agent. Drugs that interfere with microtubule function and cell division, such as taxanes and vinca alkaloids, are well-known to cause myelosuppression because they disrupt the proliferation of rapidly dividing hematopoietic progenitor cells in the bone marrow. While early interpretations suggested that neutropenia might be an "on-target" effect of c-MET inhibition on hematopoiesis [4], this explanation became less tenable in light of the drug's broader mechanistic profile. In retrospect, the clinical safety data provided strong corroborating evidence for the microtubule disruption mechanism. This pattern of toxicity should have served as a significant clinical clue, prompting a more critical interrogation of the prevailing c-MET inhibitor hypothesis earlier in the drug's development.

Synthesis and Concluding Analysis

The development and ultimate failure of Tivantinib provide a compelling and instructive case study in modern oncology drug development. The synthesis of its preclinical pharmacology, pharmacokinetic profile, and extensive clinical trial data reveals a narrative defined by a flawed foundational premise, the peril of overinterpreting early clinical signals, and the critical importance of integrating all available data streams to form a coherent understanding of a drug's behavior.

The Failure of the Biomarker Strategy

The cornerstone of the late-stage Tivantinib clinical program was a biomarker-driven strategy focused on enriching for patients with MET-high tumors. This approach was logically derived from the initial hypothesis that Tivantinib was a selective c-MET inhibitor. However, this strategy was destined to fail because it was fundamentally misaligned with the drug's primary mechanism of action. The overwhelming evidence indicates that Tivantinib's clinically relevant activity is that of an antimitotic agent, driven by microtubule disruption, which has no established biological correlation with tumor MET expression levels.

The striking survival benefit observed in the small MET-high subgroup of the Phase 2 HCC trial appears, in retrospect, to have been a statistical anomaly. Such spurious signals are a known risk in small, retrospective subgroup analyses and must be interpreted with extreme caution. When this biomarker strategy was tested prospectively in large, robust Phase 3 trials like METIV-HCC, the effect vanished completely, leading to the program's failure. This outcome underscores a critical lesson: a biomarker strategy is only as valid as the mechanistic hypothesis upon which it is built.

Tivantinib in the Context of Modern c-MET Inhibitors

It is crucial to recognize that the failure of Tivantinib does not invalidate c-MET as a therapeutic target in oncology. The clinical success of other, mechanistically distinct c-MET inhibitors provides clear validation for the pathway's importance.

• Comparison with Approved Agents: Approved drugs such as Cabozantinib, Crizotinib, and Capmatinib have demonstrated meaningful clinical benefit in patient populations with MET-driven malignancies.[50] Unlike Tivantinib, these agents are typically ATP-competitive inhibitors that have been shown to potently and effectively inhibit MET signaling in both cellular and clinical settings.[50] Crizotinib (which also targets ALK and ROS1) and Capmatinib have shown significant efficacy in NSCLC patients whose tumors harbor MET exon 14 skipping mutations, a clear MET-addicted state.[51] Cabozantinib, a multi-kinase inhibitor that targets MET, VEGFR, and AXL, is approved for renal cell carcinoma and HCC, where these pathways are relevant.[51]
• Conclusion: The success of these other drugs confirms that c-MET is a druggable and clinically relevant oncogenic driver. Tivantinib's failure was not a failure of the target but a failure of a specific molecule whose development was predicated on a misunderstanding of its pharmacology.[5]

Final Status and Lessons for Oncology Drug Development

Following the definitive negative results from its pivotal Phase 3 trials in both HCC and NSCLC, the global development of Tivantinib was officially discontinued by all corporate partners—ArQule, Daiichi Sankyo, and Kyowa Hakko Kirin—around 2017.[12] While the drug is no longer being pursued for major cancer indications, its story leaves behind several enduring lessons for the field of oncology drug development.

1. Mechanism of Action is Paramount: The Tivantinib case is a stark reminder that a drug's true biological mechanism must be rigorously and unequivocally established in relevant preclinical models before embarking on costly, biomarker-driven clinical trials. Overreliance on in vitro biochemical data without thorough validation in cellular systems can lead to fundamentally flawed development strategies.
2. Beware of Spurious Subgroup Signals: Promising efficacy signals that emerge from small, retrospectively defined subgroups in early-phase trials are notoriously difficult to replicate. Such findings should be considered hypothesis-generating and must be prospectively validated in adequately powered, well-controlled studies before being used as the basis for a pivotal trial design.
3. Integrate All Data Streams: The clinical safety profile of Tivantinib (prominent hematologic toxicity) and its preclinical cellular activity (broad, non-MET-dependent cytotoxicity) were both early and persistent clues that it was not behaving like a typical targeted MET inhibitor. A more integrated and critical analysis of all available data—preclinical, pharmacokinetic, and clinical safety—could have challenged the prevailing hypothesis much sooner in the development process.
4. The Importance of Pharmacogenomics: The significant impact of CYP2C19 polymorphisms on Tivantinib's exposure and toxicity highlights the need to proactively investigate and manage genetic factors that influence drug metabolism and disposition. For drugs with a narrow therapeutic index, particularly those metabolized by highly polymorphic enzymes, a pharmacogenomic strategy should be an integral part of clinical development from the earliest phases.

Works cited

1. Tivantinib | C23H19N3O2 | CID 11494412 - PubChem, accessed August 26, 2025, https://pubchem.ncbi.nlm.nih.gov/compound/Tivantinib
2. Tivantinib: Uses, Interactions, Mechanism of Action | DrugBank Online, accessed August 26, 2025, https://go.drugbank.com/drugs/DB12200
3. Selective Inhibitor of the c-Met Receptor Tyrosine Kinase in Advanced Hepatocellular Carcinoma: No Beneficial Effect With the Use of Tivantinib? - PubMed Central, accessed August 26, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC8600564/
4. Development of Tivantinib as Treatment for Hepatocellular Carcinoma - PMC, accessed August 26, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC4521271/
5. Tivantinib (ARQ197) Displays Cytotoxic Activity That Is Independent of Its Ability to Bind MET, accessed August 26, 2025, https://aacrjournals.org/clincancerres/article/19/9/2381/78118/Tivantinib-ARQ197-Displays-Cytotoxic-Activity-That
6. Tivantinib (ARQ 197, CAS Number: 905854-02-6) | Cayman Chemical, accessed August 26, 2025, https://www.caymanchem.com/product/17135/tivantinib
7. Tivantinib (ARQ 197) Exhibits Antitumor Activity by Directly Interacting with Tubulin and Overcomes ABC Transporter–Mediated Drug Resistance - AACR Journals, accessed August 26, 2025, https://aacrjournals.org/mct/article/13/12/2978/116604/Tivantinib-ARQ-197-Exhibits-Antitumor-Activity-by
8. Tivantinib Completed Phase 1 Trials for Solid Tumors Treatment | DrugBank Online, accessed August 26, 2025, https://go.drugbank.com/drugs/DB12200/clinical_trials?conditions=DBCOND0029860&phase=1&purpose=treatment&status=completed
9. Tivantinib - Drug Targets, Indications, Patents - Patsnap Synapse, accessed August 26, 2025, https://synapse.patsnap.com/drug/f7a8704981ef48108e8cf716111ada88
10. Tivantinib in MET-high hepatocellular carcinoma patients and the ongoing Phase III clinical trial - PubMed Central, accessed August 26, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC6095177/
11. Metiv-HCC: A phase III clinical trial evaluating tivantinib (ARQ 197), a MET inhibitor, versus placebo as second-line in patients (pts) with MET-high inoperable hepatocellular carcinoma (HCC). - ASCO Publications, accessed August 26, 2025, https://ascopubs.org/doi/10.1200/jco.2013.31.15_suppl.tps4159
12. Daiichi Sankyo and ArQule Announce the Completion of the METIV-HCC Phase 3 Study of Tivantinib in Second-Line Treatment of MET-Overexpressing Hepatocellular Carcinoma - Press Releases - Media, accessed August 26, 2025, https://www.daiichisankyo.com/media/press_release/detail/index_3377.html
13. Second-line tivantinib (ARQ 197) vs placebo in patients (Pts) with ..., accessed August 26, 2025, https://ascopubs.org/doi/10.1200/JCO.2017.35.15_suppl.4000
14. Phase III Multinational, Randomized, Double-Blind, Placebo-Controlled Study of Tivantinib (ARQ 197) Plus Erlotinib Versus Erloti - IRIS, accessed August 26, 2025, https://iris.hunimed.eu/retrieve/de164a8e-bfa6-73f5-e053-d805fe0a7be4/SANTORO_Phase%20III%20Multinational%20Randomized%20Double-Blind%20Placebo-Cont....pdf
15. Kyowa Hakko Kirin Announces Discontinuation for Developing ARQ ..., accessed August 26, 2025, https://www.kyowakirin.com/media_center/news_releases/2017/e20171006_01.html
16. TIVANTINIB - precisionFDA, accessed August 26, 2025, https://precision.fda.gov/ginas/app/ui/substances/bada554e-6b9f-43e2-a237-fa40cdba24df
17. Tivantinib | C23H19N3O2 - ChemSpider, accessed August 26, 2025, https://www.chemspider.com/Chemical-Structure.9669218.html
18. Tivantinib (ARQ 197) | c-Met Inhibitor - MedchemExpress.com, accessed August 26, 2025, https://www.medchemexpress.com/Tivantinib.html
19. TIVANTINIB, (+)- - gsrs, accessed August 26, 2025, https://gsrs.ncats.nih.gov/ginas/app/beta/substances/E9UU0ESM69
20. Tivantinib | CAS 905854-02-6 | Cayman Chemical | Biomol.com, accessed August 26, 2025, https://www.biomol.com/products/chemicals/biochemicals/tivantinib-cay17135-1
21. Tivantinib | 99.65%(HPLC) | In Stock | c-Met inhibitor - Selleck Chemicals, accessed August 26, 2025, https://www.selleckchem.com/products/arq-197.html
22. Tivantinib ( LSM-1131 ) - Small Molecules, accessed August 26, 2025, https://lincsportal.ccs.miami.edu/SmallMolecules/view/LSM-1131
23. Compound: TIVANTINIB (CHEMBL2103882) - ChEMBL - EMBL-EBI, accessed August 26, 2025, https://www.ebi.ac.uk/chembl/explore/compound/CHEMBL2103882
24. TIVANTINIB - gsrs, accessed August 26, 2025, https://gsrs-dev-public.ncats.io/ginas/app/ui/substances/bada554e-6b9f-43e2-a237-fa40cdba24df
25. Safety Data Sheet - Cayman Chemical, accessed August 26, 2025, https://cdn.caymanchem.com/cdn/msds/17135m.pdf
26. Tivantinib (ARQ 197)|C-Met inhibitor,non-ATP-competitive - APExBIO, accessed August 26, 2025, https://www.apexbt.com/tivantinib-arq-197.html
27. tivantinib | Ligand page | IUPHAR/BPS Guide to PHARMACOLOGY, accessed August 26, 2025, https://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=7948
28. Cytotoxic Activity of Tivantinib (ARQ 197) Is Not Due Solely to c-MET Inhibition, accessed August 26, 2025, https://aacrjournals.org/cancerres/article/73/10/3087/584223/Cytotoxic-Activity-of-Tivantinib-ARQ-197-Is-Not
29. Phase 1 study of safety, pharmacokinetics, and pharmacodynamics of tivantinib in combination with bevacizumab in adult patients with advanced solid tumors, accessed August 26, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC8628145/
30. A Phase 3, Randomized, Double-Blind Study of Tivantinib (ARQ 197) in Subjects with MET Diagnostic-High Inoperable Hepatocellular Carcinoma (HCC) Treated with One Prior Systemic Therapy (ARQ - Dana-Farber Cancer Institute, accessed August 26, 2025, https://www.dana-farber.org/clinical-trials/13-216
31. A Phase I Dose-Escalation Study of Tivantinib (ARQ 197) in Adult Patients with Metastatic Solid Tumors | Clinical Cancer Research - AACR Journals, accessed August 26, 2025, https://aacrjournals.org/clincancerres/article/17/24/7754/204901/A-Phase-I-Dose-Escalation-Study-of-Tivantinib-ARQ
32. Metabolism and disposition of [(14)C]tivantinib after oral ... - PubMed, accessed August 26, 2025, https://pubmed.ncbi.nlm.nih.gov/24911380/
33. TIVANTINIB - Inxight Drugs, accessed August 26, 2025, https://drugs.ncats.io/drug/PJ4H73IL17
34. Selective Inhibitor of the c-Met Receptor Tyrosine Kinase in Advanced Hepatocellular Carcinoma: No Beneficial Effect With the Use of Tivantinib? - Frontiers, accessed August 26, 2025, https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2021.731527/full
35. Tivantinib plus erlotinib versus placebo plus erlotinib in Asian patients with previously treated nonsquamous NSCLC with wild-type EGFR: First report of a phase III ATTENTION trial. - ASCO Publications, accessed August 26, 2025, https://ascopubs.org/doi/10.1200/jco.2014.32.15_suppl.8044
36. A randomized, double-blind, placebo-controlled, phase III trial of erlotinib with or without a c-Met inhibitor tivantinib (ARQ 197) in Asian patients with previously treated stage IIIB/IV nonsquamous nonsmall-cell lung cancer harboring wild-type epidermal growth factor receptor (ATTENTION study) - PubMed, accessed August 26, 2025, https://pubmed.ncbi.nlm.nih.gov/26153496/
37. Drug-drug Interaction Study of Tivantinib (ARQ 197) With Omeprazole, S-warfarin, Caffeine, Midazolam, and Digoxin in Cancer Subjects | ClinicalTrials.gov, accessed August 26, 2025, https://clinicaltrials.gov/study/NCT01517399
38. Tivantinib Completed Phase 1 / 2 Trials for Advanced Solid Tumors Treatment - DrugBank, accessed August 26, 2025, https://go.drugbank.com/drugs/DB12200/clinical_trials?conditions=DBCOND0030110&phase=1%2C2&purpose=treatment&status=completed
39. Tivantinib Completed Phase 2 Trials for Pancreatic Neoplasms Treatment - DrugBank, accessed August 26, 2025, https://go.drugbank.com/drugs/DB12200/clinical_trials?conditions=DBCOND0028305&phase=2&purpose=treatment&status=completed
40. Study Details | Tivantinib in Treating Patients With Recurrent or Metastatic Breast Cancer, accessed August 26, 2025, https://clinicaltrials.gov/study/NCT01575522
41. Study Details | Tivantinib in Treating Patients With Previously Treated Malignant Mesothelioma | ClinicalTrials.gov, accessed August 26, 2025, https://www.clinicaltrials.gov/study/NCT01861301?intr=ARQ-197&aggFilters=studyType:int&viewType=Table&rank=7
42. Safety and efficacy of MET inhibitor tivantinib (ARQ 197) combined with sorafenib in patients (pts) with NRAS wild-type or mutant melanoma from a phase I study. - ASCO Publications, accessed August 26, 2025, https://ascopubs.org/doi/10.1200/jco.2012.30.15_suppl.8519
43. Profile of tivantinib and its potential in the treatment of hepatocellular carcinoma: the evidence to date - PMC, accessed August 26, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC5118026/
44. Tivantinib Plus Erlotinib Versus Placebo Plus Erlotinib for the Treatment of Non-squamous, Non-small-cell Lung Cancer | ClinicalTrials.gov, accessed August 26, 2025, https://www.clinicaltrials.gov/study/NCT01244191
45. ArQule, Inc. Soars As FDA Panel Favors Continuation Of Liver Cancer Drug Study, Stock Up +6.78% At Market Close (January 17, 2014) - BioSpace, accessed August 26, 2025, https://www.biospace.com/arqule-inc-soars-as-fda-panel-favors-continuation-of-liver-cancer-drug-study-stock-up-6-78-percent-at-market-close-january-17-2014
46. POSITIVE PHASE 2 STUDY RESULTS FOR TIVANTINIB IN PREVIOUSLY TREATED HEPATOCELLULAR CARCINOMA TO BE PRESENTED AT ASCO - Press Releases - Media - Daiichi Sankyo, accessed August 26, 2025, https://www.daiichisankyo.com/media/press_release/detail/index_3690.html
47. Daiichi, ArQule cancer drug fails in late-stage study | BioPharma Dive, accessed August 26, 2025, https://www.biopharmadive.com/news/Arqule-Daiichi-Sankyo-Tivantinib-carcinoma-trial/436478/
48. The efficacy and safety of tivantinib in the treatment of solid tumors: a systematic review and meta-analysis - PMC, accessed August 26, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC5762579/
49. Safety and efficacy of MET inhibitor tivantinib (ARQ 197) combined with sorafenib in patients (pts) with hepatocellular carcinoma (HCC) from a phase I study. - ASCO Publications, accessed August 26, 2025, https://ascopubs.org/doi/10.1200/jco.2012.30.15_suppl.4117
50. c-Met inhibitor - Wikipedia, accessed August 26, 2025, https://en.wikipedia.org/wiki/C-Met_inhibitor
51. Safety and Tolerability of c-MET Inhibitors in Cancer - PMC - PubMed Central, accessed August 26, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC7491978/
52. ROS-1 NSCLC therapy resistance mechanism - García-Pardo - Precision Cancer Medicine, accessed August 26, 2025, https://pcm.amegroups.org/article/view/6343/html
53. CABOMETYX® (cabozantinib) Mechanism of Action, accessed August 26, 2025, https://www.cabometyxhcp.com/mechanism-of-action
54. What diseases does Cabozantinib treat? - Patsnap Synapse, accessed August 26, 2025, https://synapse.patsnap.com/article/what-diseases-does-cabozantinib-treat