PARP inhibitor therapy in patients with IDH1 mutated cholangiocarcinoma

Abstract

Background

Isocitrate dehydrogenase 1 (IDH1) missense mutations occur at a frequency of 10%-15% in intrahepatic cholangiocarcinoma (iCCA). IDH1 mutations result in accumulation of (R)-2-hydroxyglutarate, an oncometabolite that leads to DNA hypermethylation and impairment of homologous recombination (HR). Impairment of HR results in a “BRCAness” phenotype which may confer sensitivity to poly(ADP ribose) polymerase (PARP) inhibition.

Methods

We conducted a retrospective cohort review to identify patients with advanced, IDH1 mutated iCCA treated with a PARP inhibitor (PARPi) at the University of Michigan between 2018 and 2023. Patients are described with respect to prior lines of therapy, response to platinum-based chemotherapy, and progression-free survival (PFS) and overall survival (OS) from the time of PARPi initiation.

Results

Between 2018 and 2023 we identified 40 patients with IDH1 mutated iCCA of which 6 patients were treated with a PARPi as monotherapy or in combination with an ATR inhibitor or anti-PD-1 immune checkpoint inhibitor. Majority of patients (n = 5) carried an IDH1 R132C mutation per tissue-based next generation sequencing. All patients had previously received at least one line of cisplatin-based systemic therapy for advanced disease prior to treatment with PARPi. PFS and OS from time of PARPi initiation ranged from 1.4 to 18.5 months and 2.8 to 42.4 months, respectively. Best response on PARPi therapy included 2 partial responses.

Conclusion

This is the first case series to describe PARPi treatment in IDH1 mutated iCCA. Results underscore the limitation of PARPi monotherapy, potentially support combined PARPi therapies, and highlight a need for effective treatment options for patients with IDH1 mutated iCCA.

This article reports an institutional case series in patients with advanced, IDH1 mutated cholangiocarcinoma treated with PARP inhibitor therapy.

Introduction

The treatment landscape for advanced biliary tract cancers (BTC) is expanding with genomic characterization of BTC. Next-generation sequencing (NGS) has identified targetable mutations in BTC, including the isocitrate dehydrogenase 1 (IDH1) hotspot missense mutation with a frequency of 10%-15% in intrahepatic cholangiocarcinoma (iCCA).^1-3 Isocitrate dehydrogenase 1 is a cytoplasmic NADPH-dependent enzyme in the citric acid cycle which catalyzes the conversion of isocitrate to alpha-ketoglutarate (α-KG).^4,5IDH1 mutations result in increased formation of (R)-2-hydroxyglutarate ((R)-2HG), an oncometabolite which has been reported to inhibit α-KG-dependent dioxygenases leading to DNA hypermethylation and impairment of homologous recombination (HR).^6-8 This impairment of HR in IDH1 mutant cells has been shown to result in a “BRCAness” phenotype and dependence on alternate DNA damage repair pathways.⁸

There have been significant efforts thus far to target IDH1 mutated iCCA with small molecule inhibitors targeting mutant IDH1 (mIDH1) protein. Ivosidenib is an oral, small-molecule, allosteric inhibitor of mIDH1 protein that demonstrated an improvement in the primary endpoint of median progression-free survival (PFS; 2.7 vs 1.4 months, P < 0.001) compared to placebo in the phase III CLARIDHY trial.⁹ Current therapeutic strategies for IDH1 mutated BTC rely on direct inhibition of the mIDH1 protein, whereas targeting DNA damage repair (DDR) vulnerabilities conferred by the oncometabolite (R)-2HG remains under active investigation.

PARP1 is a nuclear protein involved in the repair of single strand breaks.¹⁰ The benefit from poly(ADP ribose) polymerase (PARP) inhibitors in patients with advanced cancer and deleterious somatic or germline mutations in DDR pathways has been well established.^11,12 PARP inhibitors induce “PARP trapping,” an inhibition of PARP complex release from sites of DNA damage which blocks DNA damage repair.¹³ Unrepaired single strand breaks can produce deleterious double-strand breaks which then require repair via the HR and non-homologous end-joining (NHEJ) pathways. In HR deficient cancers, repair mechanisms for double-strand breaks fall to more error prone NHEJ pathways which can lead to increased antigenicity and cell death.¹⁴

Herein, we report an institutional case series in patients with advanced, IDH1 mutated BTC treated with PARP inhibitor (PARPi) therapy.

Methods

Study design and cohort identification

This single-center, retrospective cohort study was approved by the institutional review board (HUM00149617) at the University of Michigan. Informed consent was not required due to the retrospective nature and minimal risk of the study. All patients aged at least 18 years with next generation sequencing analysis and a histologically confirmed diagnosis of BTC were reviewed for eligibility.

Data collection

Patient demographics, next generation sequencing results, disease characteristics, and treatment data were collected using DataDirect, Electronic Medical Record Search Engine (EMERSE), and manual review of electronic medical records (EMRs).¹⁵

Statistical analysis

PFS was defined as time from start of PARPi to the date of radiologic or clinical progression or death from any cause, or censored at last date of contact, if still on therapy. Overall survival (OS) was calculated as time from start of PARPi to death from any cause, or censored at last date of contact, if still alive at the cutoff date of this analysis. Best response was defined per response evaluation criteria in solid tumors (RECIST) version 1.1. Descriptive statistics were used to summarize patient characteristics and results.

Results

Patients

Between 2018 and 2023, we identified 40 patients with IDH1 mutated BTC of which 6 patients ages 44 to 76 years with advanced stage iCCA were treated with a PARPi (olaparib or rucaparib) at our institution. Five patients carried an IDH1 R132C mutation and one patient carried an IDH1 R132G mutation per tissue-based NGS. Two patients in this series received off label single agent olaparib, 3 patients were treated with the combination rucaparib and anti-PD1 antibody, nivolumab, on a clinical trial (NCT03639935), and one with olaparib and the ATR inhibitor camonsertib on another trial (NCT04972110) (Table 1). All patients had previously received at least one line of cisplatin-based systemic therapy for advanced disease prior to treatment with PARPi, with 2 patients experiencing partial response (PR) and 3 with stable disease (SD) as best response on cisplatin-based chemotherapy per RECIST v1.1. PARPi was offered at the time of progressive disease (PD) on prior treatment (n = 3) or as maintenance therapy in combination with nivolumab in patients without prior progression on platinum-based systemic therapy (n = 3). Four patients subsequently received ivosidenib, as either third-line or fifth-line treatment. Majority (n = 4) of patients had an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0 at the time of PARPi initiation.

Table 1.

Patient characteristics

Patient	Sex	Race ethnicity	TNM stage	BTC subtype	IDH1 mutation	Concurrent somatic mutations	Germline variants	TMB≠	Microsatellite statusα	Treatment prior to PARPi	Best response on chemotherapy€	PARPi treatment	ECOG PS*	Line of therapy±	Treatment-related grade ≥3 AEs^
1	F	White Non-Hispanic	IV	iCCA	R132C	PBRM1 p.M957fs insertion KDM6B p.L113fs insertion	—	7.3	MSS	Nivolumab and IpilimumabGemcitabine and Cisplatin	PD	Olaparib	2	3rd	Anemia, thrombocytopenia
2	F	White Non-Hispanic	IV	iCCA	R132C	BAP1 p.Q436*	—	NR**	MSS	Gemcitabine and Cisplatin	PR	Olaparib	1	2nd	Thrombocytopenia, neutropenia
3	F	White Non-Hispanic	IV	iCCA	R132G	BRAF p.N661K ARID1A p.P120fs	-	2.6	MSS	Gemcitabine and Cisplatin and Nab-paclitaxel	PR	Rucaparib and Nivolumab	0	M	Thrombocytopenia, AST, ALT and alkaline phosphatase elevation, anemia
4	F	White Non-Hispanic	IV	iCCA	R132C	SAV1 pR132* BAP1 p.S327fs insertion	PMS2 p.K647*	2	MSS	Gemcitabine and Cisplatin	SD	Rucaparib and Nivolumab	0	M	Rash, neutropenia
5	F	White Non-Hispanic	IIIB	iCCA	R132C	RAF1 p.L613V PTPRT p.E644fs insertion	—	11	MSS	Gemcitabine and Cisplatin	SD	Rucaparib and Nivolumab	0	M	Neutropenia
6	M	White Non-Hispanic	IB	iCCA	R132C	BAP1 Q28* PDRM1 K1020fs *115	—	NR**	MSS	Gemcitabine and Cisplatin & CPI-613	SD	Olaparib and Camonsertib	0	2nd	Anemia

^≠ TMB = tumor mutation burden, reported as mutations/Mb.

^α Microsatellite status reported as MSI high (MSI-H) or MSI stable (MSS).

**NR = not reported.

*Eastern Cooperative Oncology Group Performance Status (ECOG PS) assessed at the time of PARP inhibitor (PARPi) therapy.

^± M = maintenance therapy.

^€SD, stable disease; PD, progressive disease; PR, partial response.

^{^}AE, adverse events graded per CTCAE v5.0.

Efficacy

The PFS from time of PARPi initiation ranged from 1.4 to 18.5 months; Figure 1. PARPi was discontinued in all patients due to PD and not due to toxicity. OS, calculated from date of start of PARPi ranged from 2.8 to 42.4 months, and from start of first-line systemic therapy for advanced disease 7.5-48.0 months; Figure 2. Best response on PARPi therapy included 2 PR, 2 SD and 2 PD as depicted in Figure 1.

Figure 1.

Swimmer’s plot for progression-free survival for IDH1 mutated patients treated with a PARP inhibitor. Two patients had a PR (green); progression-free survival range is 1.4 to 18.5 months.

Figure 2.

Stacked bar graph for overall survival and treatment sequencing for IDH1 mutated patients. Overall survival was calculated from the start of systemic therapy for advanced disease to the time of death or last follow-up. Patients 2, 4, and 5 are alive and continue on treatment. Overall survival range is 7.5-48.0 months. Abbreviations: NI, nivolumab and ipilimumab; GC, gemcitabine and cisplatin; O, olaparib; Ivo, ivosidenib; A, ablation; GCA, gemcitabine and cisplatin and nab-paclitaxel; RN, rucaparib and nivolumab; FOLFOX, 5-fluorouracil and oxaliplatin; RG, regorafenib; TN, TPST-1120 and nivolumab; R, chemoradiation; GCD, gemcitabine, cisplatin, and devimistat; CamO, camonsertib and olaparib; CO, capecitabine and oxaliplatin.

Of the 2 patients with PR, one patient received rucaparib and nivolumab as maintenance therapy after a prior PR with gemcitabine, cisplatin, and nab-paclitaxel chemotherapy. The second patient with a PR was treated with olaparib and camonsertib as second-line therapy after progression on first-line gemcitabine, cisplatin, and CPI-613. These patients had a PFS of 8.5 and 7.3 months, respectively.

One patient treated with rucaparib and nivolumab as maintenance therapy had an underlying germline PMS2 mutation but had microsatellite stable disease and tumor mutational burden of 2. This patient received 10 cycles of maintenance rucaparib and nivolumab with a best response of SD, PFS of 9.2 months and OS of 34.6 months from start of gemcitabine and cisplatin.

Discussion

This institutional case series highlights the clinical application of PARPi therapy in patients with IDH1 mutated advanced iCCA. IDH1 mutations have a modest frequency in iCCA and the inhibition of HR by the oncometabolite R-2HG and subsequent deficiency in DDR provides a unique therapeutic target. Our case series is a heterogenous group of patients treated with PARPi therapy, used either as maintenance or a later line therapeutic, alone or in combination. Based on these limited data, we observed minimal to no activity of PARPi monotherapy but modest yet preliminary evidence for response and/or disease control with PARPi in combination with immune checkpoint or ATR inhibitors. Of some interest, the OS for advanced disease was observed to be appreciably longer for patients in this series but may also be ascribed to the sensitivity to cisplatin-based chemotherapy, with 5 of the patients having at least stable disease on prior platinum-based chemotherapy.

Two patients with olaparib monotherapy with progression on prior platinum therapy experienced limited to no benefit. This lack of benefit from PARPi monotherapy as subsequent line therapeutic was exemplified by the phase II NCI10129 trial (NCT03212274) which evaluated olaparib in patients with IDH1 and IDH2 mutant advanced solid tumors refractory to standard treatment.¹⁶ Unfortunately, no response was seen in 30 patients with CCA and further enrollment was closed.

Interestingly, 3 patients in our case series with the longest PFS (range, 8.5-18.5 months) were treated with PARPi in combination with immune checkpoint inhibitor (ICI) as maintenance therapy in absence of progression on prior platinum therapy. Preclinical studies in mice with Brca1-deficient, platinum-sensitive ovarian tumors have demonstrated increased tumor antigenicity with PARP inhibition. Subsequent treatment with PARPi and anti-PD-1 ICI resulted in antitumor activity greater than PARPi therapy alone, thus providing a rationale for combination of PARPi with ICI.¹⁷ A recent phase II SOLID trial (NCT03991832) evaluated olaparib and anti-PDL1 ICI, durvalumab, in patients with IDH1 mutated CCA.¹⁸ Enrolled patients had received at least one prior systemic therapy, with 90% receiving prior platinum-based therapy. Unfortunately, no responses were seen in the first 10 patients, and the study did not proceed to stage 2. In comparison, the prolonged survival and a PR in our case series may be due to platinum-sensitive disease, defined as stable or responding disease on platinum-based therapy, as compared to those on the phase II SOLID clinical trial. Indeed, the benefit of PARPi as maintenance therapy in absence of progression on first-line platinum-based chemotherapy has been substantiated in other cancers, including in patients with metastatic pancreatic cancer with underlying germline BRCA mutations.^19,20

One patient treated with olaparib and an ATR inhibitor (camonsertib) after progression on platinum-based therapy experienced a PR and PFS of 7.3 months in our cohort. ATR and PARP inhibitors have distinct effects on DDR and pre-clinical studies have highlighted a synergistic effect on tumor cell death compared to single agent therapy.^21-24 Moreover, preclinical models suggest that ATR inhibition can overcome PARPi and platinum resistance, suggesting rationale for dual targeting of DDR pathways in the setting of platinum resistance.^22,25 In an effort to reverse PARPi resistance, the phase II OLAPCO basket trial (NCT0257644) evaluated use of olaparib alone or in combination with other HR directed therapies including ceralasertib, an ATR inhibitor.²⁶ In the combination arm of olaparib and ceralasertib, one patient with IDH1 R132C mutated myxoid tumor had a best response of SD with duration of response of 8 months. Based on potential resistance to PARPi monotherapy and efforts to exploit DDR alterations, including IDH1 mutations, in CCA, there are 2 ongoing clinical trials (NCT04298021 and NCT03878095) evaluating the use of ceralasertib, an ATR inhibitor, in combination with olaparib for patients with advanced BTC.

The limitations of this case series include sample size and heterogeneity of treatment. It questions benefit of PARPi monotherapy, supports use of combination therapy and is hypothesis generating for its role in platinum sensitive as opposed to resistant tumors. IDH1 mutation may be a predictive biomarker of response to PARPi but further studies, with or without concurrent anti-PD1/L1 antibody as maintenance therapy and/or other agents targeting DDR, are needed.

Author contributions

Arathi Mohan (Data curation, Formal Analysis, Visualization, Writing—original draft), Elit Quingalahua (Data curation, Formal Analysis), Valerie Gunchick (Data curation), Simi Paul (Data curation), Chandan Kumar-Sinha (Data curation), Oxana Crysler (Investigation), Mark M. Zalupski (Investigation, Supervision, Writing—review & editing), Vaibhav Sahai (Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Resources, Supervision, Writing—review & editing)

Conflicts of interest

A.M.—Institutional grant funding from Verastem. E.Q.—None. V.G.—Cornerstone (previously Rafael). S.P.—None. C.K.S.—None. O.C.—Institutional grant funding from Servier Pharmaceuticals LLC, Surface Oncology, Genentech, AstraZeneca, Tempest Therapeutics, and sourced funding through Academic and Community Cancer Research United (ACCRU). M.M.Z.—Institutional grant funding from AstraZeneca, MedImmune, and Seattle Genetics. V.S.—Institutional research grant funding—Actuate Therapeutics, Boehringer Ingram, Bristol-Myers Squibb, Clovis, Esanik, Exelixis, Fibrogen, Ipsen, MedImmune, NCI, PanCAN, Cornerstone (previously Rafael), Relay, Repare, Servier, Syros, Transthera. Consultant—Amplity (Lynx Group), AstraZeneca, Autem, Delcath Systems, Histosonics, Ipsen, Incyte, Cornerstone (previously Rafael), Servier and Taiho.

Funding

This work was supported by Rogel Cancer Center, University of Michigan (P30CA046592) Rogel Cancer Center, University of Michigan (Rogel Scholar).

Data availability

The data underlying this article cannot be shared publicly due to the privacy of individuals that participated in the study or were treated at our institution. The data will be shared on reasonable request to the corresponding author.