Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive solid tumors. Most patients are diagnosed with advanced and inoperable disease. The median overall survival (OS) in patients with stage IV disease is limited.1,2 General OS is approximately 4 to 11 months and real-life data on OS are limited, although it is slightly higher in patients selected for clinical trials.3–5 Management of this disease is challenging, with patients presenting with severe debilitating symptoms at diagnosis, including weight loss, radiating abdominal pain, loss of appetite, depression, and deteriorated physical activity and well-being.6 Furthermore, PDAC is resistant to chemotherapy, with patients treated with first-line combination chemotherapy therapy experiencing 4 to 6 months or less of progression-free survival (PFS). The response rate to combination chemotherapy protocols is <40%.5,7
To improve the management of PDAC in an era of precision medicine, it is highly important to identify subsets of patients who can benefit from targeted treatments. In particular, BRCA1/2 germline mutations (gBRCAm) affect up to 7% of patients with PDAC.8 The BRCA1/2 proteins play a significant role in the repair of DNA double-strand breaks (DSBs). Tumors with homologous recombination repair (HRR) gene abnormalities such BRCA1/2 are sensitive to both platinum and PARP inhibitors (PARPi).9
The platinum salts (carboplatinum, cisplatin, and oxaliplatin) are well-documented DNA-damage–inducing chemotherapies that generate covalent cross-links between DNA bases. The cytotoxic effects of these agents are determined by the relative amount and specific structure of DNA adducts.10 The efficacy and response rates of platinum salts have been found to be higher in tumor cells with DSB repair apparatus deficiencies. PARP inhibition causes the failure of single-strand break (SSB) repairs. SSBs encountered by the replication fork result in fork stalling and possible fork collapse or the formation of DSBs. In the absence of BRCA1/2 functional proteins, the replication fork cannot be restarted and collapses. PARP inhibition mechanisms also include the “trapping” of the PARP-1 protein on the site of DNA damage, leading to a lack of PARP-1 protein release and recruitment of repair proteins, and expression of cytotoxic DNA complexes.9
Based on the understanding of BRCA proteins and their essential role in DNA repair, gBRCAm were found to be a specific predictive biomarker of DNA damage–based treatment by platinum-based chemotherapies. In 2005, PARPi were introduced as a clinical option, and based on their specific mode of inhibition and preclinical data, BRCA-related tumors were tested in these early-phase clinical trials.11
Data from clinical trials in ovarian cancer helped clarify the use of PARPi in solid tumors. An additional important lesson learned from PARPi in ovarian cancer trials and basket trials in other solid tumors was that there was a cross-resistance between platinum salts and PARPi in gBRCAm.12–16 These results suggested that PARPi should first be tested as a maintenance strategy to maximize its potential efficacy and obtain biologic tumor stabilization (ie, stable or in remission with chemotherapy). Recent, more efficient polychemotherapies (ie, 5-FU/irinotecan/oxaliplatin [FOLFIRINOX]) have improved the control of PDAC tumors. This has facilitated a change of paradigm, including the development of maintenance strategies that preserve a steady state between a tumor with reduced activity and the host, that does not negatively affect patients’ quality of life (QoL) by using treatments that are less toxic than platinum-based chemotherapy. In particular, retrospective data in PDAC have shown that the median OS in patients with gBRCAm is as poor as in those with sporadic PDAC. Certain retrospective clinical data suggest that patients with gBRCAm and PDAC may have long-lasting responses to platinum-based chemotherapy and an extended OS. This is supported by recent retrospective and prospective clinical data. These combined datasets have shown that although gBRCAm is a predictive biomarker, it is not prognostic.17–19
There were certain difficulties in the development of PARPi for patients with PDAC. First, the exact prevalence and potential geographic/ethnic variability of the gBRCAm were unclear, making it difficult to plan a strategic clinical trial. Furthermore, it is unclear whether a biologically guided maintenance approach can be evaluated in the presence of metastatic PDAC with a short OS.
In the phase III POLO study designed in 2014, the PARPi olaparib was given as maintenance treatment following first-line platinum-based chemotherapy in patients with gBRCAm metastatic PDAC versus placebo.20 Patients had to receive at least 16 weeks of first-line platinum-based chemotherapy, without tumor progression. No upper limit to the duration of chemotherapy was defined in the eligibility criteria of the study, giving trial investigators flexibility and facilitating enrollment. The study design included a prescreening period for eligible patients with metastatic PDAC. Of the 3,315 patients enrolled in the prescreening process, 7.5% were found to have a gBRCAm, including approximately 1.5% with a known mutation before the trial. Approximately two-thirds of the patients with PDAC had a gBRCA2m and one-third had a gBRCA1m, whereas most patients with ovarian or breast cancer had BRCA1 mutations.
In the POLO trial, 38% of patients were not eligible for randomization for maintenance therapy, most because of disease progression during first-line platinum-based chemotherapy (near 30%). These patients probably had an innate (primary) resistance to DNA-damaging chemotherapy despite the gBRCAm, for example due to the acquisition of a reverse mutation during platinum treatment, upregulation of the ATR/CHK1 pathway, or restoring RAD51 foci formation.
Finally, the 154 eligible patients for the POLO trial were randomized 3:2 to either olaparib tablets at 300 mg twice daily or placebo, and continued treatment until disease progression or unacceptable toxicity.
The primary endpoint was PFS assessed by a blinded independent central review. PFS was measured from the time of randomization, which occurred after first-line chemotherapy had been completed. Key secondary endpoints included time to second progression, objective response rate, health-related quality of life, safety and tolerability, and OS. The primary analysis was prespecified at 87 events of PFS, which would give the study a power of 80% to detect a significant difference between arms with a one-sided alpha of 0.025.
Patient characteristics were well balanced between the two arms. Multivariate analysis showed that minor imbalances between arms did not influence the results of PFS. One-third of olaparib arm patients were still on treatment at the time of data cutoff (January 15, 2019).
The primary endpoint (PFS) was 7.4 months in the olaparib arm and 3.8 months in the placebo arm (hazard ratio, 0.53; P=.0038). Prespecified analyses were performed of the proportion of patients who were progression-free at 6, 12, 18, and 24 months. From 6 months onwards, more than twice the proportion of patients in the olaparib arm were progression-free compared with the placebo arm, which is consistent with the hazard ratio of 0.53. All of the subgroup analyses of PFS favored olaparib. An additional interesting finding was that the PFS in patients in the olaparib arm was similar among those with stable disease and those with a response to first-line chemotherapy before randomization.
At the time of the primary PFS analysis, a predefined interim OS analysis was performed as part of a hierarchical testing plan. Median OS was 18.9 months with olaparib and 18.1 months with placebo, with a hazard ratio of 0.91. Subsequent treatments after tumor progression (mainly reintroduction of chemotherapy) may have influenced our OS results. In particular, 14.5% of patients in the placebo arm received subsequent treatment with a PARPi that was not part of the study design. Final OS analysis will be performed once 106 deaths have occurred.
The blinded independent central review of the objective response rate in patients with measurable disease was 11.5% for placebo and 23.1% for olaparib, with 2 complete responses. The onset of response occurred earlier in the placebo arm, suggesting that these responses were a carryover from/influenced by first-line platinum-based chemotherapy.
In the POLO trial, median duration of response was 6 months, but it was much longer (>24 months) in a subset of patients (23%) who had objective response at imaging, which is exceptional in patients with metastatic PDAC. Patient-reported health-related QoL in the first 6 months of treatment was not significantly deteriorated in those receiving olaparib based on the EORTC QLQ-C30 questionnaire using a 10-point change in the 100-point scale.21 Forty percent of patients in the olaparib arm had a grade ≥3 adverse event (AE). However, only 5.5% of AEs required treatment discontinuation. Median duration of treatment was 6 months for olaparib and 3.7 months for placebo. The toxicity profile of olaparib was similar to that identified in other tumors. The most common AEs were fatigue, nausea, and diarrhea. Anemia and fatigue were the most frequent grade ≥3 AEs.
Other PARPi are being evaluated in patients with PDAC (Table 1). In a randomized phase II study comparing modified FOLFIRI + veliparib versus FOLFIRI alone for metastatic PDAC in the second-line setting (SWOG S1513), the addition of veliparib increased toxicity and did not improve OS in biomarker-unselected patients.22
Clinical Trials: PARPi PDAC
In another study by O’Reilly et al,18 50 patients with gBRCA- or PALB2-mutated PDAC were randomized to receive the combination of gemcitabine (600 mg/m2) and cisplatin (25 mg/m2) on day 3 and 10 either alone or with veliparib (80 mg) orally twice daily from day 1 to 12 cycled every 3 weeks. The response rate among patients in the veliparib arm was not superior to those in the other arm (74.1% vs 65.2%; P=.55). Median PFS was 10.1 months (95% CI, 6.7–11.5 months) and 9.7 months (95% CI, 4.2–13.6 months; P=.73), respectively, and median OS was 15.5 months (95% CI, 12.2–24.3 months) and 16.4 months (95% CI, 11.7–23.4 months; P=.6), respectively. Overall, veliparib did not to improve the results and provided more grade 3/4 hematologic toxicity. This study underlined the valuable efficacy of combination gemcitabine + cisplatin. However, it should be noted that unlike in the POLO study, patients received PARPi as frontline therapy, with no preselection chemotherapy with a platinum agent.
In the inclusion criteria of the POLO study there was no upper limit to how many cycles of treatment a patient could receive. The decision to stop chemotherapy in whole or part was left to the discretion of the treating physician. It seems that switching to PARPi usually occurs around 6 months, based on the median time from diagnosis to randomization in the POLO study (6.9 months [range, 3.6–38.4 months]). However, based on trial design and results, this is still open for discussion between patients and physicians. Moreover, at the time of study initiation, the results of the (still unpublished) PANOPTIMOX study comparing 5-FU maintenance versus continuation of FOLFIRINOX (reported at the ASCO meeting in 2018) were not evaluable (NCT02352337); otherwise, to our knowledge, prospective evaluation of the role of FOLFIRI or capecitabine maintenance has not been performed.
PDAC resistance to PARPi is a challenge. In the POLO trial, approximately 40% of patients with controlled tumors after being randomized to platinum-based chemotherapy had a primary resistance to olaparib. Analysis of circulating cell-free DNA for BRCA reversion mutations will probably help guide therapeutic decisions in the future.23 Otherwise, combining PARPi with other agents, such as checkpoint inhibitors and bromodomain and extra-terminal domain protein inhibitors such as JQ1, could increase the efficacy of PARPi.24
Finally, the value of PARPi and the best management of progressive tumors on olaparib must be evaluated in earlier forms of PDAC, those with rarer BRCAness signatures (eg, PALB2, RAD51, ATM), or those with somatic mutations leading to HRR deficiency.
Aier I, Semwal R, Sharma A, A systematic assessment of statistics, risk factors, and underlying features involved in pancreatic cancer. Cancer Epidemiol 2019;58:104–110.
Goldstein D, El-Maraghi RH, Hammel P, nab-Paclitaxel plus gemcitabine for metastatic pancreatic cancer: long-term survival from a phase III trial. J Natl Cancer Inst 2015;107:107.
Golan T, Sella T, Margalit O, Short- and long-term survival in metastatic pancreatic adenocarcinoma, 1993–2013. J Natl Compr Canc Netw 2017;15:1022–1027.
Conroy T, Desseigne F, Ychou M, FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med 2011;364:1817–1825.
Boyd AD, Brown D, Henrickson C, Screening for depression, sleep-related disturbances, and anxiety in patients with adenocarcinoma of the pancreas: a preliminary study. ScientificWorldJournal 2012;2012:650707.
Von Hoff DD, Ervin T, Arena FP, Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med 2013;369:1691–1703.
Golan T, Kindler HL, Park JO, Geographic and ethnic heterogeneity of germline BRCA1 or BRCA2 mutation prevalence among patients with metastatic pancreatic cancer screened for entry into the POLO trial. J Clin Oncol 2020;38:1442–1454.
- Search Google Scholar
- Export Citation
)| false , Golan T , Kindler HL , Park JO Geographic and ethnic heterogeneity of germline BRCA1 or BRCA2 mutation prevalence among patients with metastatic pancreatic cancer screened for entry into the POLO trial. J Clin Oncol 2020; 38: 1442– 1454. 32073954 10.1200/JCO.19.01890
Johnstone TC, Suntharalingam K, Lippard SJ. The next generation of platinum drugs: targeted pt(II) agents, nanoparticle delivery, and pt(IV) prodrugs. Chem Rev 2016;116:3436–3486.
Gudmundsdottir K, Ashworth A. The roles of BRCA1 and BRCA2 and associated proteins in the maintenance of genomic stability. Oncogene 2006;25:5864–5874.
McMullen M, Karakasis K, Madariaga A, Overcoming platinum and PARP-inhibitor resistance in ovarian cancer. Cancers (Basel) 2020;12:12.
Kaufman B, Shapira-Frommer R, Schmutzler RK, Olaparib monotherapy in patients with advanced cancer and a germline BRCA1/2 mutation. J Clin Oncol 2015;33:244–250.
Gelmon KA, Tischkowitz M, Mackay H, Olaparib in patients with recurrent high-grade serous or poorly differentiated ovarian carcinoma or triple-negative breast cancer: a phase 2, multicentre, open-label, non-randomised study. Lancet Oncol 2011;12:852–861.
- Search Google Scholar
- Export Citation
)| false , Gelmon KA , Tischkowitz M , Mackay H Olaparib in patients with recurrent high-grade serous or poorly differentiated ovarian carcinoma or triple-negative breast cancer: a phase 2, multicentre, open-label, non-randomised study. Lancet Oncol 2011; 12: 852– 861. 10.1016/S1470-2045(11)70214-5 21862407
Fong PC, Yap TA, Boss DS, Poly(ADP)-ribose polymerase inhibition: frequent durable responses in BRCA carrier ovarian cancer correlating with platinum-free interval. J Clin Oncol 2010;28:2512–2519.
Golan T, Kanji ZS, Epelbaum R, Overall survival and clinical characteristics of pancreatic cancer in BRCA mutation carriers. Br J Cancer 2014;111:1132–1138.
O’Reilly EM, Lee JW, Zalupski M, Randomized, multicenter, phase II trial of gemcitabine and cisplatin with or without veliparib in patients with pancreas adenocarcinoma and a germline BRCA/PALB2 mutation. J Clin Oncol 2020;38:1378–1388.
- Search Google Scholar
- Export Citation
)| false , O’Reilly EM , Lee JW , Zalupski M Randomized, multicenter, phase II trial of gemcitabine and cisplatin with or without veliparib in patients with pancreas adenocarcinoma and a germline BRCA/PALB2 mutation. J Clin Oncol 2020; 38: 1378– 1388. 10.1200/JCO.19.02931
Golan T, Sella T, O’Reilly EM, Overall survival and clinical characteristics of BRCA mutation carriers with stage I/II pancreatic cancer. Br J Cancer 2017;116:697–702.
Golan T, Hammel P, Reni M, Maintenance olaparib for germline BRCA-mutated metastatic pancreatic cancer. N Engl J Med 2019;381:317–327.
Hammel P, Kindler HL, Reni M, Health-related quality of life in patients with a germline BRCA mutation and metastatic pancreatic cancer receiving maintenance olaparib. Ann Oncol 2019;30:1959–1968.
Chiorean EG, Guthrie KA, Philip PA, Randomized phase II study of second-line modified FOLFIRI with PARP inhibitor ABT-888 (Veliparib) (NSC-737664) versus FOLFIRI in metastatic pancreatic cancer (mPC): SWOG S1513 [abstract]. J Clin Oncol 2019;37(Suppl):Abstract 4014.
- Search Google Scholar
- Export Citation
)| false , Chiorean EG , Guthrie KA , Philip PA Randomized phase II study of second-line modified FOLFIRI with PARP inhibitor ABT-888 (Veliparib) (NSC-737664) versus FOLFIRI in metastatic pancreatic cancer (mPC): SWOG S1513 [abstract]. J Clin Oncol 2019;37(Suppl):Abstract 4014. 10.1200/JCO.2019.37.15_suppl.4014
Lin KK, Harrell MI, Oza AM, BRCA reversion mutations in circulating tumor DNA predict primary and acquired resistance to the PARP inhibitor rucaparib in high-grade ovarian carcinoma. Cancer Discov 2019;9: 210–219.
- Search Google Scholar
- Export Citation
)| false , Lin KK , Harrell MI , Oza AM BRCA reversion mutations in circulating tumor DNA predict primary and acquired resistance to the PARP inhibitor rucaparib in high-grade ovarian carcinoma. Cancer Discov 2019; 9: 210– 219. 30425037 10.1158/2159-8290.CD-18-0715
Miller AL, Fehling SC, Garcia PL, The BET inhibitor JQ1 attenuates double-strand break repair and sensitizes models of pancreatic ductal adenocarcinoma to PARP inhibitors. EBioMedicine 2019;44:419–430.
Sohal DPS, Kennedy EB, Cinar P, Metastatic pancreatic cancer: ASCO guideline update. J Clin Oncol 2020;38:3217–3230.