Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Giovanna Damiaa · Cristiana Sessa b
a
Istituto di Ricerche Farmacologiche ‘Mario Negri’, Milan, Italy; b Oncology Institute of Southern Switzerland, San Giovanni Hospital, Bellinzona, Switzerland
Abstract In ovarian cancer, the clinical development of anticancer agents targeting DNA repair has been associated with significant results because of the elucidation of the different types of damages and repair systems, including PARP. The discovery of the BRCA mutation and its role in ovarian cancer and the clinical application of the concept of synthetic lethality have been the rationale for the successful testing of PARP inhibitors in BRCA mutated ovarian cancer patients. The recent knowledge of the molecular features of low grade ovarian cancer and the application of the concept of synthetic lethality also in this well-defined pathological entity have prompted the clinical evaluation of a combination of PI3K/MEK inhibitors, the first results of which have been already reported. © 2014 S. Karger AG, Basel
The DNA repair system has a key role in the maintenance of genetic stability, and deficiencies in its pathways have been associated with the development of several diseases, including cancer [1, 2]. In addition, many anticancer agents interfere with DNA, and the tumor cell DNA repair capacity can affect the cellular sensitivity/resistance to radiochemotherapy. In the last two decades, the mechanisms of the human DNA repair pathways have been unraveled generating hypotheses and questions for clinical studies. In particular, two questions are of outmost importance [1, 3]. (1) Since the tumor repair status is a determinant of sensitivity to radiochemotherapy, can we identify prognostic or predictive biomarkers? (2) Can deficiency in tumor DNA repair be therapeutically targeted according to the synthetic lethality concept?
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Targeting DNA Repair
DNA damage
TLS
(Translesion DNA synthesis)
FA
DNA repair pathways
DR
BER
MMR
Damaged base O6MeG Glycosylase
APE1
PARP
MLH1-PMS2, MLH1-PMS1
PCNA, XRCC1, RFC, PCNA, EXO, RFC, LIGASES, PoI RPA, Pold/e, Ligase DNA polβ
DSBR
NER
GGR/TCR
Misparing, Bulky lesions insertion, deletion MSH2/6, MSH2/3
ATG
(Fanconi anemia)
RPA, XPA, XPC, TFIIH
Double strand breaks HR
NHEJ
MRN complex
Ku70/80
ATM/ATR
DNA-PKcs
XPA, XPB, RAD52 epistasis XPD, XPG, group, XRCC2/3, ERCC1-XPF and others PCNA, Lig1, RFC, Pol
DNA PoIµ, XRCC4, Lig4
PCNA, Lig1, RFC, Pold/e
Fig. 1. Schematic representation of the main DNA repair pathways. From Damia and D’Incalci [4].
Common types of DNA damage are produced by reactive drugs which covalently bind to DNA as parent compound or metabolites. These alkylating agents produce adducts with modification of single or different bases located in the same DNA strand (intrastrand crosslinks) or on opposite DNA strands (interstrand crosslinks) which block the replication fork and cause double-strand breaks (DSBs). DSBs are the most lethal of all DNA lesions and are caused by ionizing radiation, free radicals and many anticancer agents including cisplatin [4]. DSBs are repaired by the error-free repair system of the homologous recombination (HR) which is predominant in the G2 phase, or by the error-prone pathway nonhomologous end joining repair, which is predominant in the G0/G1 and which could be mutagenic (fig. 1). Nucleotide excision repair (NER) is a multistep process involved in the removal of large helix-distorting adducts on DNA that occur by chemical modification of DNA bases upon exposure to environmental mutagens, such as ultraviolet light, tobacco smoke, reactive oxygen species (ROS), radiation and chemotherapeutic agents. In particular, platinum antitumor activity is mediated by the formation of DNA adducts, most
90
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Main DNA Repair Pathways
of which are repaired through NER alone or in conjunction with HR and Fanconi anemia (FA) pathways. Cisplatin has a key role in the treatment of ovarian cancer, and the development of platinum-taxane combinations had improved the 5-year survival rate in advanced disease [5]. However, there is a subset of patients showing intrinsic resistance to a platinum-based treatment or patients who, after an initial response, become resistant to a platinum-based therapy. The early identification of nonresponders to a given therapy (i.e. platinum) would help in selecting the therapy, avoiding the exposure to potentially toxic ineffective treatments in some patients while treating those more likely to respond to them.
During the last decade, ERCC1 (excision repair complementation group 1) has emerged as a promising biomarker in the management of several cancers and is one of the most studied proteins as prognostic and predictive biomarker for platinumbased chemotherapy. ERCC1 works in conjunction with xeroderma pigmentosum, complementation group F (XPF), forming a heterodimer which plays a key role in the NER pathway and in the repair of intra- and interstrand crosslinks. The gene is located on chromosome 19q13.2-q13.3, contains 10 exons and codes for at least four different isoforms by alternative slicing. These isoforms are still largely uncharacterized. The heterodimer ERCC1XPF, the catalytic activity of which is harbored by XPF, is able to recognize and incise branched double-single DNA structure, allowing the cleavage and elimination of the damaged strand. XPF-ERCC1 is the last factor to join the mammalian NER incision complex. The dimeric organization is critical for stability and catalytic activity of XPF, and ERCC1 and XPF are unstable without each other [6]. Initial preclinical data suggested that ovarian cancer cell lines overexpressing ERCC1 were more resistant to the cytotoxicity of platinum-based treatment [7, 8], and that the lack of ERCC1 [9] or its downregulation by siRNA or antisense strategies could sensitize cells to platinum. These data indicate that ERCC1 might affect the cellular response to platinum and that it could predict the response to a platinumbased therapy; the hypothesis is that low ERCC1 tumor levels are associated with low tumor DNA repair capacity and higher platinum sensitivity, while high ERCC1 levels could correspond to a greater tumor DNA repair capacity and decreased platinum sensitivity. ERCC1 as prognostic or predictive biomarker has been studied at the genomic (analysis of single-nucleotide polymorphism), transcriptional (by quantitative realtime polymerase chain reaction), and protein level (by immunohistochemistry, IHC) in both retrospective and prospective studies.
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
91
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Is ERCC1 a Predictive Biomarker of Response to Platinum Compounds?
Clinical Results
92
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
The predictive/prognostic role of ERCC1 has been best studied in non-small cell lung cancer (NSCLC). In the International Adjuvant Lung Trial-bio (IALT-bio), ERCC1 was assessed by IHC in 867 NSCLC patients [11]; ERCC1-negative patients treated with platinum-based therapy survived longer than ERCC1-positive patients, implying that ERCC1-positive patients did not benefit from an adjuvant chemotherapy. Because of the lack of confirmation of the predictive role of ERCC1 of a benefit from cisplatin-based adjuvant chemotherapy in two independent randomized trials, the IALT biology cohort samples were again stained; results discordant with the initial report of 2005 were achieved suggesting a change in the performance of the 8F1 antibody [12]. The predictive and prognostic value of ERCC1 has been investigated in ovarian cancer with conflicting results. Low-level ERCC1 and BRCA1 correlated with improved survival in one study [13], and in another patients with elevated ERCC1 mRNA expression had a greater risk of disease progression than patients with low ERCC1 expression [14]. In a recently published paper, 408 ovarian cancer patients were assessed for ERCC1 positivity by IHC: no correlation could be found with clinical characteristics or platinum responsiveness [15]. The conflicting results on ERCC1 can be partly explained by the two different methods of assessment (IHC vs. RT-PCR), by the retrospective nature of the studies and by the limited sample size. There is no consensus on which method of assessment is better because the two techniques have not been used concomitantly, and there are conflicting results on the correlation between mRNA levels and protein expression by IHC. Both techniques have some limitations which make difficult the interpretation of the results and can partly explain their lack of consistencies. There are four different ERCC1 isoforms with different biological activity, and the primers used in RT-PRC detect all of them, while the only functional isoform ERCC1-202 should be specifically detected. In addition, there is a lack of consensus on the specificity of the 8F1 mouse monoclonal ERCC1, by far the most commonly used reagent in published studies. Another important drawback of IHC is the lack of a validated cutoff value to be applied in all participating institutions. Because of all these reasons, ERCC1 did not fulfill the expectations and cannot be considered a predictive biomarker of response to platinum compounds. It should not be used in clinical routine. More in general, the prospective application of a potential biomarker must be supported not only by the positive data of previous studies but also by their confirmation through a retrospective analysis of the published data.
AD
DBD
a
Catalytic domain
PARP1 Znl
Znll NLS
Znlll
BRCT
*H *Y
WGR
*E
DNA strand break detection by PARP1
b
Binding of PARP1 to DNA strand breaks AMPK activation
AMP
NUDIX
ADP-ribose
NAD*
AMP + PPi
PARP1
PARG ARH3
Nicotinamide
ATP + PRPP
PARP inhibitors
Poly (ADP-ribosyl)ation of PARP1
Transient recruitment and non-covalent and covalent modifications fo various proteins XRCCI
Recruitment of XRCCI and DNA ligase III to SSBs and repair by BER
Histone HI
Altered chromatin binding during DNA damage and transcription
DNA-PKcs, Ku70 and Ku80
ATM
DSB repair by NHEJ
DSB repair by HR and checkpoin activation
MREII
Topoisomerase 1
Other unknown targets
HR and Genomic restrating maintenance of collapsed replication forks
Fig. 2. Structural and functional characteristics of PARP1. a PARP1 with its DNA-binding (DBD), automodification (AD) and catalytic domains. b Consequences of PARP1 activation by DNA damage. From Rouleau et al. [16].
Can DNA Repair Dysregulation Be a Suitable Target?
Poly (ADP-ribose) polymerase (PARP) is a family of nuclear proteins with enzymatic scaffolding properties and recruiting ability for DNA repair proteins [16]. The most important is PARP1, which is involved in the base excision repair system, where it binds to the strand break and starts the synthesis of PAR on acceptor proteins, located on PARP1, or other proteins involved in DNA repair (fig. 2).
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
93
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
PARP1 and Synthetic Lethality
PARP1 is also involved in the repair of DSBs where it binds to the D protein kinase catalytic subunit and recruits ATM, MRE11 and topo I. The HR system works to repair DSBs through a series of enzymatic activations which involve ATM, RAD51 and BRCA1/BRCA2 tumor suppressor genes. Synthetic lethality is a phenomenon by which cell death is caused by the inactivation (by mutation or inhibition) of two genes or their products or two pathways, when inactivation of either alone is not lethal [17]. The concept of synthetic lethality was the rationale for the clinical development of PARP1 inhibitors in BRCA1/BRCA2-mutated ovarian and breast cancer. Theoretically, synthetic lethality might offer a greater selectivity because of the heterozygous status of normal tissues and the homozygous status of the tumor genotype. HR repair defects also involve mutations or inactivation of the BRCA/FA pathway, mutations in BRCA1/BRCA2 genes (like hypermethylation of the BRCA1 promoter or amplification of the EMSY gene with inactivation of BRCA2), decreased expression of proteins involved in HR (like RAD51, ATM) and PTEN deficiency. Overall, this dysfunction of the HR system corresponds to the clinical phenotype of BRCAness that some sporadic tumors share with familiar BRCA cancers, with improved response and long survival with platinum agents. The estimated frequency of BRCAness in high-grade serous ovarian cancer is about 50% (15% in women with germline BRCA1/BRCA2 mutation and 35% in patients with acquired defects in the HR pathway) and was the rationale for testing the antitumor effect of PARP inhibitors in ovarian cancer regardless of the BRCA mutation status of patients.
Olaparib (AZD 2281) is the PARP1 inhibitor at the more advanced stage of development; phase I and phase II single and combination studies have been completed and phase III studies are ready to start. A significant step in the clinical development of olaparib in ovarian cancer was the report of 46% overall response rate in the phase I-phase II proof-of-concept study performed in germline BRCA mutated patients, with also a very promising 33% response rate in platinum-resistant disease [18]. It was first thought that the presence of BRCA mutation was not predictive of response to olaparib because of the observation of 22% antitumor activity in sporadic cases of ovarian cancer [19]. However, it is clear today that the presence of BRCA mutation is associated with the greatest clinical effect. 265 platinum-sensitive recurrent high-grade serous ovarian cancers, in response to the last treatment with platinum, were treated in a randomized placebo-controlled phase II study with olaparib as maintenance [20, 21]. Germline and somatic mutation status was known in 218 and 209 patients, respectively. The greatest progression-free survival (PFS) benefit with
94
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Clinical Achievements with PARP Inhibitors
olaparib was in patients with germline mutation (11.2 vs. 4.1 months), but the PFS benefit was still present when the somatic mutation data were included (11.2 vs. 4.3 months). The results of this study indicate that the greatest improvement by giving olaparib as maintenance is achieved in BRCA-mutated patients, with a PFS of 6.9 months. Confirmatory phase III trials in patients with BRCA mutation will start at the end of 2013. The successful clinical development of PARP inhibitors has shown that a targetbased approach to treatment is feasible. To improve the results further, it is necessary to identify predictive markers for HR dysfunction and response to PARPi, and to establish the optimal timing of therapy. In addition, repeated tumor biopsies at recurrence might screen for HR dysfunction, while preclinical studies could clarify the potential interactions with other signaling pathways.
Ovarian cancer can originate from endometriosis, from the peritoneum or from the fimbriae of the fallopian tube, the latter being responsible for hereditary tumors [22]. Mutant p53 can be detected in fimbriae, and it is associated with in situ and invasive serous cancer of the tube; up to 80% of familial ovarian cancers arising from BRCA1/BRCA2 mutations could originate from the tube and implant upon or cover the ovary rather than originating from it. Different histotypes correspond to distinct entities with different genetic abnormalities, gene expression profile and clinical behavior. Kurman and Shih [21] have proposed a unifying theory on the origin and pathogenesis of epithelial ovarian cancer. Type II are high-grade serous, endometrioid, undifferentiated and malignant mixed mesodermal tumors (carcinosarcoma). They have a more aggressive behavior, advanced stage and sensitivity to platinum-based therapy [21]. P53 and BRCA1/2 mutations are almost exclusive of this class, while the PI3K pathway is hyperactivated in only a minority of cases. Type I or low-grade carcinomas are more frequently diagnosed at an early stage, have a slow growth rate and develop platinum resistance while receiving a platinumbased therapy as primary treatment. Type I group include low-grade serous and endometrioid, mucinous and well-differentiated clear cell. The distinct molecular dysregulations of type I carcinomas include KRAS mutations (20–30% mostly in mucinous), BRAF mutations (20–30% mostly in serous), PI3K in clear cell, PTEN deletion, mostly in endometrioid and clear cell types. Distinct well-established precursors have been identified, endometriosis for both endometrioid and clear cell carcinomas and borderline tumors for low-grade serous and mucinous carcinomas.
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
95
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Personalizing Therapy in Ovarian Cancer
Distinctive therapeutic strategies have been defined, with antiangiogenics and DNA-interacting agents for type II, MEK inhibitors [23] or combination of PI3K/ MEK inhibitors [24] for type I, the latter because of the stimulation of the two pathways by the ligand-induced activation of receptor tyrosine kinases and the concept of synthetic lethality. In this target-oriented therapeutic approach, the characterization of the molecular profile at baseline and during treatment by repeated biopsies, particularly at progression, could be of help in the validation of the target and of potential pharmacodynamic and/or predictive biomarkers. The most interesting results have been reported with the MEK1/2 inhibitor selumetinib in a phase II study in 51 women with recurrent low-grade serous ovarian or peritoneal carcinoma [23]. Eight patients (15%) achieved an objective response with a PFS of 11 months, and 34 (65%) had stable disease. The BRAF and KRAS mutation status could be retrospectively assessed in the primary tumor in 82% of cases, of whom 41% of cases were KRAS mutants. No correlation could be found between tumor response and mutation status. The most severe toxicities were cutaneous (12% grade 3) and gastrointestinal (10% grade 3) but not myelosuppression. A series of phase III studies comparing MEK inhibitors versus investigators’ choice chemotherapy or best supportive care will be implemented in the near future. From a clinical point of view, the results achieved with MEK inhibitors in lowgrade serous ovarian cancer are very promising because of the current lack of available effective treatments, the duration of the response and the clinical confirmation of a preclinical hypothesis. To improve further, a careful planning of future studies with adequate criteria for patient selection (with KRAS, BRAF, PI3K mutations), criteria for tumor assessment, standardized management of toxicities and repeated tumor biopsies is needed.
References
96
6 Shuck SC, Short EA, Turchi JJ: Eukaryotic nucleotide excision repair: from understanding mechanisms to influencing biology. Cell Res 2008; 18: 64– 72. 7 Gossage L, Madhusudan S: Current status of excision repair cross complementing-group 1 (ERCC1) in cancer. Cancer Treat Rev 2007;33:565–577. 8 Ferry KV, Hamilton TC, Johnson SW: Increased nucleotide excision repair in cisplatin-resistant ovarian cancer cells: role of ERCC1-XPF. Biochem Pharmacol 2000;60:1305–1313. 9 Damia G, Imperatori L, Stefanini M, D’Incalci M: Sensitivity of CHO mutant cell lines with specific defects in nucleotide excision repair to different anticancer agents. Int J Cancer 1996;66:779–783.
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
1 Jalal S, Earley JN, Turchi JJ: DNA repair: from genome maintenance to biomarker and therapeutic target. Clin Cancer Res 2011;17:6973–6984. 2 Negrini S, Gorgoulis VG, Halazonetis TD: Genomic instability – an evolving hallmark of cancer. Nat Rev Mol Cell Biol 2010;11:220–228. 3 Curtin NJ: DNA repair dysregulation from cancer driver to therapeutic target. Nat Rev Cancer 2012;12: 801–817. 4 Damia G, D’Incalci M: Targeting DNA repair as a promising approach in cancer therapy. Eur J Cancer 2007;43:1791–1801. 5 McGuire WP, Hoskins WJ, Brady MF, et al: Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996;334:1–6.
18 Fang PC, Yap TA, Boss DS, et al: 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. 19 Gelmon KA, Fischkowitz M, Mackeny M, et al: Olaparib in patients with recurrent high-grade serous or poorly differentiated ovarian carcinoma or triple negative breast cancer: a phase II multicenter open label non randomized study. Lancet Oncol 2011;12:852–861. 20 Ledermann J, Harter P, Gourley C, et al: Olaparib maintenance therapy in patients with platinum sensitive relapsed serous ovarian cancer (SOC) and a BRCA mutation (BRCAm) (abstract). J Clin Oncol 2013;31(suppl):5505. 21 Kurman RJ, Shih IM: The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. Am J Surg Pathol 2010;34:433–443. 22 Bast RC: Molecular approaches to personalizing management of ovarian cancer. Ann Oncol 2011; 22(suppl 8):VIII5–VIII15. 23 Farley JF, Brady WE, Vathipadiekal V, et al: Selumetinib in woman with recurrent low grade serous carcinoma of the ovary or peritoneum: an open label, single-arm, phase 2 study. Lancet Oncol 2013; 14: 134–140. 24 Bedard P, Tabernero J, Kurzrock R, Britten CD, Stathis A, Perez-Garcia JM, Zubel A, Le NT, Carter K, Bellew KM, Gallarati C, Niazi F, Demanse D, De Buck SS, Sessa C: A phase lb, open-label, multicenter, dose-escalation study of the oral pan-PI3K inhibitor BKM120 in combination with the oral MEK1/2 inhibitor GSK1120212 in patients (pts) with selected advanced solid tumors (abstract). J Clin Oncol 2012; 30(suppl):3003.
Cristiana Sessa Oncology Institute of Southern Switzerland San Giovanni Hospital 6500 Bellinzona (Switzerland) E-Mail Cristiana.sessa @ eoc.ch
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
97
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
10 Selvakumaran M, Pisarcik DA, Bao R, Yeung AT, Hamilton TC: Enhanced cisplatin cytotoxicity by disturbing the nucleotide excision repair pathway in ovarian cancer cell lines. Cancer Res 2003; 63: 1311– 1316. 11 Olaussen KA, Duvant A, Fouret P, Brambilla E, Audrè F, Maddag V, et al: DNA repair by ERCC 1 in non small cell lung cancer and cisplatin based adjuvant chemotherapy. N Engl J Med 2006; 355: 983– 991. 12 Friboulet L, Olaussen KA, Pignas J, et al: ERCC1 isoform expression and DNA repair in non-small cell lung cancer. N Engl J Med 2013;368:1101–1110. 13 Weberpals J, Garbuio K, O’Brien A, Clark-Knowles K, Doucette S, Antoniouk O, Goss G, Dimitroulakos J: The DNA repair proteins BRCA1 and ERCC1 as predictive markers in sporadic ovarian cancer. Int J Cancer 2009;124:806–815. 14 Walsh CS, Ogawa S, Karahashi H, Scoles DR, Pavelka JC, Tran H, Miller CW, Kawamata N, Ginther C, Dering J, Sanada M, Nannya Y, Slamon DJ, Koeffler HP, Karlan BY: ERCC5 is a novel biomarker of ovarian cancer prognosis. J Clin Oncol 2008; 26: 2952– 2958. 15 Rubatt J, Darcy MH, Tian C, et al: Pre-treatment tumor expression of ERCC1 in women with advanced stage epithelial ovarian cancer is not predictive of clinical outcomes: a gynecologic oncology group study. Gynecol Oncol 2012;125:421–426. 16 Rouleau M, Patel A, Hendzel MJ: PARP inhibition: PARP1 and beyond. Nat Rev Cancer 2010; 10: 293– 301. 17 Ashwort A: A synthetic lethality therapeutic approach: poly (ADP ribose) polymerase inhibitors for the treatment of cancer deficient in DNA double strand break repair. J Clin Oncol 2008; 26: 3785– 3790.
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Giovanna Damiaa · Cristiana Sessa b
a
Istituto di Ricerche Farmacologiche ‘Mario Negri’, Milan, Italy; b Oncology Institute of Southern Switzerland, San Giovanni Hospital, Bellinzona, Switzerland
Abstract In ovarian cancer, the clinical development of anticancer agents targeting DNA repair has been associated with significant results because of the elucidation of the different types of damages and repair systems, including PARP. The discovery of the BRCA mutation and its role in ovarian cancer and the clinical application of the concept of synthetic lethality have been the rationale for the successful testing of PARP inhibitors in BRCA mutated ovarian cancer patients. The recent knowledge of the molecular features of low grade ovarian cancer and the application of the concept of synthetic lethality also in this well-defined pathological entity have prompted the clinical evaluation of a combination of PI3K/MEK inhibitors, the first results of which have been already reported. © 2014 S. Karger AG, Basel
The DNA repair system has a key role in the maintenance of genetic stability, and deficiencies in its pathways have been associated with the development of several diseases, including cancer [1, 2]. In addition, many anticancer agents interfere with DNA, and the tumor cell DNA repair capacity can affect the cellular sensitivity/resistance to radiochemotherapy. In the last two decades, the mechanisms of the human DNA repair pathways have been unraveled generating hypotheses and questions for clinical studies. In particular, two questions are of outmost importance [1, 3]. (1) Since the tumor repair status is a determinant of sensitivity to radiochemotherapy, can we identify prognostic or predictive biomarkers? (2) Can deficiency in tumor DNA repair be therapeutically targeted according to the synthetic lethality concept?
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Targeting DNA Repair
DNA damage
TLS
(Translesion DNA synthesis)
FA
DNA repair pathways
DR
BER
MMR
Damaged base O6MeG Glycosylase
APE1
PARP
MLH1-PMS2, MLH1-PMS1
PCNA, XRCC1, RFC, PCNA, EXO, RFC, LIGASES, PoI RPA, Pold/e, Ligase DNA polβ
DSBR
NER
GGR/TCR
Misparing, Bulky lesions insertion, deletion MSH2/6, MSH2/3
ATG
(Fanconi anemia)
RPA, XPA, XPC, TFIIH
Double strand breaks HR
NHEJ
MRN complex
Ku70/80
ATM/ATR
DNA-PKcs
XPA, XPB, RAD52 epistasis XPD, XPG, group, XRCC2/3, ERCC1-XPF and others PCNA, Lig1, RFC, Pol
DNA PoIµ, XRCC4, Lig4
PCNA, Lig1, RFC, Pold/e
Fig. 1. Schematic representation of the main DNA repair pathways. From Damia and D’Incalci [4].
Common types of DNA damage are produced by reactive drugs which covalently bind to DNA as parent compound or metabolites. These alkylating agents produce adducts with modification of single or different bases located in the same DNA strand (intrastrand crosslinks) or on opposite DNA strands (interstrand crosslinks) which block the replication fork and cause double-strand breaks (DSBs). DSBs are the most lethal of all DNA lesions and are caused by ionizing radiation, free radicals and many anticancer agents including cisplatin [4]. DSBs are repaired by the error-free repair system of the homologous recombination (HR) which is predominant in the G2 phase, or by the error-prone pathway nonhomologous end joining repair, which is predominant in the G0/G1 and which could be mutagenic (fig. 1). Nucleotide excision repair (NER) is a multistep process involved in the removal of large helix-distorting adducts on DNA that occur by chemical modification of DNA bases upon exposure to environmental mutagens, such as ultraviolet light, tobacco smoke, reactive oxygen species (ROS), radiation and chemotherapeutic agents. In particular, platinum antitumor activity is mediated by the formation of DNA adducts, most
90
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Main DNA Repair Pathways
of which are repaired through NER alone or in conjunction with HR and Fanconi anemia (FA) pathways. Cisplatin has a key role in the treatment of ovarian cancer, and the development of platinum-taxane combinations had improved the 5-year survival rate in advanced disease [5]. However, there is a subset of patients showing intrinsic resistance to a platinum-based treatment or patients who, after an initial response, become resistant to a platinum-based therapy. The early identification of nonresponders to a given therapy (i.e. platinum) would help in selecting the therapy, avoiding the exposure to potentially toxic ineffective treatments in some patients while treating those more likely to respond to them.
During the last decade, ERCC1 (excision repair complementation group 1) has emerged as a promising biomarker in the management of several cancers and is one of the most studied proteins as prognostic and predictive biomarker for platinumbased chemotherapy. ERCC1 works in conjunction with xeroderma pigmentosum, complementation group F (XPF), forming a heterodimer which plays a key role in the NER pathway and in the repair of intra- and interstrand crosslinks. The gene is located on chromosome 19q13.2-q13.3, contains 10 exons and codes for at least four different isoforms by alternative slicing. These isoforms are still largely uncharacterized. The heterodimer ERCC1XPF, the catalytic activity of which is harbored by XPF, is able to recognize and incise branched double-single DNA structure, allowing the cleavage and elimination of the damaged strand. XPF-ERCC1 is the last factor to join the mammalian NER incision complex. The dimeric organization is critical for stability and catalytic activity of XPF, and ERCC1 and XPF are unstable without each other [6]. Initial preclinical data suggested that ovarian cancer cell lines overexpressing ERCC1 were more resistant to the cytotoxicity of platinum-based treatment [7, 8], and that the lack of ERCC1 [9] or its downregulation by siRNA or antisense strategies could sensitize cells to platinum. These data indicate that ERCC1 might affect the cellular response to platinum and that it could predict the response to a platinumbased therapy; the hypothesis is that low ERCC1 tumor levels are associated with low tumor DNA repair capacity and higher platinum sensitivity, while high ERCC1 levels could correspond to a greater tumor DNA repair capacity and decreased platinum sensitivity. ERCC1 as prognostic or predictive biomarker has been studied at the genomic (analysis of single-nucleotide polymorphism), transcriptional (by quantitative realtime polymerase chain reaction), and protein level (by immunohistochemistry, IHC) in both retrospective and prospective studies.
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
91
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Is ERCC1 a Predictive Biomarker of Response to Platinum Compounds?
Clinical Results
92
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
The predictive/prognostic role of ERCC1 has been best studied in non-small cell lung cancer (NSCLC). In the International Adjuvant Lung Trial-bio (IALT-bio), ERCC1 was assessed by IHC in 867 NSCLC patients [11]; ERCC1-negative patients treated with platinum-based therapy survived longer than ERCC1-positive patients, implying that ERCC1-positive patients did not benefit from an adjuvant chemotherapy. Because of the lack of confirmation of the predictive role of ERCC1 of a benefit from cisplatin-based adjuvant chemotherapy in two independent randomized trials, the IALT biology cohort samples were again stained; results discordant with the initial report of 2005 were achieved suggesting a change in the performance of the 8F1 antibody [12]. The predictive and prognostic value of ERCC1 has been investigated in ovarian cancer with conflicting results. Low-level ERCC1 and BRCA1 correlated with improved survival in one study [13], and in another patients with elevated ERCC1 mRNA expression had a greater risk of disease progression than patients with low ERCC1 expression [14]. In a recently published paper, 408 ovarian cancer patients were assessed for ERCC1 positivity by IHC: no correlation could be found with clinical characteristics or platinum responsiveness [15]. The conflicting results on ERCC1 can be partly explained by the two different methods of assessment (IHC vs. RT-PCR), by the retrospective nature of the studies and by the limited sample size. There is no consensus on which method of assessment is better because the two techniques have not been used concomitantly, and there are conflicting results on the correlation between mRNA levels and protein expression by IHC. Both techniques have some limitations which make difficult the interpretation of the results and can partly explain their lack of consistencies. There are four different ERCC1 isoforms with different biological activity, and the primers used in RT-PRC detect all of them, while the only functional isoform ERCC1-202 should be specifically detected. In addition, there is a lack of consensus on the specificity of the 8F1 mouse monoclonal ERCC1, by far the most commonly used reagent in published studies. Another important drawback of IHC is the lack of a validated cutoff value to be applied in all participating institutions. Because of all these reasons, ERCC1 did not fulfill the expectations and cannot be considered a predictive biomarker of response to platinum compounds. It should not be used in clinical routine. More in general, the prospective application of a potential biomarker must be supported not only by the positive data of previous studies but also by their confirmation through a retrospective analysis of the published data.
AD
DBD
a
Catalytic domain
PARP1 Znl
Znll NLS
Znlll
BRCT
*H *Y
WGR
*E
DNA strand break detection by PARP1
b
Binding of PARP1 to DNA strand breaks AMPK activation
AMP
NUDIX
ADP-ribose
NAD*
AMP + PPi
PARP1
PARG ARH3
Nicotinamide
ATP + PRPP
PARP inhibitors
Poly (ADP-ribosyl)ation of PARP1
Transient recruitment and non-covalent and covalent modifications fo various proteins XRCCI
Recruitment of XRCCI and DNA ligase III to SSBs and repair by BER
Histone HI
Altered chromatin binding during DNA damage and transcription
DNA-PKcs, Ku70 and Ku80
ATM
DSB repair by NHEJ
DSB repair by HR and checkpoin activation
MREII
Topoisomerase 1
Other unknown targets
HR and Genomic restrating maintenance of collapsed replication forks
Fig. 2. Structural and functional characteristics of PARP1. a PARP1 with its DNA-binding (DBD), automodification (AD) and catalytic domains. b Consequences of PARP1 activation by DNA damage. From Rouleau et al. [16].
Can DNA Repair Dysregulation Be a Suitable Target?
Poly (ADP-ribose) polymerase (PARP) is a family of nuclear proteins with enzymatic scaffolding properties and recruiting ability for DNA repair proteins [16]. The most important is PARP1, which is involved in the base excision repair system, where it binds to the strand break and starts the synthesis of PAR on acceptor proteins, located on PARP1, or other proteins involved in DNA repair (fig. 2).
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
93
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
PARP1 and Synthetic Lethality
PARP1 is also involved in the repair of DSBs where it binds to the D protein kinase catalytic subunit and recruits ATM, MRE11 and topo I. The HR system works to repair DSBs through a series of enzymatic activations which involve ATM, RAD51 and BRCA1/BRCA2 tumor suppressor genes. Synthetic lethality is a phenomenon by which cell death is caused by the inactivation (by mutation or inhibition) of two genes or their products or two pathways, when inactivation of either alone is not lethal [17]. The concept of synthetic lethality was the rationale for the clinical development of PARP1 inhibitors in BRCA1/BRCA2-mutated ovarian and breast cancer. Theoretically, synthetic lethality might offer a greater selectivity because of the heterozygous status of normal tissues and the homozygous status of the tumor genotype. HR repair defects also involve mutations or inactivation of the BRCA/FA pathway, mutations in BRCA1/BRCA2 genes (like hypermethylation of the BRCA1 promoter or amplification of the EMSY gene with inactivation of BRCA2), decreased expression of proteins involved in HR (like RAD51, ATM) and PTEN deficiency. Overall, this dysfunction of the HR system corresponds to the clinical phenotype of BRCAness that some sporadic tumors share with familiar BRCA cancers, with improved response and long survival with platinum agents. The estimated frequency of BRCAness in high-grade serous ovarian cancer is about 50% (15% in women with germline BRCA1/BRCA2 mutation and 35% in patients with acquired defects in the HR pathway) and was the rationale for testing the antitumor effect of PARP inhibitors in ovarian cancer regardless of the BRCA mutation status of patients.
Olaparib (AZD 2281) is the PARP1 inhibitor at the more advanced stage of development; phase I and phase II single and combination studies have been completed and phase III studies are ready to start. A significant step in the clinical development of olaparib in ovarian cancer was the report of 46% overall response rate in the phase I-phase II proof-of-concept study performed in germline BRCA mutated patients, with also a very promising 33% response rate in platinum-resistant disease [18]. It was first thought that the presence of BRCA mutation was not predictive of response to olaparib because of the observation of 22% antitumor activity in sporadic cases of ovarian cancer [19]. However, it is clear today that the presence of BRCA mutation is associated with the greatest clinical effect. 265 platinum-sensitive recurrent high-grade serous ovarian cancers, in response to the last treatment with platinum, were treated in a randomized placebo-controlled phase II study with olaparib as maintenance [20, 21]. Germline and somatic mutation status was known in 218 and 209 patients, respectively. The greatest progression-free survival (PFS) benefit with
94
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Clinical Achievements with PARP Inhibitors
olaparib was in patients with germline mutation (11.2 vs. 4.1 months), but the PFS benefit was still present when the somatic mutation data were included (11.2 vs. 4.3 months). The results of this study indicate that the greatest improvement by giving olaparib as maintenance is achieved in BRCA-mutated patients, with a PFS of 6.9 months. Confirmatory phase III trials in patients with BRCA mutation will start at the end of 2013. The successful clinical development of PARP inhibitors has shown that a targetbased approach to treatment is feasible. To improve the results further, it is necessary to identify predictive markers for HR dysfunction and response to PARPi, and to establish the optimal timing of therapy. In addition, repeated tumor biopsies at recurrence might screen for HR dysfunction, while preclinical studies could clarify the potential interactions with other signaling pathways.
Ovarian cancer can originate from endometriosis, from the peritoneum or from the fimbriae of the fallopian tube, the latter being responsible for hereditary tumors [22]. Mutant p53 can be detected in fimbriae, and it is associated with in situ and invasive serous cancer of the tube; up to 80% of familial ovarian cancers arising from BRCA1/BRCA2 mutations could originate from the tube and implant upon or cover the ovary rather than originating from it. Different histotypes correspond to distinct entities with different genetic abnormalities, gene expression profile and clinical behavior. Kurman and Shih [21] have proposed a unifying theory on the origin and pathogenesis of epithelial ovarian cancer. Type II are high-grade serous, endometrioid, undifferentiated and malignant mixed mesodermal tumors (carcinosarcoma). They have a more aggressive behavior, advanced stage and sensitivity to platinum-based therapy [21]. P53 and BRCA1/2 mutations are almost exclusive of this class, while the PI3K pathway is hyperactivated in only a minority of cases. Type I or low-grade carcinomas are more frequently diagnosed at an early stage, have a slow growth rate and develop platinum resistance while receiving a platinumbased therapy as primary treatment. Type I group include low-grade serous and endometrioid, mucinous and well-differentiated clear cell. The distinct molecular dysregulations of type I carcinomas include KRAS mutations (20–30% mostly in mucinous), BRAF mutations (20–30% mostly in serous), PI3K in clear cell, PTEN deletion, mostly in endometrioid and clear cell types. Distinct well-established precursors have been identified, endometriosis for both endometrioid and clear cell carcinomas and borderline tumors for low-grade serous and mucinous carcinomas.
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
95
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
Personalizing Therapy in Ovarian Cancer
Distinctive therapeutic strategies have been defined, with antiangiogenics and DNA-interacting agents for type II, MEK inhibitors [23] or combination of PI3K/ MEK inhibitors [24] for type I, the latter because of the stimulation of the two pathways by the ligand-induced activation of receptor tyrosine kinases and the concept of synthetic lethality. In this target-oriented therapeutic approach, the characterization of the molecular profile at baseline and during treatment by repeated biopsies, particularly at progression, could be of help in the validation of the target and of potential pharmacodynamic and/or predictive biomarkers. The most interesting results have been reported with the MEK1/2 inhibitor selumetinib in a phase II study in 51 women with recurrent low-grade serous ovarian or peritoneal carcinoma [23]. Eight patients (15%) achieved an objective response with a PFS of 11 months, and 34 (65%) had stable disease. The BRAF and KRAS mutation status could be retrospectively assessed in the primary tumor in 82% of cases, of whom 41% of cases were KRAS mutants. No correlation could be found between tumor response and mutation status. The most severe toxicities were cutaneous (12% grade 3) and gastrointestinal (10% grade 3) but not myelosuppression. A series of phase III studies comparing MEK inhibitors versus investigators’ choice chemotherapy or best supportive care will be implemented in the near future. From a clinical point of view, the results achieved with MEK inhibitors in lowgrade serous ovarian cancer are very promising because of the current lack of available effective treatments, the duration of the response and the clinical confirmation of a preclinical hypothesis. To improve further, a careful planning of future studies with adequate criteria for patient selection (with KRAS, BRAF, PI3K mutations), criteria for tumor assessment, standardized management of toxicities and repeated tumor biopsies is needed.
References
96
6 Shuck SC, Short EA, Turchi JJ: Eukaryotic nucleotide excision repair: from understanding mechanisms to influencing biology. Cell Res 2008; 18: 64– 72. 7 Gossage L, Madhusudan S: Current status of excision repair cross complementing-group 1 (ERCC1) in cancer. Cancer Treat Rev 2007;33:565–577. 8 Ferry KV, Hamilton TC, Johnson SW: Increased nucleotide excision repair in cisplatin-resistant ovarian cancer cells: role of ERCC1-XPF. Biochem Pharmacol 2000;60:1305–1313. 9 Damia G, Imperatori L, Stefanini M, D’Incalci M: Sensitivity of CHO mutant cell lines with specific defects in nucleotide excision repair to different anticancer agents. Int J Cancer 1996;66:779–783.
Damia · Sessa Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
1 Jalal S, Earley JN, Turchi JJ: DNA repair: from genome maintenance to biomarker and therapeutic target. Clin Cancer Res 2011;17:6973–6984. 2 Negrini S, Gorgoulis VG, Halazonetis TD: Genomic instability – an evolving hallmark of cancer. Nat Rev Mol Cell Biol 2010;11:220–228. 3 Curtin NJ: DNA repair dysregulation from cancer driver to therapeutic target. Nat Rev Cancer 2012;12: 801–817. 4 Damia G, D’Incalci M: Targeting DNA repair as a promising approach in cancer therapy. Eur J Cancer 2007;43:1791–1801. 5 McGuire WP, Hoskins WJ, Brady MF, et al: Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996;334:1–6.
18 Fang PC, Yap TA, Boss DS, et al: 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. 19 Gelmon KA, Fischkowitz M, Mackeny M, et al: Olaparib in patients with recurrent high-grade serous or poorly differentiated ovarian carcinoma or triple negative breast cancer: a phase II multicenter open label non randomized study. Lancet Oncol 2011;12:852–861. 20 Ledermann J, Harter P, Gourley C, et al: Olaparib maintenance therapy in patients with platinum sensitive relapsed serous ovarian cancer (SOC) and a BRCA mutation (BRCAm) (abstract). J Clin Oncol 2013;31(suppl):5505. 21 Kurman RJ, Shih IM: The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. Am J Surg Pathol 2010;34:433–443. 22 Bast RC: Molecular approaches to personalizing management of ovarian cancer. Ann Oncol 2011; 22(suppl 8):VIII5–VIII15. 23 Farley JF, Brady WE, Vathipadiekal V, et al: Selumetinib in woman with recurrent low grade serous carcinoma of the ovary or peritoneum: an open label, single-arm, phase 2 study. Lancet Oncol 2013; 14: 134–140. 24 Bedard P, Tabernero J, Kurzrock R, Britten CD, Stathis A, Perez-Garcia JM, Zubel A, Le NT, Carter K, Bellew KM, Gallarati C, Niazi F, Demanse D, De Buck SS, Sessa C: A phase lb, open-label, multicenter, dose-escalation study of the oral pan-PI3K inhibitor BKM120 in combination with the oral MEK1/2 inhibitor GSK1120212 in patients (pts) with selected advanced solid tumors (abstract). J Clin Oncol 2012; 30(suppl):3003.
Cristiana Sessa Oncology Institute of Southern Switzerland San Giovanni Hospital 6500 Bellinzona (Switzerland) E-Mail Cristiana.sessa @ eoc.ch
Successes and Limitations of Targeted Cancer Therapy in Ovarian Cancer Peters S, Stahel RA (eds): Successes and Limitations of Targeted Cancer Therapy. Prog Tumor Res. Basel, Karger, 2014, vol 41, pp 89–97 (DOI: 10.1159/000355905)
97
Downloaded by: Kainan University 203.64.11.45 - 1/10/2015 12:59:58 AM
10 Selvakumaran M, Pisarcik DA, Bao R, Yeung AT, Hamilton TC: Enhanced cisplatin cytotoxicity by disturbing the nucleotide excision repair pathway in ovarian cancer cell lines. Cancer Res 2003; 63: 1311– 1316. 11 Olaussen KA, Duvant A, Fouret P, Brambilla E, Audrè F, Maddag V, et al: DNA repair by ERCC 1 in non small cell lung cancer and cisplatin based adjuvant chemotherapy. N Engl J Med 2006; 355: 983– 991. 12 Friboulet L, Olaussen KA, Pignas J, et al: ERCC1 isoform expression and DNA repair in non-small cell lung cancer. N Engl J Med 2013;368:1101–1110. 13 Weberpals J, Garbuio K, O’Brien A, Clark-Knowles K, Doucette S, Antoniouk O, Goss G, Dimitroulakos J: The DNA repair proteins BRCA1 and ERCC1 as predictive markers in sporadic ovarian cancer. Int J Cancer 2009;124:806–815. 14 Walsh CS, Ogawa S, Karahashi H, Scoles DR, Pavelka JC, Tran H, Miller CW, Kawamata N, Ginther C, Dering J, Sanada M, Nannya Y, Slamon DJ, Koeffler HP, Karlan BY: ERCC5 is a novel biomarker of ovarian cancer prognosis. J Clin Oncol 2008; 26: 2952– 2958. 15 Rubatt J, Darcy MH, Tian C, et al: Pre-treatment tumor expression of ERCC1 in women with advanced stage epithelial ovarian cancer is not predictive of clinical outcomes: a gynecologic oncology group study. Gynecol Oncol 2012;125:421–426. 16 Rouleau M, Patel A, Hendzel MJ: PARP inhibition: PARP1 and beyond. Nat Rev Cancer 2010; 10: 293– 301. 17 Ashwort A: A synthetic lethality therapeutic approach: poly (ADP ribose) polymerase inhibitors for the treatment of cancer deficient in DNA double strand break repair. J Clin Oncol 2008; 26: 3785– 3790.
