Lung cancer is one of the leading causes of death worldwide. The standard of care for advanced small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) includes platinumbased chemotherapy.
Nikita T. Patel, MD
Winship Cancer Institute
Emory University, Atlanta, GA
Conor Steuer, MD
Winship Cancer Institute
Emory University, Atlanta, GA
Suresh S. Ramalingam, MD
Winship Cancer Institute
Emory University, Atlanta, GA
Lung cancer is one of the leading causes of death worldwide. The standard of care for advanced small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) includes platinumbased chemotherapy. However, the emergence of novel targeted therapies has reshaped the landscape and is shifting the field toward individualized therapy. Poly-ADP ribose polymerase (PARP)-1 is a protein that is primarily involved in DNA repair. PARP also impacts a number of important cellular functions, and has thus emerged as a target for cancer therapy. Recently reported data in lung cancer suggest that the efficacy of platinum-based chemotherapy is enhanced when used in combination with PARP inhibition. In this review, we discuss the structure and function of PARP proteins and the mechanism of action of PARP inhibitors; additionally, clinical trial data of various PARP-1 inhibitors in lung cancer are summarized. PARP-1 inhibitors have been proven to be effective at potentiating the effects of chemotherapy and radiation, justifying their investigation in lung cancer.
Lung cancer has emerged as the model for development of personalized therapy in oncology. Despite the grim survival statistics for advanced-stage lung cancer, novel targeted therapies that are directed toward specific molecular subtypes have caused major changes in the natural history of this disease.
The role of systemic therapy for nearly all stages of lung cancer is now well established. Cytotoxic chemotherapy has also undergone major changes with the development of novel treatment regimens that are associated with robust efficacy with relatively minimal toxicity. These exciting developments have been driven largely by increasing knowledge regarding critical cell signaling pathways in patients with lung cancer and identification of targetable dominant oncogenic events.
It is no longer uncommon to experience long-term survival with certain molecular subtypes of lung cancer with the utilization of targeted therapies. However, even as we move forward with individualized therapy, a substantial subset of patients with lung cancer does not have a targetable molecular abnormality. Since chemotherapy and radiation remain integral to the treatment of lung cancer, efforts to improve their efficacy and to identify predictive biomarkers have been extensively undertaken.
Poly-ADP ribose protein (PARP)-1 inhibitors are an emerging class of agents that have the potential to play an important role in the treatment of a variety of cancers. Though the initial development of these agents was focused primarily on the treatment of breast and ovarian cancers, they are now being studied against both SCLC and NSCLC. This review focuses on the biological rationale for the use of PARP-1 inhibitors and discusses the current investigations and clinical trials with these agents in lung cancer.PARP-1 is a member of the PARP family of proteins that includes 17 proteins, of which PARP-1 and PARP-2 are considered to be the most important. Primarily, they are involved in DNA repair; however, the protein family is also known to be involved in DNA transcription, cellular signaling, cell-cycle regulation, and mitosis. Of these proteins, PARP-1 is the best described.1
First reported in the early 1950s, PARP-1 binds to multiple DNA structures, including single strand breaks (SSBs) in the DNA helix, double strand breaks (DSBs), as well as supercoils, crossovers, and cruciforms.2 PARP-1 also associates with chromatin by binding to proteins related to DNA transcription, including NF-kB, CTFC, YY-1, and AP-1.2,3In the same capacity, it also binds to various chromatin- modifying enzymes.4The development of PARP inhibitors for cancer therapy primarily focuses on PARP-1’s role in DNA repair.PARP-1 has multiple domains that serve a variety of functions. The catalytic domain of the enzyme is involved in DNA repair and has a moiety that cleaves ADP from NAD.5PARP-1 has been studied in at least 3 separate mechanisms of repair. These are base excision repair (BER), SSB repair, and DSB repair.6,7
While the role of PARP-1 in SSBs and DSBs has more frequently been the focus of studies, BER appears to be the most important function of the PARP-1 enzyme. Base excision repair involves repair of only 1 to 3 bases at a time. This can cause errors in repair by insertion of an unmatched base, which therefore leads to genomic mutations. If not repaired, this can result in SSBs.8
When DNA damage occurs, PARP-1 and PARP-2 are recruited to the site to assist with BER.9 Comparatively, homologous repair (HR) uses the sister chromatids or homologous chromosomes, which are undamaged, to repair the site. This results in repair with less chance of error. Homologous repair occurs primarily in the S and G2 phases of the cell cycle. Importantly, BRCA proteins have been implicated in this pathway.9
In SSBs, PARP-1 binds to the site of the SSB and transfers ADP-ribosyl moiety to the acceptor protein, thus making polymers (PAR). These polymers assist in the recruitment of various DNA repair proteins to the break site. If not repaired, SSBs can lead to DSBs.5 When DNA is damaged, there is immediate addition of PAR to the site of the break. The extent of addition is directly proportional to the extent of damage.6,7,10In a cell that appears salvageable, PARP- 1 promotes repair and cell survival ensues. However, if the extent of DNA damage is deemed unsalvageable, PARP-1 assists in cell death. This occurs in one of 2 ways: either depletion of nicotinamide adenine dinucleotide (NAD+) and, ultimately, adenosine triphosphate (ATP), resulting in cell necrosis, or translocation of apoptosis-inducing factor from the mitochondria to the nucleus, resulting in apoptosis.6
Because of its multiple mechanisms of action, PARP-1 interacts with many proteins involved with the detection of DNA damage. These include ATM kinase and p53. Similarly, it has been found to interact with proteins involved in DNA repair and protein recruitment, for example, XRCC-1.7This pathway of repair can be disrupted or inhibited by protein or enzymes deficiencies. ATM deficiency is known to disrupt the SSB repair process and has been shown to be present in 10% to 15% of NSCLCs.11,12Of note, PTEN deficiency as well asBRCAdeficiency have also been noted to disrupt the repair pathway.12Given these findings, inhibition of the PARP-1 enzyme with subsequent disruption of the repair pathway to induce cell death is an active area of cancer research.In addition to DNA repair, the PARP family has been implicated in other functions within the cell. These include cell division, specifically mitosis, and the regulation of cell membrane structures as well as the structure of the actin cytoskeleton.13A review by Schreiber et al divided the PARP family by their function, classifying the proteins as DNA-dependent PARPs (PARP-1 and PARP-2), tankyrases, CCCH-type PARPs, and macro-PARPs.14The remaining PARPs are currently unclassified.
Tankyrase is a protein within the PARP family (defined by its catalytic PARP domain) found at the telomere of human genes. It binds to a telomere protein, TRF-1 (telomeric repeat binding factor-1) and regulates telomere length through ADP-ribosylation. Promotion of ADP-ribosylation of TRF-1 has been shown to inhibit its ability to bind to human telomeres, therefore acting as a negative regulator of telomere length.15
The CCCH-type PARP consists of tiPARP, PARP-12, and PARP-13. These proteins share a similar domain consisting of zinc fingers, a WWE domain (a proteinprotein motif that contains 2 Trp residues and one Glu residue), and the PARP catalytic domain.14It has been localized in the hippocampus and may play a role in learning, behavior, and memory.16In addition, isoforms of PARP-13 are currently being investigated in retroviral resistance.17
Macro-PARPs are defined by a subfamily of proteins that link 1 to 3 macro domains to a PARP domain. They are considered to be cofactors in DNA transcription. The macro domain of macro-PARPs is involved in transcriptional repression and X chromosome inactivation.18Various macro-PARPs have been studied, each with a slightly different function. Some mediate transcriptional activation and upregulate T-cells, while others are found to have repressive activity.19,20They have also been implicated in B-cell motility.21 The multiple functions of the PARP family create a potential target for future treatments. Currently, PARP-1 is the best studied, and its inhibition has been shown to increase cell death.PARP inhibitors work through competitive inhibition, and compete with nicotinamide for the catalytic domain of the enzyme.5,22Without inhibition, the PARP-1 enzyme would bind to the site of DNA damage and, through the addition of NAD+, cause release of the enzyme from the site. This process recruits repair enzymes to DNA damage sites. PARP-1 inhibitors bind to the enzyme and prevent this release, therefore preventing access to the site by DNA repair enzymes (Figure 1). Murai et al describes this mechanism of “PARP trapping.” Their study suggested that when PARP-1 is recruited to the site of DNA damage, the PARP-1 inhibitor binds to the NAD moiety, described as a DNA-PARP complex. These complexes are cytotoxic to the cell and result in cell death.23
BRCAmutations have been studied extensively in regards to PARP inhibition.BRCA-1plays a role in surveillance for DNA damage, whileBRCA-2has a direct role in DNA repair through HR. A cell that is BRCA-deficient does not have the capacity to repair its DNA through HR. DNA repair mediated by PARP-1 serves as an alternative pathway inBRCA-deficient cells.5
First coined in 1946, the term synthetic lethality describes a process whereby 2 separate mutations that may not be significant on their own lead to cell death when combined. This concept has been well described in many cancer treatments. This phenomenon can be seen in PARP-1 inhibition andBRCAdeficiency. PARP-1 loss in a normal cell does not lead to cell death because the cell can overcome this deficiency by utilizing other DNA repair mechanisms. However, if the tumor is deficient in both PARP-1 (through inhibition) andBRCA(through mutation), the cells are left unrepaired, leading to cell death.9Similarly, a tumor that is deficient in a mechanism for HR, such as ATM deficiency or p53 deficiency, will exhibit synthetic lethality when exposed to a PARP inhibitor.24This concept opens the doors for treatment for not only breast and gynecologic cancers, which primarily exhibit theBRCAdeficiency, but also for tumor cells that have other mutations in the HR mechanism.
In recent years, the PARP inhibitors have demonstrated promising activity in gynecological cancers. Initially developed for cancers that were found to beBRCA-deficient, they are now found to have activity in cancers that do not exhibitBRCAmutations. Phase II trials have been completed with oral PARP inhibitors that studied efficacy and tolerability of olaparib in advancedBRCA-mutated ovarian cancer.25Importantly, olaparib has also been studied in ovarian cancers that do not exhibitBRCAmutations.
BMN-673 is currently undergoing phase I trials to evaluate toxicites.
Most phase I trials above were investigations in solid tumor malignancies, not limited to lung cancer.
Both Ledermann et al and Gelmon et al conducted phase II trials evaluating olaparib as maintenance therapy in relapsed, platinum-sensitive,BRCA-mutation negative ovarian cancer. They concluded that olaparib significantly improves progression-free survival (PFS). In the study conducted by Ledermann et al, 265 patients with high-grade serous ovarian cancer who had received 2 or more platinum-based regimens and had had a partial response (PR) or complete response (CR) to their most recent platinum-based regimen were enrolled. Of these, 136 patients were randomized to the olaparib group and 129 patients to the placebo group. Median PFS was significantly longer with olaparib than with placebo (8.4 months vs 4.8 months, respectively) upon completion of chemotherapy.26,27These promising data, specifically in cancers withoutBRCAmutations, prompted the study of PARP inhibitors in a variety of different cancers that respond to platinum therapy, including, but not limited to, lung cancer. PARP inhibitors are generally well tolerated in trials in patients with ovarian cancer (Table 1), and the promising results have prompted investigation in other solid tumors.Expression of PARP-1 is higher relative to other DNA repair enzymes in SCLC tumors. In vitro studies by Byers et al showed equal reduction in growth of SCLC cell lines by standard chemotherapy (cisplatin and etoposide) or with a PARP inhibitor, AZD2281. Investigators measured levels of PAR in the cell lines and found a dose-dependent inhibition of PAR with the administration of AZD2281. However, when chemotherapy and PARP-1 inhibitors were used in combination, there was a synergistic inhibition of cell survival.28In contrast, NSCLC exhibits a range of PARP-1 activity, and the study showed that expression of PARP-1 activity is directly correlated to its sensitivity to PARP inhibition.28
Paul et al demonstrated the concept of synthetic lethality inBRCA-deficient NSCLC. By using RNA interference, NSCLC cells were madeBRCA1-deficient and subsequently were found to be more sensitive to PARP-1 inhibition. Additionally, they showed thatBRCA-deficient, platinum-resistant cells were still sensitive to PARP inhibition. The authors suggest that in a normal cell, BAX and BAK proteins are in place to prevent mitochondrial apoptosis, while in aBRCA-deficient cell, these regulatory mechanisms are bypassed and cell death occurs independent of mitochondrial apoptosis.29Several PARP inhibitors are currently undergoing clinical trials in lung cancer to study these concepts (Table 2).BMN 673
In vitro, BMN 673 (BioMarin Pharmaceutical, Novato, CA) selectively targets tumor cells withBRCA1/2or PTEN deficiencies. It has been shown to be 20 to 200 times more potent than some of the other PARP inhibitors currently in clinical trials. It has also been suggested that BMN 673 is better at creating PARP-DNA complexes, resulting in better PARP trapping compared with the mechanism of competitive inhibition.30A phase I study is presently recruiting patients to determine the maximum tolerated dose (MTD) of BMN 673 in patients with recurrent or advanced solid tumor malignancies, including SCLC.31Early results show that BMN 673 has antitumor activity in patients with advanced previously treated SCLC. In a preliminary report, objective response was noted in 2 of 11 patients with refractory SCLC.32Currently, a phase I study of the combination of BMN 673 with carboplatin and paclitaxel in patients with advanced solid tumors is in development.33Other planned phase I studies include the combination of BMN673 with temozolomide and with irinotecan.
CR indicates complete response; MTD,maximum tolerated dose; DLT,dose-limiting toxicities. PD, pharmacodynamics; PFS, progression-free survival; PK, pharmacokinetics; PR, partial response.
Olaparib (AZD 2281; AstraZeneca, London, England), an oral PARP inhibitor, has been studied primarily inBRCA-positive breast and ovarian cancer. The dose-limiting toxicities include mood swings, fatigue, and thrombocytopenia.27In a separate phase I study at the NCI conducted by Rajan et al, olaparib was combined with cisplatin and gemcitabine in 23 patients with solid tumors that were refractory to previous treatments. The combination was not well tolerated due to severe myelosuppression. Approximately 35% of the patients had NSCLC. The trial demonstrated reduction in levels of PAR from over 100,000 pg/10 mg protein to about 40,000 pg/10 mg protein when olaparib was administered for 4 days. However, in some cases, levels recovered prior to the next cycle of chemotherapy.34Investigations in triple-negative breast cancer with PARP inhibitors show thatEGFRandBRCA-1may be found in the same protein complex and mutations in these proteins facilitate synthetic lethality of PARP inhibition. This concept has been extrapolated to NSCLC with similarEGFRmutations.35A currently accruing phase Ib/II trial is evaluating the combination of olaparib with gefitinib in patients with advanced NSCLC that have an activatingEGFRmutation.36In addition, a phase II study is currently recruiting patients with chemosensitive advanced NSCLC, to compare olaparib versus placebo as maintenance therapy in advanced NSCLC.37
Veliparib (ABT-888; Abbvie, Abbott Park, IL), an oral PARP inhibitor, is unique in that it has the ability to cross the blood-brain barrier. In preclinical studies, it potentiates platinum agents, temozolomide, and radiation therapy.38Veliparib was first studied in rats, at which time it was noted to have excellent bioavailability following oral absorption, with wide distribution in tissues. Its primary clearance is renal, followed by metabolic clearance. It is excreted unchanged in the urine, suggesting that the active compound is not a metabolite of the drug.39Kummar et al conducted a phase O trial with veliparib, where 13 patients with advanced malignancies received the drug. Good bioavailability was demonstrated, and the investigators concluded that the drug is well tolerated.40In addition, significant reduction in PARP levels were noted in the tumor tissue in response to veliparib administration.
Veliparib is being studied in a number of ongoing clinical trials for variety of cancers. In stage 3 surgically unresectable NSCLC, it is being studied in combination with thoracic radiotherapy, carboplatin, and paclitaxel in the combined modality setting by the Southwest Oncology Group (SWOG) (Figure 3).41
Appleman et al reported on a phase I trial studying the tolerability of carboplatin, paclitaxel, and veliparib in 68 patients with advanced solid tumors, 15 of them with NSCLC. Dose-limiting toxicities (febrile neutropenia and hyponatremia) were noted in 2 of the 7 patients that received the maximum dose of 120 mg. Veliparib was well tolerated at the 80-mg dose, with a dose-limiting toxicity (febrile neutropenia) in 1 of 9 patients. Partial response was observed in 11 patients (2 with lung cancer), CR in 2 patients (1 with breast cancer and 1 with urothelial cancer), and stable disease in 35 patients. It was concluded that the addition of veliparib to carboplatin and paclitaxel was generally well tolerated and has a similar safety profile to carboplatin/paclitaxel alone.42
A phase I study of veliparib in combination with cisplatin and etoposide in patients with extensive-stage SCLC (N = 9) was presented at the 2014 American Society of Clinical Oncology annual meeting. The study showed safety for the combination and efficacy in previously untreated SCLC; of the 7 evaluable patients, stable disease was noted in 2 patients, PR in 4 patients, and CR in 1 patient. A randomized phase II efficacy study is currently under way (Figure 4).43In addition, a randomized phase II study is currently recruiting patients with refractory or relapsed SCLC to evaluate the efficacy of temozolomide with or without veliparib.44
Recently, the results of a randomized, placebocontrolled phase II study in NSCLC with veliparib in combination with carboplatin and paclitaxel were reported. Patients with advanced or metastatic NSCLC are randomized to receive carboplatin/paclitaxel plus veliparib or placebo (Figure 2). Treatment was continued for a maximum of 6 cycles. Maintenance therapy was not included as part of the study treatment regimen. Patients were stratified by smoking history and histology.
Of the 158 patients enrolled, 49% had squamous NSCLC. For the intent-to-treat population, although the median PFS was numerically higher with veliparib, the differences did not reach statistical significance (5.8 months vs 4.2 months, HR 0.74;P=.193). Similar trends were also observed for overall survival (OS) favoring veliparib (11.7 months vs 9.1 months, HR 0.77;P=.205). The improvement in PFS and OS was more pronounced in patients with squamous cell histology (6.1 months vs 4.1 months [HR 0.77] and 10.3 months vs 8.4 months [HR 0.71], respectively). However, in patients with non-squamous histology, the addition of veliparib did not result in improvements in PFS or OS.
The objective response rate was approximately 30% in both study arms. However, the median duration of response was significantly longer with veliparib, with a HR of 0.11. This suggests that the addition of veliparib is likely to be effective in patients with platinum-sensitive tumors, where the clinical benefit is more pronounced, but not in platinumrefractory tumors.
Adverse events including alopecia, anemia, neutropenia, and peripheral neuropathy were reported, but were noted to be relatively similar whether or not the patient received veliparib. These observations have now led to a phase III study of carboplatin and paclitaxel with veliparib or placebo for first-line therapy of advanced SCLC.45A separate study has recently been initiated to evaluate the efficacy of the veliparib-carboplatin-paclitaxel combination in advanced NSCLC patients with a smoking history.
Rucaparib (AG-014699/PF-01367338) was the first PARP inhibitor to be studied in combination with chemotherapy. In patients with melanoma treated with intravenous rucaparib, an objective response was seen in 18% of 40 patients. However, these patients suffered from significant myelosuppression.46Phase I trials administered rucaparib to patients with advanced breast and ovarian cancers. It was tolerated well, and the most common toxicities were nausea and fatigue.47Although there are currently no trials in lung cancer, phase II trials in gynecological cancers are ongoing.
Iniparib was initially thought to be a PARP-1 inhibitor and was studied extensively in patients with triple-negative breast cancer. Despite promising results in a randomized phase II study, however, iniparib failed to improve survival when given in combination with chemotherapy in a phase III study for patients with advanced triple-negative breast cancer. A phase III trial evaluating gemcitabine and carboplatin with or without iniparib was completed in patients with previously untreated stage 4 squamous NSCLC (ECLIPSE study). The study failed to demonstrate improvement in OS with the addition of iniparib to chemotherapy.
1. On days of chemotherapy, the morning veliparib/placebo is to be administered after premedications for etoposide, prior to etoposide IV.
2. Recommended phase II dose. Dose for phase II was determined in the phase I portion of the study. ULN = upper limit of normal.
While these results are disappointing, recent reports indicate that, although iniparib does not inhibit PARP, it might exert anticancer activity through other mechanisms unrelated to PARP inhibition.48,49Further clinical development of this drug has been discontinued, however, based on the negative results of the phase III study mentioned earlier. While there are many lessons to be learned from the iniparib development history, the implications of this to other PARP inhibitors is likely to be minimal based on the new knowledge about its mechanism of action.The PARP protein plays a crucial role in DNA repair. Clinical trials are currently under way to exploit this role, and promising results have been reported in a randomized phase II study in advanced NSCLC. This has prompted phase III studies in patients with advanced NSCLC. As continued research into HR pathways and mutations within NSCLC emerge, new uses for PARP inhibition can be applied. These therapies have proven to be well tolerated upon oral administration, making a compelling rationale for the continued study of these agents in lung cancer. Preclinical and early clinical studies indicate that PARP inhibitors enhance the efficacy of platinum agents in tumors that have intrinsic platinum sensitivity, but not in platinum-refractory tumors. Therefore, identification of biomarkers that correlate with platinum sensitivity might be critical for optimal utilization of PARP inhibitors.
Presently, the biomarker discovery efforts have focused on baseline expression of DNA repair genes, ATM deficiency, homologous recombination repair capacity, and BRCA expression. Since BRCA mutations are not common in lung cancer, the level of expression of BRCA might be particularly relevant. Cells with a BRCA-mutated phenotype, based on low BRCA expression, might be sensitive to PARP inhibition. Specimens obtained from patients who participated in recent lung cancer trials with PARP inhibitors will prove to be valuable to evaluate these biomarkers and correlate them with efficacy.