Evaluating the Sensitivity of Sporadic Pancreatic Cancer to poly(ADP-ribose) polymerase (PARP) Inhibition (Velaparib, Olaparib, AG14361) as Single Agents and as Chemo-sensitizers

After resection of pancreatic cancer local recurrence occurs in 50%-80% of the cases while metastasis develops 75% of the time. Current, adjuvant therapy often consists of gemcitabine, cisplatin and/or 5-fluorouracil which add a modest increase in median survival by 4-5 months. In this study, we treated human pancreatic cancer cells with poly (ADP-ribose) polymerase (PARP) Inhibitors (AG14361, Veliparib and Olaparib) alone or with gemcitabine, cisplatin or 5 – fluorouracil. Methods: CFPAC-1 and BXPC-3, HPAC human pancreatic cancer cell lines were treated for 72 hours with PARP inhibitors alone or in combination with gemcitabine, cisplatin, or 5 – fluorouracil. Validated MTT assays were used to form dose response curves from which the IC50 values were calculated. Results: The PARP1 IC50 values for CFPAC-1, BXPC-3 and HPAC pancreatic cancer cell lines were AG14361 (14.3 μM, 12.7 μM, 38.3 μM), Veliparib (52.6 μM, 100.9 μM 102.0 μM) and Olaparib (79.5 μM, 184.8 μM, 200.2 μM). The IC50 values of cisplatin, were decreased up to 60 fold in the presence of clinically relevant amounts of PARP inhibitor while 5-flourouracil IC50 values were decreased up to 6000 fold in the presence of clinically relevant amounts of PARP inhibitor. Gemcitabine was inhibited up to 73% by PARP inhibitors. Conclusions: Sporadic human pancreatic cancer cells are sensitive to PARP inhibition. PARP inhibitors significantly enhanced the cytotoxicity of cisplatin and 5-fluorouracil while inhibiting gemcitabine. There is little correlation between endogenous PARP activity and the effectiveness of PARP inhibitors.


Introduction
Recently, Olaparib has been shown to increase progression free survival in BRCA-mutated metastatic pancreatic cancer. 1 Adenocarcinoma of the pancreas remains a deadly disease with a dismal outcome in the overwhelming majority of cases with a 5 -year survival of only 5%. 2 The primary therapeutic intervention in treating pancreatic cancer has been surgery with adjuvant and chemotherapy adding perhaps 2-5 months of additional overall survival when compared to surgery alone 3,4 . 5-Fluorouracil (5-FU) has had modest activity against pancreatic cancer in the adjuvant setting and in the treatment of metastatic and until recently was the mainstay of chemotherapy especially when used in combination with radiotherapy. 5.6 Gemcitabine was shown to be able to improve median survival when used as an adjuvant over 5-FU by one month and 1-year survival by 16% over 5-FU and since, has been the mainstay of single agent adjuvant therapy. 7 The addition of cisplatin to gemcitabine improves the median survival over gemcitabine alone by a month and a half. 8 The efficacy of chemotherapy in pancreatic cancer clearly lags behind other diseases such as breast and colorectal cancer and current treatment protocols provide only transient relief.
Poly(ADP-ribose) polymerase (PARP) inhibitors have shown efficacy as single agents, and chemosensitizers in breast and ovarian cancer. 9 In addition, preclinical evidence indicates that PARP inhibitors may serve as potent radiosensitizers. 10 The PARP enzyme binds to single strand breaks (SSB) formed during base excision repair (BER) and forms complexes with pol β, XRCC1 and DNA ligase III during the repair process. 11 The blockage of the PARP enzyme results in an eventual double strand break during DNA replication when a DNA SSB is encountered at the replication fork. 12 These double stranded breaks may then be repaired efficiently by the process of homologous recombination (HR). 13 In the absence of homologous recombination the cell may be repaired by the error prone nonhomologous end joining which may lead to the accumulation of chromatid breaks leading to the loss of viability. 14 BRCA1 and BRCA2 hereditary cancers have been shown to be highly sensitive to PARP inhibitors due to their defect in homologous recombination. 15 17.18 It may be that many sporadic cancers are deficient in homologous recombination. Adding to the more generalizable use of PARP inhibitors to sporadic cancer is the finding that some keto-phenanthradine type PARP inhibitors may also cause G2/M arrest in cancer cells. 19 It is known that cancer cells that are deficient in HR are sensitive to DNA damaging agent alone and hypersensitive to these agents after the cancer cells have been chemosensitized with PARP inhibitors. 20 In this study we look at the use of AG14361, Veliparib and Olaparib alone or in combination with 5-FU, gemcitabine and cisplatin three of the most common agents in the treatment of pancreatic cancer. 21 PARP inhibitors with platinum-based compounds have been approved for the treatment of BRCA deficient ovarian and breast cancers which are deficient in homologous recombination provided much of the impetus for this study. 22 Here we expand the potential use of PARP inhibitors from BRCA mutated pancreatic cancers to sporadic pancreatic cancer.

Cell Lines & Medications
BXPC-3, CFPAC, HPAC human pancreatic cancer cell lines were obtained from American Type Tissue Culture in Manassas VA. 5-Fluorouracil, Cisplatin, and Gemcitabine were purchased from Sigma-Aldrich, St. Louis MO. The PARP-1 inhibitor AG14361 was obtained from Organix inc. in Woburn MA while ABT888 (Veliparib) and AZD2281 (Olaparib) were obtained from Selleck Chemical in Houston TX. Cells were grown in DMEM media with 10% fetal bovine serum in a 5% CO2 incubator at 37C. Cells at 80-85% confluence were trypsinized, washed with PBS and plated for each experiment.
Determining IC50 values

1.
Standard Curve: standard curves where made by plating 25k to 300k cells for each cell line for 4-12 hours -until the cell were attached to the wells. They were then exposed to a solution of 1 mg/ml thiazolyl blue tetrazolium (MTT) for 30 minutes. This was followed by decanting the MTT and adding propranolol to each well for 30 minutes. The absorption for each well was read with the Perkin-Elmer 1420 multi-label counter. Each data point was replicated at least 10 times.

2.
IC50 value: IC50 values were determined by plating 50 k cells into 12 well plates until they were attached. At this point the cells were incubated for 72 hours to various doses of drug.
There were at least two controls containing vehicle for each experiment. The number of cells in each well at each specific drug concentration was then determined by the previously described MTT assay. Each experiment was replicated at least 9 times with the tenth time being a direct cell count by hemacytometer. The IC 50 values were then calculated by graphing the % inhibition vs log(dose) using Sigma Plot and 3 rd order equations with the required correlation coefficient r 2 value having to be > 0.950.

Determination of PARP activity
The "PARP in vivo Pharmacodynamic Assay II kit" from Trevigen inc, Gaithersburg MD was used to determine the amount of PARP activity in each cell line prior to and after treatment with PARP inhibitors. Each assay was replicated 5 times. Endogenous Poly (ADPribose) in each cell line versus the IC50 if each PARP inhibitor was evaluated using the Pearson Product Moment Correlation analysis.

Results
The ability of the PARP inhibitors AG14361, Veliparib and Olaparib to inhibit the growth of pancreatic cancer cells was evaluated utilizing the BxPC-3, CFPAC-1 and HPAC cell lines.
BxPC-3 cell line was from an adenocarcinoma of the pancreas from a 61 year old female. 23 The  1).

Discussion
Clinical trials which are the gold standard in developing new therapies for pancreatic cancer often advanced without sufficient pre-clinical evidence which might be valuable in predicting the outcome of clinical trials. Some studies have for instance combined PARP inhibition with Gemcitabine even though these are clearly antagonistic in this study. 30 The IC50 values of the three PARP inhibitors in this study were low yet the Ki for each of these three PARP inhibitors  Experimentally, the rate of repair by the NER pathway was shown to be dependent on the PARP enzyme. 32 We next looked at combining PARP inhibitors with the current mainstay of adjuvant pancreatic cancer treatment gemcitabine. There is evidence from breast cancer preclinical and clinical trials that PARP inhibitors enhance the efficacy of combined gemcitabine/platinum regimens. 33,34 Gemcitabine is metabolized to difluorodeoxycytidine di and tri phosphate where it inhibits ribonucleotide reductase and incorporates into DNA strands. In this study PARP inhibitors antagonize gemcitabine as a single agent in pancreatic cancer cells significantly by cutting its potency in half. The mechanism of this inhibition is unknown. pancreatic cancer which up to now has been relatively disappointing. It is also significant to note that PARP inhibitors are radiosensitizers this is significant as radiation therapy is currently an important part of current treatment for pancreatic cancer. 35 One question that has been asked is, "is there a molecular marker which one can use to