Int. J. Cancer: 125, (2009)

Transcription

1 Int. J. Cancer: 125, (2009) ' 2009 UICC Low drug resistance to both platinum and taxane chemotherapy on an in vitro drug resistance assay predicts improved survival in patients with advanced epithelial ovarian, fallopian and peritoneal cancer Koji Matsuo 1,2 *, Virginia K. Bond 1, Michele L. Eno 1, Dwight D. Im 1,2 and Neil B. Rosenshein 1,2 1 Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Maryland School of Medicine, Baltimore, MD 2 The Gynecologic Oncology Center, Mercy Medical Center, Baltimore, MD The objective of this study was to evaluate the role of an in vitro drug resistance assay to platinum and taxane in the management of advanced epithelial ovarian, fallopian and primary peritoneal cancer. All patients with FIGO Stage IIIc and IV who received postoperative chemotherapy with platinum and taxane for more than 4 courses after the initial cytoreductive surgery between 1995 and 2008 were evaluated. Patients who received neoadjuvant chemotherapy were not included. An in vitro drug resistance assay (EDR Assay, Oncotech, Tustin, CA) was used to determine drug resistance for each patient s tumor tissue. Level of drug resistance was described as extreme (EDR), intermediate (IDR), or low (LDR). Response to chemotherapy and survival were correlated to the EDR Assay. Of the 335 patients who underwent primary cytoreductive surgery, 173 cases met the criteria for statistical evaluation. The 58 patients (33.5%) whose tumors had LDR to both platinum and taxane had statistically improved progressionfree survival and overall survival (OS) compared with the 115 patients (66.5%) who demonstrated IDR or EDR to platinum and/ or taxane (5-year OS rates, 41.1% vs. 30.9%, p ). The 5-year OS rates for the 28 (16.2%) cases that had optimal cytoreduction with LDR to both platinum and taxane was significantly improved over the 62 (35.8%) cases that were suboptimally cytoreduced with IDR or EDR to platinum and/or taxane (54.1% vs. 20.4%, respectively, p < 0.001). In conclusion, LDR to both platinum and taxane chemotherapy, as determined by an in vitro drug resistance assay, independently predicts improved survival in patients with advanced epithelial ovarian, fallopian and peritoneal cancer, especially in those patients who undergo optimal primary cytoreduction. ' 2009 UICC Key words: ovarian cancer; in vitro drug resistance assay; platinum; taxane; chemotherapy There are approximately 25,000 women annually diagnosed with primary ovarian, fallopian and primary peritoneal cancer in the United States. 1 More than 75% of these cases present in an advanced stage. This is illustrated by the poor prognosis of this condition with a 5-year overall survival (OS) rate of Stage IIIc 28.9% and Stage IV 13.4%. 2 Platinum-based chemotherapy has improved the duration of survival significantly. The standard management of advanced disease is maximal cytoreductive surgery followed by systemic combination chemotherapy with platinum and taxane. 3 The response rate to primary chemotherapy in advanced disease is up to 80%. The majority of patients enter remission, unfortunately, 75% of these patients relapse. 4 Given such a high recurrence rate, prediction of response to combination chemotherapy with platinum and taxane has been a strong area of research interest for investigators. As an alternative to empiric chemotherapy, in vitro chemosensitivity and drug resistance assays are designed to offer a choice of chemotherapeutic agents as determined by the tumor tissue response to drugs under laboratory conditions. 5 Chemosensitivity offers relatively low positive predictive accuracy and no survival advantage. 6,7 Unlike sensitivity tests, drug resistance assays, with the percent cell inhibition more than 1 SD below the median termed as extreme drug resistance (EDR), show greater than 99% accuracy in identifying chemotherapeutic agents with no clinical response. 8,9 There is only 1 study that concluded a significantly increased risk of progression and death in subjects demonstrating EDR to platinum in primary ovarian cancer. 9 Otherwise, most previous studies of drug resistance assays did not show a significant improvement in OS in patients with primary ovarian cancer treated with assay-directed chemotherapy. 8,10 15 Although platinum and taxane-based chemotherapy is the standard management of primary ovarian cancer, there are limited studies available that investigate the role of an in vitro drug resistance assay in the prediction of response to the primary platinum and taxane combination chemotherapy and survival for ovarian cancer. 6 Eltabbakh et al 6 did not demonstrate that the presence of EDR to paclitaxel precludes the response to combination chemotherapy with paclitaxel and cisplatin in ovarian cancer. The aim of our study was to evaluate the role of an in vitro drug resistance assay to platinum and taxane and its impact on the survival outcome in advanced epithelial ovarian, fallopian and primary peritoneal cancer. Material and methods A retrospective study was conducted using a database of in vitro drug resistance assay results for specimens obtained at Mercy Medical Center in Baltimore for epithelial ovarian, fallopian and primary peritoneal cancer between January 1, 1995 and March 31, In our facility, tissue samples from initial cytoreductive surgeries for ovarian, fallopian and primary peritoneal cancer are routinely evaluated for drug resistance, using an in vitro assay. These results are only considered if there is recurrence. Patient demographics, clinicopathologic data, response to chemotherapy and follow-up outcomes were abstracted from medical records and pathology reports. Survival outcome was correlated with in vitro drug resistance assay results. The standard protocol for our facility in the management of ovarian, fallopian and primary peritoneal cancer is initiation of first-line chemotherapy 3 4 weeks after the primary cytoreductive surgery. Inclusion criteria were FIGO Stage IIIc and IV who received postoperative chemotherapy with platinum (carboplatin or cisplatin) and taxane (paclitaxel or docetaxel) for more than 4 courses after the initial cytoreductive surgery. Exclusion criteria were neoadjuvant chemotherapy, histopathology with mucinous carcinoma and low malignancy potential and coexisting cancer or recent history of cancer within 5 years. Mucinous carcinoma of the ovary is often associated with tumor, with origin from the gastrointestinal system and was, for this reason, excluded from the statistical evaluation. In vitro drug resistance assay (EDR Assay 1, Oncotech, Tustin, CA) results were evaluated for platinum and taxane, as well as cyclophosphamide, doxil, etoposide, gemcitabine and topotecan. 16 *Correspondence to: Obstetrics, Gynecology, and Reproductive Sciences, University of Maryland Medical Center, University of Maryland School of Medicine, 22 South Greene Street, Box 290, Baltimore, MD 21201, USA. Fax: koji.matsuo@gmail.com Received 20 January 2009; Accepted after revision 4 June 2009 DOI /ijc Published online 15 June 2009 in Wiley InterScience ( wiley.com). Publication of the International Union Against Cancer

2 2722 MATSUO ET AL. FIGURE 1 Inclusion and exclusion criteria. Relative in vitro tumor cell proliferation observed in the presence of each tested drug was determined compared with the control medium and designated low drug resistance (LDR), less than median growth; intermediate drug resistance (IDR), median to 11.0 SD; and extreme drug resistance (EDR), 11.0 SD or greater. Tissue specimens were collected during primary cytoreductive surgery for frozen histologic evaluation and for fixation to the EDR Assay medium by pathologists. The standard methodology of the EDR Assay is as follows. 16 Fresh viable tumor tissue is minced and undergoes an enzymatic reaction to disaggregate the tumor cells. The tumor cells are plated in soft agar, which preferentially favors tumor cell proliferation. Cells are exposed to tumor type-specific antineoplastic agents for 5 days in a carefully controlled environment. Drug exposures in excess of the maximum tolerated are used. Because of the reduced rate of drug metabolism, in vitro tumor exposure to chemotherapy agents is 5 80 times greater than in vivo. Tritiated thymidine is introduced during the last 2 days of culture as a measure of cell proliferation. Treated cells are compared with untreated controls. In vitro drug resistance assay results for platinum and taxane were evaluated for: (i) crude assay results; and (ii) results for actual given chemotherapy agents. In the crude assay results, discrepancy results from LDR to either IDR or EDR and from IDR to EDR were evaluated between carboplatin and cisplatin as well as paclitaxel and docetaxel. Survival outcome for drug assay results with actual given chemotherapy drugs was determined. Expression of the following biomarkers was tested: multidrug resistance gene product-1, mutation of p53, human epidermal growth factor receptor-2, epidermal growth factor receptor, CD31 for angiogenesis index (per 3200 field), estrogen receptor, and progesterone receptor. Percentage of positive expression of biomarker (%) and histoscore were evaluated. DNA profiles with presence of aneuploidity, DNA index and S-phase fraction (%) were also evaluated. Optimal cytoreductive surgery was defined as no residual tumor measuring greater than 1 cm in maximal dimension at the end of surgical procedure. Date of progression was determined by clinical examination, imaging study either with CT or with PET scan, and/or CA-125 level. Platinum resistance was defined as the presence of recurrence within 6 months from the date of last chemotherapy. Response to chemotherapy was defined by the new Response Evaluation Criteria in Solid Tumours guidelines. 17 Progression-free survival (PFS) was defined as the time interval from the date of initial cytoreductive surgery to the date of documented first recurrence or progression of disease. If there was no recurrence, PFS was determined as the date of last follow-up. OS was defined as the interval between the initial surgery and the date of death or last follow-up visit. Continuous variables were assessed for the normal distribution by Kolmogorov-Smirnov test and expressed either by mean (6SD) or by median (range) as appropriate. Categorical variables were evaluated by using Fisher s exact test with odds ratio and 95% confidence interval (95% CI). Univariate analysis with liner regression test was used to assess all the corrected variables and primary outcome (PFS and OS). For the significant variables for univariate analysis, multivariate analysis with Cox log rank test was further performed to determine the difference in survival. Kaplan Meier test was used to estimate survival curves. All statistical tests were 2 tailed, and p values of less than 0.05 were considered statistically significant. The statistical significance of the data was determined using Statistical Package for Social Scientists software (SPSS, version 12.0, Chicago, IL). The study protocol was approved by the institutional review board at Mercy Medical Center, Baltimore, Maryland. Results A total of 335 cases were identified during the study period. One hundred ninety-two (57.3%) of these cases received more than 4 courses of combination chemotherapy consisting of platinum and taxane for FIGO Stage IIIc and IV epithelial ovarian, fallopian and primary peritoneal cancer (Fig. 1). Of those 192 cases,

3 DRUG RESISTANCE ASSAY AND CANCER CHEMOTHERAPY 2723 there were 8 (4.2%) cases with failed assays and 11 (5.7%) cases that assay results were not available for the actual platinum or taxane given. The remaining 173 (51.6%) cases were evaluated for significance. The patient demographics were shown in Table I. The majority of cases were epithelial ovarian cancer (87.9%) with FIGO Stage IIIc (86.1%), serous carcinoma (82.7%), and these patients received 6 courses of carboplatin and paclitaxel (52.3%). Histology subtypes did not show a significant difference in in vitro drug resistance assay results (serous vs. nonserous, platinum EDR, 7.7% vs. 10.0%, p ; Taxane EDR, 17.5% vs. 13.3%, p ). Tumor grade was not correlated with the in vitro drug resistance assay results (high grade vs. nonhigh grade, platinum EDR, TABLE I PATIENT DEMOGRAPHICS Subjects n Age FHx of breast, ovarian cancer 16 (9.2%) PHx of breast cancer 1 9 (5.2%) Type of cancer Ovarian cancer 152 (87.9%) Primary peritoneal cancer 19 (11.0%) Fallopian cancer 2 (1.2%) Lymph nodes metastasis 69 (39.9%) Lympho-vascular invasion 57 (32.9%) FIGO stage IIIc 149 (86.1%) IV 24 (13.9%) Histologic type Serous 143 (82.7%) Endometrioid 5 (2.9%) Undifferentiated 3 (1.7%) Clear cell 2 (1.2%) Mixed 18 (10.4%) Other 2 (1.2%) Tumor size 8.5 cm (1.2 25) High- grade tumor 134 (79.3%) Preoperative CA U/mL (8 26,000) Optimal surgery 81 (46.8%) Bowel resection 60 (34.7%) Type of chemotherapy Carboplatin 1 Paclitaxel 90 (52.3%) Carboplatin 1 Docetaxel 53 (30.6%) Cisplatin 1 Paclitaxel 30 (17.3%) Number of cycles 6 (4 10) Platinum resistance 55 (31.8%) Site of recurrence n (72.8%) Extrapelvis 72 (47.4%) Pelvis 21 (16.7%) Chest 15 (8.7%) Other 8 (4.6%) 1 Not include past history within 5 years. 7.5% vs. 11.1%, p ; taxane EDR, 13.5% vs. 27.8%, p ). Optimal surgery was documented in 46.8% of cases. Response to chemotherapy was seen in 83.8%. With a median period of 12.6 ( ) months, progression of disease was seen in 72.8% of cases. Median follow-up period was 28.8 ( ) months and 78 (45.1%) patients died. The 2, 3, and 5-year OS rates were 76.4, 63.7, and 33.4% in the study population. Crude results of the in vitro drug resistance assay for platinum and taxane are shown in Table II. The proportion of EDR for platinum was less than 10%. There was no difference in the proportion of assay results in platinum agents (p ) and taxane agents (p ). Among 134 assays for carboplatin and cisplatin, the different results counted for 29.9%: LDR to IDR 17.9%, LDR to EDR 3.7%, and IDR to EDR 8.2%. A greater difference (43.6%) was seen in the 101 assays for paclitaxel and docetaxel: LDR to IDR 23.8%, LDR to EDR 6.9%, and IDR to EDR 12.9% (p ). The actual chemotherapy agents given are shown in the second portion of Table II. LDR to platinum and taxane was found in 65.3% and 50.9%, respectively. LDR to both platinum and taxane did count theoretically for 79 (45.7%) cases in the crude assay results, whereas there were 58 (33.5%) cases of LDR to both platinum and taxane given as the actual chemotherapy regimen (p ). Platinum resistance was seen in 55 (31.8%) patients and was not associated with EDR for platinum in the in vitro drug resistance assay (platinum EDR: carboplatin 12.8%, cisplatin 12.5%, p , p , respectively) (Table II). Presence of EDR to platinum did not predict platinum resistance. Response rate to chemotherapy among the cases with EDR to platinum was not significantly different from the cases with non- EDR (71.4% vs. 86.2%, p ). Presence of EDR to taxane also did not show statistical significance for response compared with non-edr to taxane (86.2% vs. 84.7%, p 5 1.0). Response to chemotherapy demonstrating EDR to at least 1 of these 2 agents (n 5 34) was similar to non-edr cases (82.1% vs. 85.8%, p ). Univariate analysis was performed to identify the significant factors affecting PFS and survival time (Tables III and IV). Among collected demographics, bowel resection (negatively), response to chemotherapy and optimal surgery were the significant variables correlated to both PFS and OS. EDR to platinum and/or taxane was not associated with survival outcome. LDR to platinum was not associated with survival status. LDR to taxane was associated with PFS (R , p ) and OS (R , p ). For combination effects, LDR to both platinum and taxane was significantly associated with survival outcome (PFS R , p ; OS R , p ). Among the tested biomarkers, only the expression of PR was correlated to both PFS and OS. Multivariate analysis was performed to identify significant variables associated with survival outcome (Table V). Among the TABLE II CHARACTERISTICS OF CRUDE RESULTS OF IN VITRO DRUG RESISTANCE ASSAY Carboplatin Cisplatin Platinum Paclitaxel Docetaxel Taxane Crude assay results EDR 16 (9.3%) 10 (7.4%) 36 (20.8%) 15 (14.9%) IDR 46 (26.6%) 33 (24.4%) 53 (30.6%) 39 (38.6%) LDR 110 (63.6%) 92 (68.1%) 84 (48.6%) 47 (46.5%) Total 172 (99.4%) 135 (78.5%) 173 (100%) 101 (58.4%) Assay results for actual given agents EDR 12 (8.4%) 2 (6.7%) 14 (8.1%) 23 (19.2%) 6 (11.3%) 29 (16.8%) IDR 39 (27.3%) 7 (23.3%) 46 (26.6%) 34 (28.3%) 22 (41.5%) 56 (32.4%) LDR 92 (64.3%) 21 (70.0%) 113 (65.3%) 63 (52.5%) 25 (47.2%) 88 (50.9%) Total 143 (82.7%) 30 (17.4%) 173 (100%) 120 (69.4%) 53 (30.6%) 173 (100%) Platinum resistance EDR 6 (12.8%) 1 (12.5%) 7 (12.7%) IDR 14 (29.8%) 3 (37.5%) 17 (30.9%) LDR 27 (57.4%) 4 (50.0%) 31 (56.4%) Total 47 (32.8%) 8 (26.7%) 55 (31.8%) LDR, low drug resistance; IDR, intermediate drug resistance; EDR, extreme drug resistance.

4 2724 MATSUO ET AL. TABLE III UNIVARIATE ANALYSIS FOR THE CLINICAL VARIABLES AND SURVIVAL OUTCOME Progression-free survival p Overall survival p Age Family history of breast cancer Personal history of breast cancer Preoperative CA Tumor size Stage IV Type of cancer Serous carcinoma High-grade tumor Lymph nodes metastasis Lympho-venous invasion Bowel resection Optimal surgery Number of chemotherapy cycles Carboplatin 1 Paclitaxel Response to chemotherapy < In vitro drug resistance assay EDR to platinum EDR to taxane EDR to platinum or taxane EDR to platinum and taxane LDR to platinum LDR to taxane LDR to platinum and taxane Multiple LDR Multiple EDR Linear regression test (R 2 is shown). LDR, low drug resistance; EDR, extreme drug resistance. TABLE IV UNIVARIATE ANALYSIS FOR BIOMARKER EXPRESSIONS AND SURVIVAL OUTCOME Progression-free survival p Overall survival p MDR-1 % cells MDR-1 staining intensity MDR-1 Result DNA aneuploidity DNA profile DNA index DNA profile S-phase fraction S-phase fraction >11% HER2 % cells HER2 staining intensity HER2 result ER % cells ER staining intensity ER histoscore ER result PR % cells PR staining intensity PR histoscore PR result p53 gene product % Cells p53 gene product staining intensity p53 gene product result EGFR % cells EGFR staining intensity EGFR result CD Linear regression analysis (R 2 is shown). MDR-1, multidrug resistance gene-1; HER2, human epidermal growth factor receptor 2; ER, estrogen receptor; PR, progesterone receptor; EGFR, epidermal growth factor receptor. tested variables, response to chemotherapy was the strongest variable for PFS (median time 17.5 vs. 5.6 months, hazard ratio 28.1, 95% CI , p < 0.001), followed by LDR to both platinum and taxane (median time 19.2 vs months, hazard ratio 3.48, 95% CI , p ) (Fig. 2a). For OS, LDR to both platinum and taxane was the strongest variable (5-year OS rates 41.1% vs. 30.9%, odds ratio 3.32, 95% CI , p ) (Fig. 2b), followed by optimal surgery (48.8% vs. 23.2%, hazard ratio 2.67, 95% CI , p ). Combined effects of cytoreductive surgery with in vitro drug resistance assay were investigated. The proportion of subgroups was as follows: optimal surgery with LDR to both platinum and taxane (n 5 28, 16.2%; Group A); optimal surgery with non-ldr to platinum and/or taxane (n 5 53, 30.6%; Group B); suboptimal surgery with LDR to both platinum and taxane (n 5 30, 17.3%; Group C), and suboptimal surgery with non-ldr to platinum and/or taxane (n 5 62, 35.8%; Group D). Survival outcomes among the 4 groups showed significant statistical difference. Median PFS times in Groups A, B C, and D were 21.2, 19.7, 17.9 vs months, respectively (p ) (Fig. 3a). Five-year OS rates in Groups A, B, C and D were 54.1, 43.1, 28.9 and 20.4%, respectively (p < 0.001) (Fig. 3b).

5 DRUG RESISTANCE ASSAY AND CANCER CHEMOTHERAPY 2725 TABLE V MULTIVARIATE ANALYSIS FOR SIGNIFICANT VARIABLES Progression-free survival Median time Hazard ratio 95% CI p Age Preoperative CA Bowel surgery 13.7 vs Response to chemotherapy 17.5 vs <0.001 Optimal surgery 19.2 vs LDR to platinum and taxane 19.2 vs PR receptor histoscore Overall survival 5-year OS Hazard ratio 95% CI p High-grade tumor 26.1% vs. 37.3% Bowel resection 33.3% vs. 37.2% Response to chemotherapy 36.5% vs. 30.0% Optimal surgery 48.8% vs. 23.2% LDR to platinum and taxane 41.1% vs. 30.9% EGFR % Cells HER2 staining intensity PR histoscore Cox log rank analysis. Median time (months). OS, overall survival; LDR, low drug resistance; HER2, human epidermal growth factor receptor 2; PR, progesterone receptor; EGFR, epidermal growth factor receptor. Discussion The survival benefit of LDR to both platinum and taxane was first demonstrated in the analysis of Stage IIIc and IV, epithelial ovarian, fallopian and primary peritoneal cancer patients who underwent initial cytoreductive surgery followed by more than 4 courses of combination chemotherapy with platinum and taxane. This suggests that there may be a certain subgroup of patients in whom the use of an in vitro drug resistance assay is beneficial. The drug resistance assay was originally designed to identify the drug agents that are unlikely to be effective. 8 Positive correlation between assay-oriented chemotherapy and its response has been reported in ovarian cancer. 8,10 15 However, survival benefit of assay-oriented chemotherapy is not yet proven. 8,10 15 Similar results were observed in our study, and presence of EDR to platinum and/or taxane was not correlated to survival outcome. Based on theory for which the assay was designed, 8 we initially analyzed the chemotherapy response and survival outcome by the conventional approach by comparing EDR to non-edr (IDR1LDR). However, as shown in Table III, presence of EDR in the assay did not show a survival benefit in epithelial ovarian, fallopian and peritoneal cancer. Because of this finding, we looked back at the original report by Kern et al 8 (Table VI). Although this table obviously shows a high nonresponse rate with EDR chemotherapeutic agents (99.2%), these results also give a chemotherapy response rate of 51.8% with LDR agents. This response rate is far superior to the total response rate (LDR vs. total, 51.8% vs. 29.3%, p < 0.001). Therefore, we believe the following is true of the in vitro drug resistance assay: (i) EDR agents are less likely to show antitumor effect; and (ii) LDR agents may possibly show a response to tumor. It is, therefore, reasonable to choose a chemotherapeutic agent with LDR rather than EDR in the in vitro drug resistance assay results. In fact, previous data showed an improved disease-free interval in the patients treated with LDR agents in recurrent ovarian cancer. 4 Selective choice of chemotherapeutic agents for LDR with non-edr is unique to our study and may be a new use of the in vitro drug resistance assay. Based on this evidence, we further hypothesized that LDR to both platinum and taxane improves survival outcome in epithelial ovarian, fallopian and peritoneal cancer. To test this hypothesis, we classified the assay results to LDR and non-ldr (IDR1EDR). As shown in Table VI, response rate to the IDR drug is far lower than that of total response rate among the tested patients (IDR vs. total, 15.8% vs. 29.3%, p ). Response rate for the combined group with EDR and IDR is still far lower than that of total response (EDR1IDR vs. total, 7.5% vs. 29.3%, p < 0.001). FIGURE 2 (a) Progression-free survival for low drug resistance to platinum and taxane. Black line, LDR to both platinum and taxane; dash line, non-ldr to both platinum and taxane. (b) Overall survival for low drug resistance to platinum and taxane. Black line, LDR to both platinum and taxane; dash line, non-ldr to platinum and/or taxane. Five-year survival rates, 41.0% vs. 30.9%, odds ratio 3.32, 95% CI , p LDR, low drug resistance. Finally, the chemotherapy response rate in LDR agents is significantly higher than that of non-ldr agents (LDR vs. non-ldr, 51.8% vs. 7.5%, odds ratio 14.4, 95% CI , p < 0.001).

6 2726 MATSUO ET AL. TABLE VI IN VITRO DRUG RESISTANCE ASSAY AND CHEMOTHERAPY RESPONSE 1 Nonresponse Response Total LDR 107 (49.2%) 115 (51.8%) 2, (49.3%) IDR 85 (84.2%) 16 (15.8%) (22.4%) EDR 126 (99.2%) 1 (0.8%) 127 (28.2%) Total 318 (70.7%) 132 (29.3%) 450 (100%) 1 Adapted and modified from the original article by Kern et al [8]. 2 LDR vs. total response, 51.8% vs. 29.3%, p < LDR vs. IDR1EDR, 51.8% vs. 7.5%, p < IDR vs. total response, 15.8% vs. 29.3%, p LDR, low drug resistance; EDR, extreme drug resistance. FIGURE 3 (a) Progression-free survival for combination effects of optimal surgery and in vitro drug resistance assay. Median time (months), p Black, optimal with LDR (21.2 months); Black dash, optimal with non-ldr (19.7 months); Gray, suboptimal with LDR (17.9 months); Gary dash, suboptimal with non-ldr (12.7 months). (b) Overall survival for combination effects of optimal surgery and in vitro drug resistance assay. Five-year survival rate, p < Black, optimal with LDR (54.1%); Black dash, optimal with non-ldr (43.1%); Gray, suboptimal with LDR (28.9%); Gary dash, suboptimal with non-ldr (20.4%). LDR, low drug resistance. Thus, we believe our classification to either LDR or non-ldr is quite useful. Our results uniquely demonstrate an improved survival outcome with LDR agents in epithelial ovarian, fallopian and peritoneal cancer. Although chemotherapeutic agents with LDR possibly show an improved response rate, it is still cautioned to elect to choose the chemotherapeutic agents with LDR. Although improved survival is reported in our results as well as the previous report, 4 these were retrospective study designs for potential confounding factors may not be controlled. Furthermore, host factors, such as vascularization of the tumor, expression of enzymes that detoxify the drug, and protein binding in the circulation, may interfere with response and adverse events associated with chemotherapy may still occur. All these host factors could render the drug inactive before it reaches the tumor. Therefore, a tumor expressing LDR may never even see the drug and have a chance to respond. These factors contribute to the higher positive predictive value of resistance assays as compared with sensitivity assays. Among the previous studies, the effect of chemotherapy was mainly evaluated for platinum monotherapy. However, in the clinical setting, combination chemotherapy with platinum and taxane is the standard treatment for primary ovarian cancer following cytoreductive surgery. 3 For this reason, we investigated LDR to both platinum and taxane as a predictor for clinical outcome in ovarian cancer. In vitro drug resistance assays and chemosensitivity assays are not equivalent tests. The former represents tumor cell growth inhibition, whereas the latter represents cell death via drug-induced apoptosis. The drug resistance assay is used to exclude agents that are less likely to be effective rather than to select the agent predicting the clinical response. 4 Thus, presence of LDR does not mean that the chemotherapeutic agent induces tumor cell death. There is only 1 study that has investigated the role of an in vitro drug resistance assay for primary chemotherapy with platinum and taxane for epithelial ovarian cancer. 6 The study failed to conclude the usefulness of the assay in the response to chemotherapy with platinum and taxane (EDR vs. non-edr, 81.3% vs. 86.4%, p ). The size of this study was underpowered (EDR, n 5 16 vs. non-edr, n 5 59). 6 By using the reported chemotherapy response rates, the sample size analysis estimates n 5 68 for both groups with alpha error level of 5% and beta error level of 80%. In our analysis of 173 cases (EDR, n 5 34 vs. non-edr, n 5 139), similar results were observed, and presence of EDR to platinum and/or taxane was not associated with chemotherapy response (EDR vs. no EDR, 82.1% vs. 85.8%, p ). Although our subject size is still underpowered, our results may support that the EDR assay does not preclude response to the first-line chemotherapy for epithelial ovarian cancer. 6 The strengths of our study are the homogeneity of the patient population and larger sample size compared with previous studies. The weakness of our study is that this is a retrospective study that potentially missed confounding factors. Further, correlating in vitro drug sensitivity or resistance testing to in vivo biological survival is a very difficult process. Many patients will receive more than 1 regimen of chemotherapy throughout their treatment course. Although response to first-line chemotherapy has one of the most important roles in postoperative management, this dilemma needs to be addressed. Although only 33.5% of cases in which LDR agents were administered, there were 45.7% of cases that theoretically showed LDR to both platinum and taxane (p ). These theoretical combinations included carboplatin with paclitaxel, carboplatin with docetaxel, cisplatin with paclitaxel and cisplatin and docetaxel. Although carboplatin with paclitaxel provides the same chemotherapeutic efficacy with less drug toxicity compared with cisplatin and paclitaxel, 18 there is not sufficient data at this time to support the use of combination chemotherapy with cisplatin and docetaxel. Further investigations of this option are needed. Efforts to maximize the optimal cytoreductive surgery for advanced stage of ovarian cancer have been attempted for decades. A meta-analysis of 53 studies between 1989 and 1998 reported mean percentage of optimal cytoreductive surgery as 41.9% (range 0 100%). 19 Positive correlation between optimal surgery and survival time was noted. 19 In 2008, the average of optimal surgery was reported as 56% in the observation of 490 cases. 20 Maximal cytoreduction was more evident after 2001, and 78% of cases were reported as optimal surgery. 20 In our 14 years of experience, the average of optimal surgery was 46.8% and

7 DRUG RESISTANCE ASSAY AND CANCER CHEMOTHERAPY year OS rate was 33.4% in our population. If we assume that maximal cytoreductive surgery of 78% is achieved in our patients, the theoretical calculated 5-year OS is 42.8% (74 of 173). Furthermore, if we assume that all patients underwent maximal cytoreductive surgery, the theoretical survival is 48.1%. These averages are far superior to the current survival rates of advanced ovarian cancer (Stage IIIc 28.9% and Stage IV 13.4%). 2 Surgical effort to maximize the survival chance is mandatory. In summary, the presence of LDR to both platinum and taxane in this in vitro drug resistance assay impacts the survival outcome in advanced epithelial ovarian, fallopian and primary peritoneal cancer. However, routine chemosensitivity and drug resistance assay in clinical practice is not recommended by the American Society of Clinical Oncology, and these tests are limited only in the setting of clinical trail. 21 A prospective randomized controlled trial is needed. Acknowledgements The abstract of the present work was presented at 100th Annual Meeting of American Association for Cancer Research (AACR), Denver, Colorado, April 18 22, 2009 (no.750). The authors thank William A. Ricketts, PhD, Exiqon, Inc., for technical advice in the discussion. References 1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun MJ. Cancer statistics, CA Cancer J Clin 2008;58: Heintz AP, Odicino F, Maisonneuve P, Beller U, Benedet JL, Creasman WT, Ngan HY, Pecorelli S. Carcinoma of the ovary. Int J Gynaecol Obstet 2003;83(Suppl 1): McGuire WP, Hoskins WJ, Brady MF, Kucera PR, Partridge EE, Look KY, Clarke-Pearson DL, Davidson M. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996;334: Loizzi V, Chan JK, Osann K, Cappuccini F, DiSaia PJ, Berman ML. Survival outcomes in patients with recurrent ovarian cancer who were treated with chemoresistance assay-guided chemotherapy. Am J Obstet Gynecol 2003;189: Samson DJ, Seidenfeld J, Ziegler K, Aronson N. Chemotherapy sensitivity and resistance assays: a systematic review. J Clin Oncol 2004; 22: Eltabbakh GH, Piver MS, Hempling RE, Recio FO, Lele SB, Marchetti DL, Baker TR, Blumenson LE. Correlation between extreme drug resistance assay and response to primary paclitaxel and cisplatin in patients with epithelial ovarian cancer. Gynecol Oncol 1998;70: M aenp a a JU, Heinonen E, Hinkka SM, Karnani P, Klemi PJ, Korpijaakko TA, Kuoppala TA, Laine AM, L ahde MA, Nuoranne EK, Puolakka JO, Ranta TT, Rhen KE, Salmi AM, Suonio E, Taina EJ, Tuominen JE, Tuomisto L, V ayrynen MA. The subrenal capsule assay in selecting chemotherapy for ovarian cancer: a prospective randomized trial. Gynecol Oncol 1995;57: Kern DH, Weisenthal LM. Highly specific prediction of antineoplastic drug resistance with an in vitro assay using suprapharmacologic drug exposures. J Natl Cancer Inst 1990;82: Holloway RW, Mehta RS, Finkler NJ, Li KT, McLaren CE, Parker RJ, Fruehauf JP. Association between in vitro platinum resistance in the EDR assay and clinical outcomes for ovarian cancer patients. Gynecol Oncol 2002;87: Brown E, Markman M. Tumor chemosensitivity and chemoresistance assays. Cancer 1996;77: Bellamy WT. Prediction of response to drug therapy of cancer. A review of in vitro assays. Drugs 1992;44: Von Hoff DD, Sandbach JF, Clark GM, Turner JN, Forseth BF, Piccart MJ, Colombo N, Muggia FM. Selection of cancer chemotherapy for a patient by an in vitro assay versus a clinician. J Natl Cancer Inst 1990;82: Von Hoff DD, Kronmal R, Salmon SE, Turner J, Green JB, Bonorris JS, Moorhead EL, Hynes HE, Pugh RE, Belt RJ, Alberts DS. A Southwest Oncology Group study on the use of a human tumor cloning assay for predicting response in patients with ovarian cancer. Cancer 1991;67: Cree IA, Kurbacher CM. Individualizing chemotherapy for solid tumors is there any alternative?. Anticancer Drugs 1997;8: Weisenthal LM, Kern DH. Prediction of drug resistance in cancer chemotherapy: the Kern and DiSC assays. Oncology (Williston Park) 1991;5: Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, Verweij J, Van Glabbeke M, van Oosterom AT, Christian MC, Gwyther SG. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer. National Cancer Institute of the United States. National Cancer Institute of Canada. J Natl Cancer Inst. 2000;92: Ozols RF, Bundy BN, Greer BE, Fowler JM, Clarke-Pearson D, Burger RA, Mannel RS, DeGeest K, Hartenbach EM, Baergen R; Gynecologic Oncology Group. Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 2003;21: Bristow RE, Tomacruz RS, Armstrong DK, Trimble EL, Montz FJ. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol 2002;20: Zivanovic O, Eisenhauer EL, Zhou Q, Iasonos A, Sabbatini P, Sonoda Y, Abu-Rustum NR, Barakat RR, Chi DS. The impact of bulky upper abdominal disease cephalad to the greater omentum on surgical outcome for stage IIIC epithelial ovarian, fallopian tube, and primary peritoneal cancer. Gynecol Oncol 2008;108: Schrag D, Garewal HS, Burstein HJ, Samson DJ, Von Hoff DD, Somerfield MR; ASCO Working Group on Chemotherapy Sensitivity and Resistance Assays. American Society of Clinical Oncology Technology Assessment: chemotherapy sensitivity and resistance assays. J Clin Oncol 2004;22: