- Email: [email protected]
Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for epithelial ovarian cancers likely to have suboptimal resection
YGYNO-976536; No. of pages: 8; 4C: Gynecologic Oncology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for epithelial ovarian cancers likely to have suboptimal resection Sarah E. Gill, M.D. a, Michaela E. McGree, B.S. b, Amy L. Weaver, M.S. b, William A. Cliby, M.D. a, Carrie L. Langstraat, M.D. a,⁎ a b
Division of Gynecologic Surgery, Mayo Clinic, Rochester, MN, USA Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA
H I G H L I G H T S • NACT/IDS is associated with improved survival in select patients. • Patients treated with NACT/IDS have reduced perioperative morbidity. • Younger age and normalization of CA-125 prior to IDS may improve survival.
a r t i c l e
i n f o
Article history: Received 16 September 2016 Received in revised form 11 November 2016 Accepted 12 November 2016 Available online xxxx Keywords: Neoadjuvant chemotherapy Interval debulking surgery Ovarian cancer Unresectable
a b s t r a c t Objective. To: a) identify prognostic factors in patients with epithelial ovarian cancer treated with neoadjuvant chemotherapy (NACT) and interval debulking surgery (IDS), and b) compare post-surgical survival between patients treated with NACT/IDS for presumed unresectable disease and stage IIIC/IV patients who underwent suboptimal primary debulking surgery (PDS). Methods. This was a retrospective study of consecutive stage IIIC or IV patients undergoing IDS after NACT at Mayo Clinic from January 2007 to December 2013. A subset of patients receiving NACT/IDS for the indication of unresectable disease were matched 1:1 on age and stage to a cohort of patients who underwent suboptimal PDS between 2003 and 2011. Hazard ratios and corresponding 95% confidence intervals were estimated from Cox proportional hazards models. Results. We identified 87 patients treated with NACT/IDS: the median OS and PFS following surgery was 2.4 and 1.0 years, respectively. Factors associated with significantly worse OS were older age (adjusted HR 1.60 per 10-year increase in age, 95% CI 1.18, 2.16) and elevated CA-125 before IDS (adjusted HR 2.30 for CA-125 N 35 U/mL, 95% CI 1.25, 4.23). Number of adjuvant chemotherapy cycles administered did not have a significant effect on survival. In the matched cohort analysis of presumed unresectable cases undergoing NACT/IDS vs suboptimal PDS cases (n = 45 each), the NACT/IDS group had a significant OS advantage (HR 0.53; 95% CI 0.32, 0.88), and fewer patients experienced a 30-day postoperative Accordion grade 3/4 complication (11% vs 36%, P = 0.01). Conclusions. Younger age and normalization of CA-125 prior to IDS are associated with improved survival with NACT/IDS. For primary EOC where resection to residual disease of 1 cm or less is unlikely, NACT/IDS is associated with improved survival and reduced perioperative morbidity compared to PDS. As these patients are likely best served by NACT/IDS, more reliable predictors of resectability would be valuable. © 2016 Elsevier Inc. All rights reserved.
1. Introduction
⁎ Corresponding author at: Division of Gynecologic Surgery, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. E-mail address: [email protected] (C.L. Langstraat).
Currently, the gold standard for primary management of advanced ovarian cancer is primary debulking surgery (PDS) followed by platinum-based adjuvant chemotherapy, with significantly improved overall survival in patients receiving an optimal cytoreduction [1–9].
http://dx.doi.org/10.1016/j.ygyno.2016.11.021 0090-8258/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
2
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
However, neoadjuvant chemotherapy (NACT) followed by interval debulking surgery (IDS) and adjuvant chemotherapy has been strongly advocated by some groups as a reasonable alternative, particularly for patients with reversible contraindications to primary surgery (poor nutritional status, poor performance status, recent thromboembolism, or myocardial infarction, etc.). Patients with a low probability of optimal cytoreduction may also benefit from NACT. However, there is a paucity of data available to directly compare survival and morbidity outcomes of suboptimal PDS vs NACT/IDS for unresectable disease. Although NACT has gained popularity as a primary treatment for EOC, there is disagreement in the literature regarding the significance of previously reported benefits of this treatment strategy as compared with standard therapy. Several studies, including randomized controlled trials (RCTs), report similar median progression-free survival (PFS) and overall survival (OS) in patients receiving NACT/IDS vs standard PDS followed by platinum-based adjuvant chemotherapy [10– 12]. In the European Organization for Research and Treatment of Cancer (EORTC) trial, median OS and PFS were similar in patients with stage IIIC and IV EOC in the NACT group (median OS 30 months and median PFS 12 months) and the primary surgery group (median OS 29 months and median PFS 12 months) [10]. The CHORUS trial also reported noninferior OS for women with stage III and IV EOC receiving NACT versus primary surgery (median OS 24.1 months vs 22.6 months, respectively) [12]. However, survival and surgical outcomes (e.g. residual disease) in the PDS arms were substantially lower than many single institution and phase III trials [10,13–14]. A retrospective study by Chi et al. analyzed the outcomes of patients undergoing PDS at their institution during the same time period as the EORTC trial, reporting a median PFS of 17 months and median OS of 50 months. They concluded that the median survival observed in the NACT arm of the EORTC trial was more similar to the median survival of patients undergoing suboptimal PDS at their institution [15], a conclusion also reported in other studies [16– 19]. Several meta-analyses also suggest that NACT is associated with poorer outcomes when compared to standard PDS followed by adjuvant chemotherapy [17–18]. These criticisms have led many to wonder if the results of these RCTs are relevant to their own patients. Many physicians advocate for the use of NACT/IDS arguing that it reduces perioperative morbidity [20], increases rates of optimal cytoreduction [21], and is associated with non-inferior survival rates compared with PDS [11–12]. Currently, the most common reasons physicians choose to administer NACT are low likelihood of optimal cytoreduction and the presence of reversible contraindications to surgery [17,22–25]. Commonly cited indications for administering NACT include unresectable or stage IV disease, recent thromboembolism or myocardial infarction, recent surgery, poor nutritional status, and poor performance status, although no data exist in the literature to identify specific factors predictive of survival in patients receiving NACT/IDS. As the use of NACT for treatment of advanced EOC becomes more common, more evidence is needed to inform best practices for individualized groups of patients. The objectives of our study were to identify prognostic factors in patients undergoing NACT/IDS. In addition, we wanted to specifically compare survival and postoperative outcomes in patients likely to have unresectable disease undergoing NACT/IDS and stage IIIC/IV patients who had undergone suboptimal PDS. 2. Materials and methods 2.1. Study group and inclusion criteria After obtaining approval from the Institutional Review Board at Mayo Clinic (Rochester, MN), we retrospectively identified all patients who underwent IDS from January 2007 to December 2013 at our institution after receiving NACT. Patients with biopsy-proven ovarian, fallopian tube, or primary peritoneal carcinoma of presumed International Federation of Gynecology and Obstetrics (FIGO) stage IIIC or IV by laparoscopy, laparotomy, or preoperative imaging studies were
included. For each patient included in the study, the decision to use NACT for primary treatment was made by both a gynecologic oncologist and medical oncologist at our institution or by a medical oncologist and/ or gynecologist oncologist at an outside institution. Patients were excluded if they had a non-epithelial or borderline histology, had a nongynecologic primary cancer, had a partial primary cytoreduction, or did not consent for review of their medical records for research. 2.2. Surgical approach and follow up Surgical resectability was determined individually by surgeons at our institution by radiologic and laboratory criteria as previously described [26] and also with laparoscopic exploration in indeterminate cases. Patients were included that received NACT as primary treatment of their ovarian cancer as a result of determinations made by physicians both internally and externally. Once PDS was chosen as the appropriate primary intervention, maximal surgical effort was employed to achieve optimal cytoreduction as previously described [27], or until optimal resection was deemed infeasible. During this timeframe, the percentage of patients who received a suboptimal cytoreduction after being deemed to have resectable disease was approximately 7%. Follow up was recommended per National Comprehensive Cancer Network (NCCN) guidelines for all patients regardless of primary treatment modality. Upon recurrence, follow up was performed as deemed appropriate according to the individual patient's presentation and by each individual surgeon or medical oncologist participating in the patient's care. 2.3. Data collection The medical records of all patients who met the inclusion criteria were reviewed and data including pertinent demographic, clinical, surgical, pathologic, and follow-up information was collected. Operative findings were recorded by each surgeon on a standardized form for all patients undergoing IDS for ovarian cancer at our institution, and this report includes quantitative reporting of initial and residual disease, sites of initial and residual disease, and a listing of operative procedures performed. Patients who had no gross disease at IDS as defined by the operating surgeon were categorized as having complete clinical response (CCR). Optimal cytoreduction was defined as residual disease measuring 1 cm or less in maximal dimension at the conclusion of IDS. Pelvic and paraaortic lymphadenectomy was performed if necessary to achieve maximal cytoreduction at the primary surgeon's discretion. As part of the data collection, all concomitant surgical procedures that were performed were recorded and a surgical complexity score was assigned to each patient as previously described [28]. Postoperative complications within 30 days of surgery were abstracted and scored with the Accordion Severity Classification of postoperative complications (contracted classification) [29]. Recurrence or progression was defined by evidence of recurrent or progressive disease on imaging or pathologic confirmation on biopsy. 2.4. Statistical analysis The outcomes of interest were post-surgical OS (primary) and PFS (secondary), starting from the date of IDS. If no documentation of recurrence or progression was found, the duration of follow-up was censored at the date of last clinically relevant follow-up. If no documentation of death was found, the duration of follow-up was censored at the date of last follow-up by any means. OS and PFS were estimated using the Kaplan-Meier method and factors were evaluated for their association with OS based on fitting univariate and multivariable Cox proportional hazards models. For the multivariable analysis, variables with a P value b 0.20 based on univariate were considered and variables were retained in the final model if their P value was b0.05 using stepwise and backward variable selection methods. Associations were
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
summarized using the hazard ratio (HR) and corresponding 95% confidence interval (CI) estimated from the model parameters. The association between number of postoperative adjuvant cycles and OS was evaluated based on handling the number of cycles as a time-dependent covariate in a Cox model using the counting process approach. At time zero, the time-dependent covariate for the number of cycles was zero, and the value of the time-dependent covariate for each patient was increased by one at the time of each subsequent cycle. The Cox model was fit using age as the time scale to more comprehensively adjust for the effect of age. Lastly, we compared outcomes between primary treatment strategies (NACT/IDS vs PDS followed by platinum-based adjuvant chemotherapy) in patients thought to have unresectable disease. The subset of patients receiving NACT/IDS for the indication of presumed unresectable disease were matched 1:1 on age at surgery (±4 years) and stage to a contemporary cohort of patients who underwent suboptimal PDS between 2003 and 2011. Demographic, clinical, surgical, pathologic, and follow-up information was collected and compared between the matched cohorts. OS, PFS, and 90-day mortality were compared between the two groups based on fitting Cox proportional hazards models. Duration of follow-up started at the date of the surgery in each group (IDS or PDS). Logistic regression models were fit to compare 30-day morbidity (defined as a complication with an Accordion Severity Classification of 3 or 4) between the two groups; associations were summarized using the odds ratio (OR) and corresponding 95% CI. All calculated P values were two-sided and P values b 0.05 were considered statistically significant. Statistical analysis was performed using the SAS version 9.3 software package (SAS Institute, Inc.; Cary, NC). 3. Results 3.1. Neoadjuvant chemotherapy and interval debulking surgery cohort A total of 107 patients presented for IDS at our institution after receiving NACT during the study period. Twenty patients were excluded from analysis for the following reasons: FIGO stage IC disease (n = 6), no consent for medical records to be used for research (n = 5), failure to receive NACT (n = 3), failure to undergo IDS (n = 2), non-epithelial histology (n = 2), endometrial primary (n = 1), and previously attempted PDS (n = 1). Among the 87 patients meeting inclusion criteria, the average age at IDS was 66.6 years, average body mass index (BMI) was 27.9 kg/m2, and most were Caucasian (92.0%). The average interval between initiation of NACT and IDS was 103 days (range, 29 to 464 days). Relevant comorbidities were present in 50 (57.5%) patients, including cardiovascular disease, hypertension, type 2 diabetes, and history of venous thromboembolism or stroke (Table 1). Approximately half of the patients were considered to have FIGO stage IIIC disease (51.7%) the remaining cases were FIGO stage IV disease, due to either malignant pleural effusion or biopsy proven disease outside the abdominal cavity. Most had FIGO grade 3 (96.6%) disease, serous histology (83.9%), and ovarian primary (56.3%). Pre-NACT CA-125 levels, drawn at the time of diagnosis, were markedly elevated at a median of 859.9 U/mL but decreased to a median of 45.0 U/mL pre-IDS. Initial mean albumin levels also improved, rising from pre-NACT to pre-IDS serum analyses (mean ± SD, 3.5 ± 0.6 g/dL vs 4.1 ± 0.3 g/dL). Over 50% of patients had an American Society of Anesthesiologists (ASA) score of 3 at the time of IDS. The determination to either administer NACT or perform PDS was made at our institution in 75% (65/87) of the cohort, while 22 patients were referred for IDS after receiving NACT at outside institutions. Of the 22 who were referred after NACT, 10 received NACT for the indication of unresectable disease and 4 for bulky stage IV disease. Of those given NACT for unresectable disease, 3 had an exploratory surgery to assess the extent of disease before NACT was administered. Eight (9.2%) patients had CCR and most patients (68, 78.2%) were noted to have N 1 cm of disease at IDS after NACT; all patients were resected to 1 cm or less residual disease (RD).
3
Table 1 Patient demographics and oncologic characteristics in NACT/IDS cohort. Characteristic
(N = 87)
Age at IDS (years), mean (SD) BMI (kg/m2), mean (SD) Caucasian, N (%) Comorbidities, N (%) Hypertension Cardiovascular disease Type 2 diabetes mellitus History of VTE History of stroke ASA score at IDS, N (%) 2 3 4 Unknown FIGO stage, N (%) IIIC IV FIGO grade, N (%) 1 2 3 Unknown Serous histology, N (%) Type of cancer, N (%) Ovarian Primary peritoneal Fallopian tube CA-125 level (U/mL), median (IQR) Pre-NACT (N = 86) Pre-IDS (N = 85) Albumin (g/dL) Pre-NACT (N = 64), mean (SD) b3.5, N (%) N3.5, N (%) Pre-IDS (N = 76), mean (SD) b3.5, N (%) N3.5, N (%)
66.6 (10.7) 27.9 (7.1) 80 (92.0) 37 (42.5) 14 (16.1) 10 (11.5) 7 (8.0) 2 (2.3) 40 (46.0) 45 (51.7) 1 (11) 1 (11) 45 (51.7) 42 (48.3) 1 (11) 0 (0.0) 84 (96.6) 2 (2.3) 73 (83.9) 49 (56.3) 34 (39.1) 4 (4.6) 859.9 (415.0, 2508.0) 45.0 (16.0, 173.0) 3.5 (0.6) 30/64 (46.9) 34/64 (53.1) 4.1(0.3) 4/76 (5.3) 72/76 (94.7)
Abbreviations: NACT, neoadjuvant chemotherapy; IDS, interval debulking surgery; SD, standard deviation; BMI, body mass index; VTE, venous thromboembolism; ASA, American Society of Anesthesiologists; FIGO, International Federation of Gynecology and Obstetrics; IQR, interquartile range.
At IDS, complex surgical procedures were necessary to achieve optimal cytoreduction in a significant proportion of patients: diaphragm stripping or resection (27.6%), large bowel resection (23.0%), stripping of pelvic peritoneum (18.4%), splenectomy (11.5%), stripping of abdominal peritoneum (6.9%), small bowel resection (6.9%), liver resection (2.3%), resection of stomach and proximal pyloris (1.1%), and partial bladder resection (1.1%). Significant surgical effort was required after NACT, as indicated by the surgical complexity scores of intermediate in 63.2% and high in 14.9%. Intraoperative complications included one patient with diaphragmatic resection and pancreatic laceration and two patients with major vascular injury. Most patients had no residual disease (66.7%) after IDS, with the remainder receiving an optimal cytoreduction. Among the 87 patients who underwent NACT/IDS, 46 are currently deceased and the median duration of follow-up for the remaining 41 patients was 2.2 years (IQR, 1.1–3.2 years). The median OS was 2.4 years and median PFS was 1.0 year. Table 2 summarizes the factors that were evaluated for an association with overall survival. In univariate analysis, older age, lower ASA score at IDS, and elevated preoperative CA-125 (N 35 U/mL) were associated with poorer OS (Fig. 1A and C). Based on multivariable analysis, older age (adjusted HR 1.60 per 10year increase in age, 95% CI 1.18, 2.16) and elevated CA-125 before IDS (adjusted HR 2.30 for CA-125 N35 U/mL, 95% CI 1.25, 4.23) were significantly associated with worse OS. The amount of initial disease at IDS was not associated with survival, however the number of patients with CCR was small (n = 8). It is notable that for those cases with CCR the median survival was 5.4 years (Fig. 1B). We did observe improved
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
4
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Table 2 Univariate and multivariate analysis of factors associated with death (due to any cause) among patients receiving NACT/IDS. Characteristic
Age at IDS (years) Neoadjuvant chemotherapy cycles b3 (N = 8) 3 (N = 44) N3 (N = 35) ASA score at IDS b3 (N = 40) ≥3 (N = 46) FIGO stage IIIC (N = 45) IV (N = 42) Preoperative CA-125 (U/mL) ≤35 (N = 37) N35 (N = 48) Preoperative albumin (g/dL) ≥3.5 (N = 72) b3.5 (N = 4) Surgical complexity Low (N = 19) Intermediate (N = 55) High (N = 13) Initial disease None (N = 8) ≤1 cm (N = 11) N1 cm (N = 68) Residual disease None (N = 58) ≤1 cm (N = 29)
Univariate analysis
Multivariate analysis
HR (95% CI)
P
HR (95% CI)
1.51 (1.12, 2.03)a
0.007 1.60 (1.18, 2.16) 0.78
0.002
0.03 Referent 0.52 (0.29, 0.95) 0.30 1.37 (0.76, 2.47) Referent 0.02
3.2. Matched cohort: NACT/IDS for unresectable disease vs suboptimal PDS
P
1.39 (0.54, 3.60) Referent 1.15 (0.61, 2.17)
Referent 1.99 (1.09, 3.64)
subset of patients, the risk of death was not associated with the number of cycles of chemotherapy given (HR: 1.11; 95% CI 0.94, 1.31; P = 0.23).
0.008 Referent 2.30 (1.25, 4.23)
0.89 Referent 0.91 (0.22, 3.78) 0.54 Referent 1.24 (0.56, 2.71) 1.70 (0.65, 4.46) 0.39 Referent 2.36 (0.62, 8.95) 1.99 (0.70, 5.65)
Within the 87 patients in the NACT/IDS cohort, we identified 58 cases in the NACT/IDS cohort who were triaged to NACT due to surgeon impression of unresectable disease with maximal surgery. We sought to compare the outcomes of these patients with an age- and stagematched cohort of stage IIIC/IV patients that underwent PDS and were unable to be cytoreduced to 1 cm or less. Among the 90 patients (45 cases from each matched cohort), a higher proportion of patients in the NACT/IDS cohort had an ASA score b3 (60.0% vs 37.8% in PDS cohort; Table 3). Patients in the NACT/IDS cohort also had less ascites (22.2% vs 82.2%) and more complex surgery (surgical complexity intermediate to high: 82.2% vs 42.2% in the PDS cohort; Table 3). Among PDS patients, 57.8% had low complexity surgery, likely representing truncated cases because of unresectable disease. Between matched cohorts, we observed a significant post-surgical survival advantage for NACT/IDS with a median OS of 2.5 years vs 1.4 years (HR for NACT/IDS vs PDS: 0.53; 95% CI 0.32, 0.88; Fig. 2B), with no significant difference in PFS with a median PFS of 1.0 years in both groups (HR for NACT/IDS vs PDS: 0.83; 95% CI 0.49, 1.39, Fig. 2A; Table 4). Morbidity within 30 days of surgery was evaluated using the Accordion Severity Classification of postoperative complications (contracted classification); 11.4% (5/44) of patients in the NACT/IDS cohort had a documented Accordion grade 3 or 4 complication vs 35.6% (16/45) in the PDS cohort (OR for NACT/IDS vs PDS: 0.23; 95% CI 0.08, 0.71; Table 4). Mortality within the first 90 postoperative days was also significantly worse for the PDS group, with one death (2.2%) in the NACT/IDS cohort and nine deaths (20.0%) in the PDS cohort (HR for NACT/IDS vs PDS: 0.10; 95% CI 0.01, 0.82; Table 4). 4. Conclusions
0.08 Referent 1.74 (0.93, 3.23)
Abbreviations: NACT, neoadjuvant chemotherapy; IDS, interval debulking surgery; ASA, American Society of Anesthesiologists; FIGO, International Federation of Gynecology and Obstetrics; HR, hazard ratio; CI, confidence interval. ⁎Variables with a P value b 0.20 based on univariate analysis were considered in a multivariable analysis using stepwise and backward variable selection methods. Variables were retained in the final multivariable model if their P value was b0.05. a Hazard ratio per 10-year increase in age.
survival based on post-IDS RD (0 vs RD b1 cm; HR 1.74; 95% CI 0.93, 3.23; P = 0.08) but this did not reach statistical significance (Table 2, Fig. 1D). Approximately half (50.6%) of cases received three pre-IDS NACT cycles, with a range 1–8 cycles administered. All NACT regimens were platinum and taxane-based consistent with standard primary treatment protocols. The majority of patients received postoperative adjuvant chemotherapy (81/87: 6 patients did not receive adjuvant chemotherapy because of complete clinical and pathologic response after NACT (n = 4), transition to palliative care for progressive disease shortly after surgery (n = 1), or loss to follow-up (n = 1)). For those receiving adjuvant chemotherapy, 36/81 received three cycles, 20/81 received 1 or 2 cycles, 14/81 received four cycles, and 11/81 received five or more cycles. Standard postoperative adjuvant regimens of carboplatin and paclitaxel (either every three-week dosing or dose dense) were administered to 65/ 81 patients, and other regimens included standard platinum-based doublets or single agents. Of the 87 patients in the NACT cohort, 74 patients had complete information on the number of adjuvant chemotherapy cycles administered and the date of the last cycle administered. In this
Primary debulking surgery to minimal residual disease followed by adjuvant platinum-based chemotherapy offers the longest survival to patients with ovarian cancer. However, the use of NACT may benefit selected groups of patients, such as those with unresectable disease or patients with high risk of surgical morbidity and mortality. As the use of NACT increases, it is imperative that additional studies be performed to clarify which patients will benefit most from this approach and to answer important questions such as the ideal duration of pre-IDS chemotherapy, number of post-IDS cycles, and to define which patients benefit from surgery. In this single institution study we were able to identify several prognostic factors for outcomes after NACT- specifically age and normalization of CA-125 prior to IDS. We found a minimal difference in survival based on RD at IDS, which prompts the question of the value of IDS. We also demonstrate that in matched stage IIIC/IV cases unlikely to undergo successful PDS to b1 cm, the perioperative outcomes are clearly superior using a NACT approach, while there are no clear survival benefits from PDS. Factors associated with longer survival after NACT/IDS are younger age and normalization of CA-125 before IDS. The longest survival was observed in patients with a complete clinical response (median OS 5.4 vs 2.2 years), however, due to a limited number of cases (n = 8) this did not reach statistical significance. Consistent with the EORTC randomized trial, debulking to no residual disease showed a trend toward improved survival. Previous studies have shown a correlation between survival and the number of preoperative NACT cycles [18] however, we failed to identify an association between prolonged treatment with chemotherapy and survival. It is interesting to observe that survival did not differ among patients receiving the standard number of chemotherapy cycles (total of 6 cycles) and those receiving additional cycles, considering that patients receiving prolonged treatment likely show clinical signs of more aggressive disease.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
5
Fig. 1. Factors associated with OS advantage among 87 patients treated with NACT/IDS. The following factors were associated with poorer overall survival in the NACT/IDS cohort: A. Elevated CA-125 before IDS (median OS for CA-125 N35 U/mL 1.9 vs 2.6 years; P = 0.02) B. Older age at IDS (median OS for increasing age quartiles 5.4, 2.2, 2.2, and 1.7 years; P = 0.08) C. Lack of complete clinical response (CCR) at IDS (median OS 2.2 years vs 5.4 years; P = 0.17). D. Residual disease after IDS (median OS 1.9 vs 2.6 years; P = 0.08).
In agreement with other studies [30], we observed a dramatic reduction in surgical complications in patients treated with NACT/IDS despite having higher complexity surgery. We also observed nearly 10-fold higher odds of 90-day mortality after PDS compared to NACT/IDS. Some of these differences are likely due to improvements in serum albumin and performance status after NACT/IDS, suggesting that focused preoperative strategies to improve nutrition and performance status
Table 3 Summary of patient demographics and oncologic characteristics among patients receiving NACT/IDS for unresectable disease and patients who underwent suboptimal PDS.a Characteristics
NACT/IDS cohort (N = 45)
PDS cohort (N = 45) At time of NACT At time of IDS At time of PDS
Age (years), mean (SD) ASA score, N (%) b3 N3 Albumin, N (%) b3.5 g/dL N3.5 g/dL Not available FIGO stage, N (%) IIIC IV Ascites, N (%) Surgical complexity score, N (%) Low Intermediate High
64.9 (11.4)
65.2 (11.5)
65.3 (11.7)
0 (0.0) 45 (100.0)
27 (60.0) 18 (40.0)
17 (37.8) 28 (62.2)
10/30 (33.3) 20/30 (66.7) 15
2/39 (5.1) 37/39 (94.9) 6
5/29 (17.2) 24/29 (82.8) 16
– – –
22 (48.9) 23 (51.1) 10 (22.2)
22 (48.9) 23 (51.1) 37 (82.2)
– – –
8 (17.8) 29 (64.4) 8 (17.8)
26 (57.8) 15 (33.3) 4 (8.9)
Abbreviations: ASA, American Society of Anesthesiologists; FIGO, International Federation of Gynecology and Obstetrics; IDS, interval debulking surgery; NACT, neoadjuvant chemotherapy; PDS, primary debulking surgery; SD, standard deviation. a Matched on age (±4 years) and stage.
may be of added value. When comparing the survival curves for NACT and PDS for unresectable disease (Fig. 2A and B) it is interesting to note that the PFS curves are the same, however there is a noticeable separation in the OS curves in the first few months, likely reflecting surgical mortality or changes in treatment goals/plans due to disease progression or poor tolerance to surgery. Outside of these first several months, the OS curves appear similar. While in the NACT/IDS group, we do not see this early mortality, we must point out this is survival post-surgery only. We did not have data on cases that did not survive until IDS, or did not tolerate chemotherapy. Extrapolating from an available RCT having intent to treat data, we assume this is a much smaller initial mortality. The evidence supporting NACT in cases of unresectable disease is strong and adds emphasis to the need for improved markers for resectability and complexity of surgery to achieve low RD, and improved care pathways to reduce surgical morbidity and mortality. Central to the decision to perform PDS or treat primarily with NACT is the assessment of resectability, especially as preoperative extent of tumor burden may be an independent prognostic factor [31–32]. Several methods have been proposed in the literature, including radiologic methods [33–34], peritoneal cancer index (PCI) [35], and laparoscopic scoring [36]. According to the SGO practice guidelines, imaging with CT imaging has limitations for predicting optimal resection because it underestimates the distribution of disease, while PET and MRI require further evaluation before their use can be endorsed. The ASCO and SGO practice guidelines regarding the use of NACT and IDS in women with advanced epithelial ovarian cancer do advocate laparoscopy as a tool to predict surgical resectability and to predict morbidity [37–38], while noting insufficient evidence to fully endorse its standard use. While each of these models aide in the determination of resectability, more reliable methods are needed. The strengths of this study are its focus on identifying factors associated with improved survival in patients receiving NACT/IDS and also the
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
6
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 2. Comparison of PFS and OS following surgery between patients with unresectable disease who underwent NACT/IDS versus patients who underwent suboptimal PDS. A. PFS and B. OS among matched cohorts who either had unresectable disease who underwent NACT/IDS or who underwent suboptimal PDS (median PFS, 1.0 vs 1.0 year; median OS, 2.5 vs 1.4 years).
comparison of outcomes for two treatment strategies in patients with unresectable disease using a uniform aggressive surgical approach. This study is limited by its retrospective design and the limitation of the NACT group to only patients receiving NACT that came to IDS. While we expect the percentage of cases not coming to IDS is small, we did not have data on this complete subset of patients. Hence this study did not include an important group of patients who underwent NACT but did not undergo IDS (secondary to progression or refusal, etc.). To minimize the influence of this important “immortal time
bias”, estimates of survival were calculated from the date of surgery in both the PDS and NACT/IDS cohorts. Absent this approach, calculation of survival from the date of diagnosis would have introduced a guaranteed or immortal time bias favoring the NACT/IDS cohort [39]. We recognize that this method discounts the survival time patients incur during preoperative chemotherapy. The ideal approach would be to account for all patients at the time of assignment to NACT and include in an intent-to-treat analysis, but this is often not possible in retrospective studies. We also recognize that 25% of the patients included in this study
Table 4 Comparison of outcomes following surgery between NACT/IDS cohort and PDS cohort.a Outcome Overall survival Progression-free survival 90-day mortality Grade 3 or 4 postoperative complication within 30 days of surgery
Unadjusted analysis HR (95% CI)b
P
Adjusted analysisc HR (95% CI)b
P
0.53 (0.32, 0.88) 0.83 (0.49, 1.39) 0.10 (0.01, 0.82) OR (95% CI)b 0.23 (0.08, 0.71)
0.01 0.47 0.03 P 0.01
0.53 (0.31, 0.89) 0.79 (0.44, 1.41) 0.15 (0.02, 1.23) OR (95% CI)b 0.15 (0.04, 0.60)
0.02 0.43 0.08 P 0.007
Abbreviations: CI, confidence interval; HR, hazard ratio; IDS, interval debulking surgery; NACT, neoadjuvant chemotherapy; OR, odds ratio; PDS, primary debulking surgery. a Matched on age (±4 years) and stage, 45 patients each. b HR and OR for NACT/IDS vs PDS. c Based on fitting a multivariable model adjusted for surgical complexity (low, intermediate, vs high).
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
were referred for IDS after NACT. Studies have shown that women who present to non-tertiary care centers may receive NACT/IDS while PDS might be the preferred option in tertiary care centers [40]. The patients referred for IDS after NACT in this study all had imaging studies and/or exploratory surgeries to make the determination of unresectability in referring institutions. For the purposes of this study, we assumed the referring institution had similar decision-making practices to our institution. Additional studies are needed to further explore the role of NACT/ IDS in the treatment of advanced EOC. In our study, the most common indication for NACT/IDS was unresectable disease. For the indication of likely unresectable disease, NACT/IDS offers improved survival and reduced morbidity with increased surgical complexity over PDS. However, other studies have clearly shown the benefit of PDS in patients with complete resection of disease [11,15]. Thus, the decision to proceed with NACT/IDS versus PDS hinges on the question of whether a patient has resectable disease. More reliable predictors of resectability are essential in classifying disease preoperatively and should be the subject of further study. The second group of patients likely to benefit from NACT are those with reversible risk factors for serious morbidity and mortality: namely malnutrition and poor performance status due to cancer. We and others have published extensively to refine models to predict surgical morbidity and mortality [41], and these should be used to triage patients. During NACT, aggressive attempts to improve nutrition and fitness should be explored. The role and reversibility of frailty and sarcopenia in ovarian cancer are relevant future research topics. Finally, the value of IDS in patients treated with NACT must be explored. Studies have shown that interval debulking surgeries after unsuccessful attempted resection have minimal impact on survival [11, 42]. This brings into question the added value of IDS after NACT when the indication is unresectable disease. Interestingly, patients with CCR had the longest median OS when compared to patients with residual measurable disease after receiving NACT. To further clarify the role of IDS in this subgroup of patients, comparing those with CCR undergoing IDS with other patients achieving CCR not undergoing IDS would be extremely informative and crucial in defining the role of IDS. In conclusion, patients thought to have unresectable disease are likely best served by NACT/IDS as a primary treatment strategy. Conflict of interest statement We have no conflicts of interest.
References [1] K. Bertelsen, Tumor reduction surgery and long-term survival in advanced ovarian cancer: a DACOVA study, Gynecol. Oncol. 38 (1990) 203–209. [2] C.T. Griffiths, Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma, Natl. Cancer Inst. Monogr. 42 (1975) 101–104. [3] C.T. Griffiths, A.F. Fuller, Intensive surgical and chemotherapeutic management of advanced ovarian cancer, Surg. Clin. North Am. 58 (1978) 131–142. [4] N.F. Hacker, J.S. Berek, L.D. Lagasse, R.K. Nieberg, R.M. Elashoff, Primary cytoreductive surgery for epithelial ovarian cancer, Obstet. Gynecol. 61 (1983) 413–420. [5] G. Delgado, D.H. Oram, E.S. Petrilli, Stage III epithelial ovarian cancer: the role of maximal surgical reduction, Gynecol. Oncol. 18 (1984) 293–298. [6] W.J. Hoskins, Epithelial ovarian carcinoma: principles of primary surgery, Gynecol. Oncol. 55 (1994) S91–S96. [7] W.P. McGuire, W.J. Hoskins, M.F. Brady, P.R. Kucera, E.E. Partridge, K.Y. Look, et al., Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer, N. Engl. J. Med. 334 (1996) 1–6. [8] R.E. Bristow, R.S. Tomacruz, D.K. Armstrong, E.L. Trimble, F.J. Montz, Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis, J. Clin. Oncol. 20 (2002) 1248–1259. [9] R.F. Ozols, B.N. Bundy, B.E. Greer, J.M. Fowler, D. Clarke-Pearson, R.A. Burger, et al., 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. 21 (2003) 3194–3200. [10] I. Vergote, C.G. Trope, F. Amant, G.B. Kristensen, T. Ehlen, N. Johnson, et al., Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer, N. Engl. J. Med. 363 (2010) 943–953. [11] M.E.L. Van der Burg, M. Van Lent, M. Buyse, A. Kobierska, N. Colombo, G. Favalli, et al., The effect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer, N. Engl. J. Med. 332 (1995) 629–634.
7
[12] S. Kehoe, J. Hook, M. Nanakivell, G.C. Jayson, H. Kitchener, T. Lopes, et al., Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial, Lancet 386 (9990) (2015) 249–257. [13] D.S. Chi, E.L. Eisenhauer, J. Lang, J. Huh, L. Haddad, N.R. Abu-Rustum, et al., What is the optimal goal of primary cytoreductive surgery for bulky stage IIIC epithelial ovarian carcinoma (EOC)? Gynecol. Oncol. 103 (2006) 559–564. [14] W.E. Winter III, G.L. Maxwell, C. Tian, J.W. Carlson, R.F. Ozols, P.G. Rose, et al., Gynecologic oncology group. Prognostic factors for stage III epithelial ovarian cancer: a Gynecologic Oncology Group study, J. Clin. Oncol. 25 (2007) 3621–3627. [15] D.S. Chi, F. Musa, F. Dao, O. Zivanovic, Y. Sonoda, M.M. Leitao, et al., An analysis of patients with bulky advanced stage ovarian, tubal, and peritoneal carcinoma treated with primary debulking surgery (PDS) during an identical time period as the randomized EORTC-NCIC trial of PDS vs neoadjuvant chemotherapy (NACT), Gynecol. Oncol. 124 (2012) 10–14. [16] I.B. Vergote, I. De Wever, J. Decloedt, W. Tjalma, M. Van Gramberen, P. van Dam, Neoadjuvant chemotherapy versus primary debulking surgery in advanced ovarian cancer, Semin. Oncol. 2 (3) (2000) 31–36. [17] P.E. Schwartz, J.T. Chambers, R. Makuch, Neoadjuvant chemotherapy for advanced ovarian cancer, Gynecol. Oncol. 53 (1994) 33–37. [18] R. Bristow, D.S. Chi, Platinum-based neoadjuvant chemotherapy and interval surgical cytoreduction for advanced ovarian cancer: a meta-analysis, Gynecol. Oncol. 106 (2006) 1070–1076. [19] W. Kuhn, S. Rutke, K. Späthe, B. Schmalfeldt, G. Florack, B. von Hundelshausen, et al., Neoadjuvant chemotherapy followed by tumor debulking prolongs survival for patientswith poor prognosis in International Federation of Gynecology and Obstetrics Stage IIIC ovarian carcinoma, Cancer 92 (2001) 2585–2591 10. [20] J.Y. Hou, M.G. Kelly, H. Yu, J.N. McAlpine, M. Azodi, T.J. Rutherford, et al., Neoadjuvant chemotherapy lessens surgical morbidity in advanced ovarian cancer and leads to improved survival in stage IV disease, Gynecol. Oncol. 105 (2007) 211–217. [21] S. Kang, B.H. Nam, Does neoadjuvant chemotherapy increase optimal cytoreduction rate in advanced ovarian cancer? Meta-analysis of 21 studies, Ann. Surg. Oncol. 16 (2009) 2315–2320. [22] W.J. Hoskins, W.P. McGuire, M.F. Brady, et al., The effect of diameter of largest residual disease on survival after primary cytoreductive surgery in patients with suboptimal residual epithelial ovarian carcinoma, Am. J. Obstet. Gynecol. 170 (4) (1994) 974–979 (discussion 979–80). [23] T. Van Gorp, F. Amant, P. Neven, et al., The role of neoadjuvant chemotherapy versus primary surgery in the management of stage III ovarian cancer, Cancer Treat. Res. 134 (2007) 387–402. [24] I. Vergote, T. van Gorp, F. Amant, et al., Neoadjuvant chemotherapy for ovarian cancer, Oncology (Williston Park) 19 (12) (2005) 1615–1622 (discussion 1623–30). [25] G.C.E. Stuart, H. Kitchener, M. Bacon, A. duBois, M. Friedlander, J. Ledermann, et al., 2010 Gynecologic Cancer InterGroup (GCIG) consensus statement on clinical trials in ovarian cancer: report from the Fourth Ovarian Cancer Consensus Conference, Int. J. Gynecol. Cancer 21 (4) (2011) 750–755. [26] G.D. Aletti, et al., Ovarian cancer surgical resectability: relative impact of disease, patient status, and surgeon, Gynecol. Oncol. 100 (1) (2006) 33–37. [27] G.D. Aletti, et al., Aggressive surgical effort and improved survival in advanced-stage ovarian cancer, Obstet. Gynecol. 107 (1) (2006) 77–85. [28] G.D. Aletti, S.C. Dowdy, K.C. Podratz, W.A. Cliby, Relationship among surgical complexity, short-term morbidity, and overall survival in primary surgery for advanced ovarian cancer, Am. J. Obstet. Gynecol. 197 (2007) 676. [29] S.M. Strasberg, D.C. Linehan, W.G. Hawkins, The accordion severity grading system of surgical complications, Ann. Surg. 250 (2) (2009) 177–186. [30] A. Fagotti, G. Ferrandina, G. Vizzielli, F. Fanfani, V. Gallotta, V. Chiantera, B. Costantini, P.A. Margariti, S. Gueli Alletti, F. Cosentino, L. Tortorella, G. Scambia, Phase III randomised clinical trial comparing primary surgery versus neoadjuvant chemotherapy in advanced epithelial ovarian cancer with high tumour load (SCORPION trial): final analysis of perioperative outcome, Eur. J. Cancer 59 (2016 May) 22–33. [31] N.S. Horowitz, A. Miller, B. Rungruang, S.D. Richard, N. Rodriguez, M.A. Bookman, C.A. Hamilton, T.C. Krivak, G.L. Maxwell, Does aggressive surgery improve outcomes? Interaction between preoperative disease burden and complex surgery in patients with advanced-stage ovarian cancer: an analysis of GOG 182, J. Clin. Oncol. 10; 33 (8) (2015) 937e43. [32] G. Vizzielli, B. Costantini, L. Tortorella, M. Petrillo, F. Fanfani, V. Chiantera, A. Ercoli, R. Iodice, G. Scambia, A. Fagotti, Influence of intraperitoneal dissemination assessed by laparoscopy on prognosis of advanced ovarian cancer: an exploratory analysis of a single-institution experience, Ann. Surg. Oncol. 21 (12) (2014) 3970e7. [33] R.E. Bristow, L.R. Duska, N.C. Lambrou, E.K. Fishman, M.J. O'Neill, E.L. Trimble, et al., A model for predicting surgical outcome in patients with advanced ovarian carcinoma using computed tomography, Cancer 89 (2000) 1532–1540. [34] J.I. Meyer, A.W. Kennedy, R. Friedman, A. Ayoub, R.C. Zepp, Ovarian carcinoma: value of CT in predicting success of debulking surgery, Am. J. Roentgenol. 165 (1995) 875–878. [35] A.-A.K. Tentes, et al., Peritoneal cancer index: a prognostic indicator of survival in advanced ovarian cancer, European Journal of Surgical Oncology (EJSO) 29 (1) (2003) 69–73. [36] A. Fagotti, et al., Prospective validation of a laparoscopic predictive model for optimal cytoreduction in advanced ovarian carcinoma, Am. J. Obstet. Gynecol. 199 (6) (2008) 642-e1. [37] G. Vizzielli, B. Costantini, L. Tortorella, I. Pitruzzella, V. Gallotta, F. Fanfani, S. Gueli Alletti, F. Cosentino, C. Nero, G. Scambia, A. Fagotti, A laparoscopic risk-adjusted model to predict major complications after primary debulking surgery in ovarian cancer: a single-institution assessment, Gynecol. Oncol. 142 (1) (2016) 19–24. [38] A.A. Wright, K. Bohlke, D.K. Armstrong, M.A. Bookman, W.A. Cliby, R.L. Coleman, D.S. Dizon, J.J. Kash, L.A. Meyer, K.N. Moore, A.B. Olawaiye, J. Oldham, R. Salani, D.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
8
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Sparacio, W.P. Tew, I. Vergote, M.I. Edelson, Neoadjuvant chemotherapy for newly diagnosed, advanced ovarian cancer: Society of Gynecologic Oncology and American Society of Clinical Oncology Clinical Practice Guideline, Gynecol. Oncol. 143 (1) (2016 Oct) 3–15. [39] S. Suissa, Immortal time bias in pharmaco-epidemiology, Am. J. Epidemiol. 167 (2008) 492–499. [40] G. Vizzielli, F. Fanfani, V. Chiantera, L. Tortorella, A. Lucidi, M. Petrillo, B. Costantini, G. Scambia, A. Fagotti, Does the diagnosis center influence the prognosis of ovarian
cancer patients submitted to neoadjuvant chemotherapy? Anticancer Res. 35 (5) (2015 May) 3027–3032. [41] A. Kumar, J.N. Bakkum-Gamez, A.L. Weaver, M.E. McGree, W.A. Cliby, Impact of obesity on surgical and oncologic outcomes in ovarian cancer, Gynecol. Oncol. 135 (1) (2014) 19–24. [42] P.G. Rose, S. Nerenstone, M.F. Brady, D. Clarke-Pearson, G. Olt, S.C. Rubin, et al., Secondary surgical cytoreduction for advanced ovarian carcinoma, N. Engl. J. Med. 351 (24) (2004) 2489–2497.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for epithelial ovarian cancers likely to have suboptimal resection Sarah E. Gill, M.D. a, Michaela E. McGree, B.S. b, Amy L. Weaver, M.S. b, William A. Cliby, M.D. a, Carrie L. Langstraat, M.D. a,⁎ a b
Division of Gynecologic Surgery, Mayo Clinic, Rochester, MN, USA Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA
H I G H L I G H T S • NACT/IDS is associated with improved survival in select patients. • Patients treated with NACT/IDS have reduced perioperative morbidity. • Younger age and normalization of CA-125 prior to IDS may improve survival.
a r t i c l e
i n f o
Article history: Received 16 September 2016 Received in revised form 11 November 2016 Accepted 12 November 2016 Available online xxxx Keywords: Neoadjuvant chemotherapy Interval debulking surgery Ovarian cancer Unresectable
a b s t r a c t Objective. To: a) identify prognostic factors in patients with epithelial ovarian cancer treated with neoadjuvant chemotherapy (NACT) and interval debulking surgery (IDS), and b) compare post-surgical survival between patients treated with NACT/IDS for presumed unresectable disease and stage IIIC/IV patients who underwent suboptimal primary debulking surgery (PDS). Methods. This was a retrospective study of consecutive stage IIIC or IV patients undergoing IDS after NACT at Mayo Clinic from January 2007 to December 2013. A subset of patients receiving NACT/IDS for the indication of unresectable disease were matched 1:1 on age and stage to a cohort of patients who underwent suboptimal PDS between 2003 and 2011. Hazard ratios and corresponding 95% confidence intervals were estimated from Cox proportional hazards models. Results. We identified 87 patients treated with NACT/IDS: the median OS and PFS following surgery was 2.4 and 1.0 years, respectively. Factors associated with significantly worse OS were older age (adjusted HR 1.60 per 10-year increase in age, 95% CI 1.18, 2.16) and elevated CA-125 before IDS (adjusted HR 2.30 for CA-125 N 35 U/mL, 95% CI 1.25, 4.23). Number of adjuvant chemotherapy cycles administered did not have a significant effect on survival. In the matched cohort analysis of presumed unresectable cases undergoing NACT/IDS vs suboptimal PDS cases (n = 45 each), the NACT/IDS group had a significant OS advantage (HR 0.53; 95% CI 0.32, 0.88), and fewer patients experienced a 30-day postoperative Accordion grade 3/4 complication (11% vs 36%, P = 0.01). Conclusions. Younger age and normalization of CA-125 prior to IDS are associated with improved survival with NACT/IDS. For primary EOC where resection to residual disease of 1 cm or less is unlikely, NACT/IDS is associated with improved survival and reduced perioperative morbidity compared to PDS. As these patients are likely best served by NACT/IDS, more reliable predictors of resectability would be valuable. © 2016 Elsevier Inc. All rights reserved.
1. Introduction
⁎ Corresponding author at: Division of Gynecologic Surgery, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. E-mail address: [email protected] (C.L. Langstraat).
Currently, the gold standard for primary management of advanced ovarian cancer is primary debulking surgery (PDS) followed by platinum-based adjuvant chemotherapy, with significantly improved overall survival in patients receiving an optimal cytoreduction [1–9].
http://dx.doi.org/10.1016/j.ygyno.2016.11.021 0090-8258/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
2
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
However, neoadjuvant chemotherapy (NACT) followed by interval debulking surgery (IDS) and adjuvant chemotherapy has been strongly advocated by some groups as a reasonable alternative, particularly for patients with reversible contraindications to primary surgery (poor nutritional status, poor performance status, recent thromboembolism, or myocardial infarction, etc.). Patients with a low probability of optimal cytoreduction may also benefit from NACT. However, there is a paucity of data available to directly compare survival and morbidity outcomes of suboptimal PDS vs NACT/IDS for unresectable disease. Although NACT has gained popularity as a primary treatment for EOC, there is disagreement in the literature regarding the significance of previously reported benefits of this treatment strategy as compared with standard therapy. Several studies, including randomized controlled trials (RCTs), report similar median progression-free survival (PFS) and overall survival (OS) in patients receiving NACT/IDS vs standard PDS followed by platinum-based adjuvant chemotherapy [10– 12]. In the European Organization for Research and Treatment of Cancer (EORTC) trial, median OS and PFS were similar in patients with stage IIIC and IV EOC in the NACT group (median OS 30 months and median PFS 12 months) and the primary surgery group (median OS 29 months and median PFS 12 months) [10]. The CHORUS trial also reported noninferior OS for women with stage III and IV EOC receiving NACT versus primary surgery (median OS 24.1 months vs 22.6 months, respectively) [12]. However, survival and surgical outcomes (e.g. residual disease) in the PDS arms were substantially lower than many single institution and phase III trials [10,13–14]. A retrospective study by Chi et al. analyzed the outcomes of patients undergoing PDS at their institution during the same time period as the EORTC trial, reporting a median PFS of 17 months and median OS of 50 months. They concluded that the median survival observed in the NACT arm of the EORTC trial was more similar to the median survival of patients undergoing suboptimal PDS at their institution [15], a conclusion also reported in other studies [16– 19]. Several meta-analyses also suggest that NACT is associated with poorer outcomes when compared to standard PDS followed by adjuvant chemotherapy [17–18]. These criticisms have led many to wonder if the results of these RCTs are relevant to their own patients. Many physicians advocate for the use of NACT/IDS arguing that it reduces perioperative morbidity [20], increases rates of optimal cytoreduction [21], and is associated with non-inferior survival rates compared with PDS [11–12]. Currently, the most common reasons physicians choose to administer NACT are low likelihood of optimal cytoreduction and the presence of reversible contraindications to surgery [17,22–25]. Commonly cited indications for administering NACT include unresectable or stage IV disease, recent thromboembolism or myocardial infarction, recent surgery, poor nutritional status, and poor performance status, although no data exist in the literature to identify specific factors predictive of survival in patients receiving NACT/IDS. As the use of NACT for treatment of advanced EOC becomes more common, more evidence is needed to inform best practices for individualized groups of patients. The objectives of our study were to identify prognostic factors in patients undergoing NACT/IDS. In addition, we wanted to specifically compare survival and postoperative outcomes in patients likely to have unresectable disease undergoing NACT/IDS and stage IIIC/IV patients who had undergone suboptimal PDS. 2. Materials and methods 2.1. Study group and inclusion criteria After obtaining approval from the Institutional Review Board at Mayo Clinic (Rochester, MN), we retrospectively identified all patients who underwent IDS from January 2007 to December 2013 at our institution after receiving NACT. Patients with biopsy-proven ovarian, fallopian tube, or primary peritoneal carcinoma of presumed International Federation of Gynecology and Obstetrics (FIGO) stage IIIC or IV by laparoscopy, laparotomy, or preoperative imaging studies were
included. For each patient included in the study, the decision to use NACT for primary treatment was made by both a gynecologic oncologist and medical oncologist at our institution or by a medical oncologist and/ or gynecologist oncologist at an outside institution. Patients were excluded if they had a non-epithelial or borderline histology, had a nongynecologic primary cancer, had a partial primary cytoreduction, or did not consent for review of their medical records for research. 2.2. Surgical approach and follow up Surgical resectability was determined individually by surgeons at our institution by radiologic and laboratory criteria as previously described [26] and also with laparoscopic exploration in indeterminate cases. Patients were included that received NACT as primary treatment of their ovarian cancer as a result of determinations made by physicians both internally and externally. Once PDS was chosen as the appropriate primary intervention, maximal surgical effort was employed to achieve optimal cytoreduction as previously described [27], or until optimal resection was deemed infeasible. During this timeframe, the percentage of patients who received a suboptimal cytoreduction after being deemed to have resectable disease was approximately 7%. Follow up was recommended per National Comprehensive Cancer Network (NCCN) guidelines for all patients regardless of primary treatment modality. Upon recurrence, follow up was performed as deemed appropriate according to the individual patient's presentation and by each individual surgeon or medical oncologist participating in the patient's care. 2.3. Data collection The medical records of all patients who met the inclusion criteria were reviewed and data including pertinent demographic, clinical, surgical, pathologic, and follow-up information was collected. Operative findings were recorded by each surgeon on a standardized form for all patients undergoing IDS for ovarian cancer at our institution, and this report includes quantitative reporting of initial and residual disease, sites of initial and residual disease, and a listing of operative procedures performed. Patients who had no gross disease at IDS as defined by the operating surgeon were categorized as having complete clinical response (CCR). Optimal cytoreduction was defined as residual disease measuring 1 cm or less in maximal dimension at the conclusion of IDS. Pelvic and paraaortic lymphadenectomy was performed if necessary to achieve maximal cytoreduction at the primary surgeon's discretion. As part of the data collection, all concomitant surgical procedures that were performed were recorded and a surgical complexity score was assigned to each patient as previously described [28]. Postoperative complications within 30 days of surgery were abstracted and scored with the Accordion Severity Classification of postoperative complications (contracted classification) [29]. Recurrence or progression was defined by evidence of recurrent or progressive disease on imaging or pathologic confirmation on biopsy. 2.4. Statistical analysis The outcomes of interest were post-surgical OS (primary) and PFS (secondary), starting from the date of IDS. If no documentation of recurrence or progression was found, the duration of follow-up was censored at the date of last clinically relevant follow-up. If no documentation of death was found, the duration of follow-up was censored at the date of last follow-up by any means. OS and PFS were estimated using the Kaplan-Meier method and factors were evaluated for their association with OS based on fitting univariate and multivariable Cox proportional hazards models. For the multivariable analysis, variables with a P value b 0.20 based on univariate were considered and variables were retained in the final model if their P value was b0.05 using stepwise and backward variable selection methods. Associations were
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
summarized using the hazard ratio (HR) and corresponding 95% confidence interval (CI) estimated from the model parameters. The association between number of postoperative adjuvant cycles and OS was evaluated based on handling the number of cycles as a time-dependent covariate in a Cox model using the counting process approach. At time zero, the time-dependent covariate for the number of cycles was zero, and the value of the time-dependent covariate for each patient was increased by one at the time of each subsequent cycle. The Cox model was fit using age as the time scale to more comprehensively adjust for the effect of age. Lastly, we compared outcomes between primary treatment strategies (NACT/IDS vs PDS followed by platinum-based adjuvant chemotherapy) in patients thought to have unresectable disease. The subset of patients receiving NACT/IDS for the indication of presumed unresectable disease were matched 1:1 on age at surgery (±4 years) and stage to a contemporary cohort of patients who underwent suboptimal PDS between 2003 and 2011. Demographic, clinical, surgical, pathologic, and follow-up information was collected and compared between the matched cohorts. OS, PFS, and 90-day mortality were compared between the two groups based on fitting Cox proportional hazards models. Duration of follow-up started at the date of the surgery in each group (IDS or PDS). Logistic regression models were fit to compare 30-day morbidity (defined as a complication with an Accordion Severity Classification of 3 or 4) between the two groups; associations were summarized using the odds ratio (OR) and corresponding 95% CI. All calculated P values were two-sided and P values b 0.05 were considered statistically significant. Statistical analysis was performed using the SAS version 9.3 software package (SAS Institute, Inc.; Cary, NC). 3. Results 3.1. Neoadjuvant chemotherapy and interval debulking surgery cohort A total of 107 patients presented for IDS at our institution after receiving NACT during the study period. Twenty patients were excluded from analysis for the following reasons: FIGO stage IC disease (n = 6), no consent for medical records to be used for research (n = 5), failure to receive NACT (n = 3), failure to undergo IDS (n = 2), non-epithelial histology (n = 2), endometrial primary (n = 1), and previously attempted PDS (n = 1). Among the 87 patients meeting inclusion criteria, the average age at IDS was 66.6 years, average body mass index (BMI) was 27.9 kg/m2, and most were Caucasian (92.0%). The average interval between initiation of NACT and IDS was 103 days (range, 29 to 464 days). Relevant comorbidities were present in 50 (57.5%) patients, including cardiovascular disease, hypertension, type 2 diabetes, and history of venous thromboembolism or stroke (Table 1). Approximately half of the patients were considered to have FIGO stage IIIC disease (51.7%) the remaining cases were FIGO stage IV disease, due to either malignant pleural effusion or biopsy proven disease outside the abdominal cavity. Most had FIGO grade 3 (96.6%) disease, serous histology (83.9%), and ovarian primary (56.3%). Pre-NACT CA-125 levels, drawn at the time of diagnosis, were markedly elevated at a median of 859.9 U/mL but decreased to a median of 45.0 U/mL pre-IDS. Initial mean albumin levels also improved, rising from pre-NACT to pre-IDS serum analyses (mean ± SD, 3.5 ± 0.6 g/dL vs 4.1 ± 0.3 g/dL). Over 50% of patients had an American Society of Anesthesiologists (ASA) score of 3 at the time of IDS. The determination to either administer NACT or perform PDS was made at our institution in 75% (65/87) of the cohort, while 22 patients were referred for IDS after receiving NACT at outside institutions. Of the 22 who were referred after NACT, 10 received NACT for the indication of unresectable disease and 4 for bulky stage IV disease. Of those given NACT for unresectable disease, 3 had an exploratory surgery to assess the extent of disease before NACT was administered. Eight (9.2%) patients had CCR and most patients (68, 78.2%) were noted to have N 1 cm of disease at IDS after NACT; all patients were resected to 1 cm or less residual disease (RD).
3
Table 1 Patient demographics and oncologic characteristics in NACT/IDS cohort. Characteristic
(N = 87)
Age at IDS (years), mean (SD) BMI (kg/m2), mean (SD) Caucasian, N (%) Comorbidities, N (%) Hypertension Cardiovascular disease Type 2 diabetes mellitus History of VTE History of stroke ASA score at IDS, N (%) 2 3 4 Unknown FIGO stage, N (%) IIIC IV FIGO grade, N (%) 1 2 3 Unknown Serous histology, N (%) Type of cancer, N (%) Ovarian Primary peritoneal Fallopian tube CA-125 level (U/mL), median (IQR) Pre-NACT (N = 86) Pre-IDS (N = 85) Albumin (g/dL) Pre-NACT (N = 64), mean (SD) b3.5, N (%) N3.5, N (%) Pre-IDS (N = 76), mean (SD) b3.5, N (%) N3.5, N (%)
66.6 (10.7) 27.9 (7.1) 80 (92.0) 37 (42.5) 14 (16.1) 10 (11.5) 7 (8.0) 2 (2.3) 40 (46.0) 45 (51.7) 1 (11) 1 (11) 45 (51.7) 42 (48.3) 1 (11) 0 (0.0) 84 (96.6) 2 (2.3) 73 (83.9) 49 (56.3) 34 (39.1) 4 (4.6) 859.9 (415.0, 2508.0) 45.0 (16.0, 173.0) 3.5 (0.6) 30/64 (46.9) 34/64 (53.1) 4.1(0.3) 4/76 (5.3) 72/76 (94.7)
Abbreviations: NACT, neoadjuvant chemotherapy; IDS, interval debulking surgery; SD, standard deviation; BMI, body mass index; VTE, venous thromboembolism; ASA, American Society of Anesthesiologists; FIGO, International Federation of Gynecology and Obstetrics; IQR, interquartile range.
At IDS, complex surgical procedures were necessary to achieve optimal cytoreduction in a significant proportion of patients: diaphragm stripping or resection (27.6%), large bowel resection (23.0%), stripping of pelvic peritoneum (18.4%), splenectomy (11.5%), stripping of abdominal peritoneum (6.9%), small bowel resection (6.9%), liver resection (2.3%), resection of stomach and proximal pyloris (1.1%), and partial bladder resection (1.1%). Significant surgical effort was required after NACT, as indicated by the surgical complexity scores of intermediate in 63.2% and high in 14.9%. Intraoperative complications included one patient with diaphragmatic resection and pancreatic laceration and two patients with major vascular injury. Most patients had no residual disease (66.7%) after IDS, with the remainder receiving an optimal cytoreduction. Among the 87 patients who underwent NACT/IDS, 46 are currently deceased and the median duration of follow-up for the remaining 41 patients was 2.2 years (IQR, 1.1–3.2 years). The median OS was 2.4 years and median PFS was 1.0 year. Table 2 summarizes the factors that were evaluated for an association with overall survival. In univariate analysis, older age, lower ASA score at IDS, and elevated preoperative CA-125 (N 35 U/mL) were associated with poorer OS (Fig. 1A and C). Based on multivariable analysis, older age (adjusted HR 1.60 per 10year increase in age, 95% CI 1.18, 2.16) and elevated CA-125 before IDS (adjusted HR 2.30 for CA-125 N35 U/mL, 95% CI 1.25, 4.23) were significantly associated with worse OS. The amount of initial disease at IDS was not associated with survival, however the number of patients with CCR was small (n = 8). It is notable that for those cases with CCR the median survival was 5.4 years (Fig. 1B). We did observe improved
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
4
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Table 2 Univariate and multivariate analysis of factors associated with death (due to any cause) among patients receiving NACT/IDS. Characteristic
Age at IDS (years) Neoadjuvant chemotherapy cycles b3 (N = 8) 3 (N = 44) N3 (N = 35) ASA score at IDS b3 (N = 40) ≥3 (N = 46) FIGO stage IIIC (N = 45) IV (N = 42) Preoperative CA-125 (U/mL) ≤35 (N = 37) N35 (N = 48) Preoperative albumin (g/dL) ≥3.5 (N = 72) b3.5 (N = 4) Surgical complexity Low (N = 19) Intermediate (N = 55) High (N = 13) Initial disease None (N = 8) ≤1 cm (N = 11) N1 cm (N = 68) Residual disease None (N = 58) ≤1 cm (N = 29)
Univariate analysis
Multivariate analysis
HR (95% CI)
P
HR (95% CI)
1.51 (1.12, 2.03)a
0.007 1.60 (1.18, 2.16) 0.78
0.002
0.03 Referent 0.52 (0.29, 0.95) 0.30 1.37 (0.76, 2.47) Referent 0.02
3.2. Matched cohort: NACT/IDS for unresectable disease vs suboptimal PDS
P
1.39 (0.54, 3.60) Referent 1.15 (0.61, 2.17)
Referent 1.99 (1.09, 3.64)
subset of patients, the risk of death was not associated with the number of cycles of chemotherapy given (HR: 1.11; 95% CI 0.94, 1.31; P = 0.23).
0.008 Referent 2.30 (1.25, 4.23)
0.89 Referent 0.91 (0.22, 3.78) 0.54 Referent 1.24 (0.56, 2.71) 1.70 (0.65, 4.46) 0.39 Referent 2.36 (0.62, 8.95) 1.99 (0.70, 5.65)
Within the 87 patients in the NACT/IDS cohort, we identified 58 cases in the NACT/IDS cohort who were triaged to NACT due to surgeon impression of unresectable disease with maximal surgery. We sought to compare the outcomes of these patients with an age- and stagematched cohort of stage IIIC/IV patients that underwent PDS and were unable to be cytoreduced to 1 cm or less. Among the 90 patients (45 cases from each matched cohort), a higher proportion of patients in the NACT/IDS cohort had an ASA score b3 (60.0% vs 37.8% in PDS cohort; Table 3). Patients in the NACT/IDS cohort also had less ascites (22.2% vs 82.2%) and more complex surgery (surgical complexity intermediate to high: 82.2% vs 42.2% in the PDS cohort; Table 3). Among PDS patients, 57.8% had low complexity surgery, likely representing truncated cases because of unresectable disease. Between matched cohorts, we observed a significant post-surgical survival advantage for NACT/IDS with a median OS of 2.5 years vs 1.4 years (HR for NACT/IDS vs PDS: 0.53; 95% CI 0.32, 0.88; Fig. 2B), with no significant difference in PFS with a median PFS of 1.0 years in both groups (HR for NACT/IDS vs PDS: 0.83; 95% CI 0.49, 1.39, Fig. 2A; Table 4). Morbidity within 30 days of surgery was evaluated using the Accordion Severity Classification of postoperative complications (contracted classification); 11.4% (5/44) of patients in the NACT/IDS cohort had a documented Accordion grade 3 or 4 complication vs 35.6% (16/45) in the PDS cohort (OR for NACT/IDS vs PDS: 0.23; 95% CI 0.08, 0.71; Table 4). Mortality within the first 90 postoperative days was also significantly worse for the PDS group, with one death (2.2%) in the NACT/IDS cohort and nine deaths (20.0%) in the PDS cohort (HR for NACT/IDS vs PDS: 0.10; 95% CI 0.01, 0.82; Table 4). 4. Conclusions
0.08 Referent 1.74 (0.93, 3.23)
Abbreviations: NACT, neoadjuvant chemotherapy; IDS, interval debulking surgery; ASA, American Society of Anesthesiologists; FIGO, International Federation of Gynecology and Obstetrics; HR, hazard ratio; CI, confidence interval. ⁎Variables with a P value b 0.20 based on univariate analysis were considered in a multivariable analysis using stepwise and backward variable selection methods. Variables were retained in the final multivariable model if their P value was b0.05. a Hazard ratio per 10-year increase in age.
survival based on post-IDS RD (0 vs RD b1 cm; HR 1.74; 95% CI 0.93, 3.23; P = 0.08) but this did not reach statistical significance (Table 2, Fig. 1D). Approximately half (50.6%) of cases received three pre-IDS NACT cycles, with a range 1–8 cycles administered. All NACT regimens were platinum and taxane-based consistent with standard primary treatment protocols. The majority of patients received postoperative adjuvant chemotherapy (81/87: 6 patients did not receive adjuvant chemotherapy because of complete clinical and pathologic response after NACT (n = 4), transition to palliative care for progressive disease shortly after surgery (n = 1), or loss to follow-up (n = 1)). For those receiving adjuvant chemotherapy, 36/81 received three cycles, 20/81 received 1 or 2 cycles, 14/81 received four cycles, and 11/81 received five or more cycles. Standard postoperative adjuvant regimens of carboplatin and paclitaxel (either every three-week dosing or dose dense) were administered to 65/ 81 patients, and other regimens included standard platinum-based doublets or single agents. Of the 87 patients in the NACT cohort, 74 patients had complete information on the number of adjuvant chemotherapy cycles administered and the date of the last cycle administered. In this
Primary debulking surgery to minimal residual disease followed by adjuvant platinum-based chemotherapy offers the longest survival to patients with ovarian cancer. However, the use of NACT may benefit selected groups of patients, such as those with unresectable disease or patients with high risk of surgical morbidity and mortality. As the use of NACT increases, it is imperative that additional studies be performed to clarify which patients will benefit most from this approach and to answer important questions such as the ideal duration of pre-IDS chemotherapy, number of post-IDS cycles, and to define which patients benefit from surgery. In this single institution study we were able to identify several prognostic factors for outcomes after NACT- specifically age and normalization of CA-125 prior to IDS. We found a minimal difference in survival based on RD at IDS, which prompts the question of the value of IDS. We also demonstrate that in matched stage IIIC/IV cases unlikely to undergo successful PDS to b1 cm, the perioperative outcomes are clearly superior using a NACT approach, while there are no clear survival benefits from PDS. Factors associated with longer survival after NACT/IDS are younger age and normalization of CA-125 before IDS. The longest survival was observed in patients with a complete clinical response (median OS 5.4 vs 2.2 years), however, due to a limited number of cases (n = 8) this did not reach statistical significance. Consistent with the EORTC randomized trial, debulking to no residual disease showed a trend toward improved survival. Previous studies have shown a correlation between survival and the number of preoperative NACT cycles [18] however, we failed to identify an association between prolonged treatment with chemotherapy and survival. It is interesting to observe that survival did not differ among patients receiving the standard number of chemotherapy cycles (total of 6 cycles) and those receiving additional cycles, considering that patients receiving prolonged treatment likely show clinical signs of more aggressive disease.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
5
Fig. 1. Factors associated with OS advantage among 87 patients treated with NACT/IDS. The following factors were associated with poorer overall survival in the NACT/IDS cohort: A. Elevated CA-125 before IDS (median OS for CA-125 N35 U/mL 1.9 vs 2.6 years; P = 0.02) B. Older age at IDS (median OS for increasing age quartiles 5.4, 2.2, 2.2, and 1.7 years; P = 0.08) C. Lack of complete clinical response (CCR) at IDS (median OS 2.2 years vs 5.4 years; P = 0.17). D. Residual disease after IDS (median OS 1.9 vs 2.6 years; P = 0.08).
In agreement with other studies [30], we observed a dramatic reduction in surgical complications in patients treated with NACT/IDS despite having higher complexity surgery. We also observed nearly 10-fold higher odds of 90-day mortality after PDS compared to NACT/IDS. Some of these differences are likely due to improvements in serum albumin and performance status after NACT/IDS, suggesting that focused preoperative strategies to improve nutrition and performance status
Table 3 Summary of patient demographics and oncologic characteristics among patients receiving NACT/IDS for unresectable disease and patients who underwent suboptimal PDS.a Characteristics
NACT/IDS cohort (N = 45)
PDS cohort (N = 45) At time of NACT At time of IDS At time of PDS
Age (years), mean (SD) ASA score, N (%) b3 N3 Albumin, N (%) b3.5 g/dL N3.5 g/dL Not available FIGO stage, N (%) IIIC IV Ascites, N (%) Surgical complexity score, N (%) Low Intermediate High
64.9 (11.4)
65.2 (11.5)
65.3 (11.7)
0 (0.0) 45 (100.0)
27 (60.0) 18 (40.0)
17 (37.8) 28 (62.2)
10/30 (33.3) 20/30 (66.7) 15
2/39 (5.1) 37/39 (94.9) 6
5/29 (17.2) 24/29 (82.8) 16
– – –
22 (48.9) 23 (51.1) 10 (22.2)
22 (48.9) 23 (51.1) 37 (82.2)
– – –
8 (17.8) 29 (64.4) 8 (17.8)
26 (57.8) 15 (33.3) 4 (8.9)
Abbreviations: ASA, American Society of Anesthesiologists; FIGO, International Federation of Gynecology and Obstetrics; IDS, interval debulking surgery; NACT, neoadjuvant chemotherapy; PDS, primary debulking surgery; SD, standard deviation. a Matched on age (±4 years) and stage.
may be of added value. When comparing the survival curves for NACT and PDS for unresectable disease (Fig. 2A and B) it is interesting to note that the PFS curves are the same, however there is a noticeable separation in the OS curves in the first few months, likely reflecting surgical mortality or changes in treatment goals/plans due to disease progression or poor tolerance to surgery. Outside of these first several months, the OS curves appear similar. While in the NACT/IDS group, we do not see this early mortality, we must point out this is survival post-surgery only. We did not have data on cases that did not survive until IDS, or did not tolerate chemotherapy. Extrapolating from an available RCT having intent to treat data, we assume this is a much smaller initial mortality. The evidence supporting NACT in cases of unresectable disease is strong and adds emphasis to the need for improved markers for resectability and complexity of surgery to achieve low RD, and improved care pathways to reduce surgical morbidity and mortality. Central to the decision to perform PDS or treat primarily with NACT is the assessment of resectability, especially as preoperative extent of tumor burden may be an independent prognostic factor [31–32]. Several methods have been proposed in the literature, including radiologic methods [33–34], peritoneal cancer index (PCI) [35], and laparoscopic scoring [36]. According to the SGO practice guidelines, imaging with CT imaging has limitations for predicting optimal resection because it underestimates the distribution of disease, while PET and MRI require further evaluation before their use can be endorsed. The ASCO and SGO practice guidelines regarding the use of NACT and IDS in women with advanced epithelial ovarian cancer do advocate laparoscopy as a tool to predict surgical resectability and to predict morbidity [37–38], while noting insufficient evidence to fully endorse its standard use. While each of these models aide in the determination of resectability, more reliable methods are needed. The strengths of this study are its focus on identifying factors associated with improved survival in patients receiving NACT/IDS and also the
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
6
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 2. Comparison of PFS and OS following surgery between patients with unresectable disease who underwent NACT/IDS versus patients who underwent suboptimal PDS. A. PFS and B. OS among matched cohorts who either had unresectable disease who underwent NACT/IDS or who underwent suboptimal PDS (median PFS, 1.0 vs 1.0 year; median OS, 2.5 vs 1.4 years).
comparison of outcomes for two treatment strategies in patients with unresectable disease using a uniform aggressive surgical approach. This study is limited by its retrospective design and the limitation of the NACT group to only patients receiving NACT that came to IDS. While we expect the percentage of cases not coming to IDS is small, we did not have data on this complete subset of patients. Hence this study did not include an important group of patients who underwent NACT but did not undergo IDS (secondary to progression or refusal, etc.). To minimize the influence of this important “immortal time
bias”, estimates of survival were calculated from the date of surgery in both the PDS and NACT/IDS cohorts. Absent this approach, calculation of survival from the date of diagnosis would have introduced a guaranteed or immortal time bias favoring the NACT/IDS cohort [39]. We recognize that this method discounts the survival time patients incur during preoperative chemotherapy. The ideal approach would be to account for all patients at the time of assignment to NACT and include in an intent-to-treat analysis, but this is often not possible in retrospective studies. We also recognize that 25% of the patients included in this study
Table 4 Comparison of outcomes following surgery between NACT/IDS cohort and PDS cohort.a Outcome Overall survival Progression-free survival 90-day mortality Grade 3 or 4 postoperative complication within 30 days of surgery
Unadjusted analysis HR (95% CI)b
P
Adjusted analysisc HR (95% CI)b
P
0.53 (0.32, 0.88) 0.83 (0.49, 1.39) 0.10 (0.01, 0.82) OR (95% CI)b 0.23 (0.08, 0.71)
0.01 0.47 0.03 P 0.01
0.53 (0.31, 0.89) 0.79 (0.44, 1.41) 0.15 (0.02, 1.23) OR (95% CI)b 0.15 (0.04, 0.60)
0.02 0.43 0.08 P 0.007
Abbreviations: CI, confidence interval; HR, hazard ratio; IDS, interval debulking surgery; NACT, neoadjuvant chemotherapy; OR, odds ratio; PDS, primary debulking surgery. a Matched on age (±4 years) and stage, 45 patients each. b HR and OR for NACT/IDS vs PDS. c Based on fitting a multivariable model adjusted for surgical complexity (low, intermediate, vs high).
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
were referred for IDS after NACT. Studies have shown that women who present to non-tertiary care centers may receive NACT/IDS while PDS might be the preferred option in tertiary care centers [40]. The patients referred for IDS after NACT in this study all had imaging studies and/or exploratory surgeries to make the determination of unresectability in referring institutions. For the purposes of this study, we assumed the referring institution had similar decision-making practices to our institution. Additional studies are needed to further explore the role of NACT/ IDS in the treatment of advanced EOC. In our study, the most common indication for NACT/IDS was unresectable disease. For the indication of likely unresectable disease, NACT/IDS offers improved survival and reduced morbidity with increased surgical complexity over PDS. However, other studies have clearly shown the benefit of PDS in patients with complete resection of disease [11,15]. Thus, the decision to proceed with NACT/IDS versus PDS hinges on the question of whether a patient has resectable disease. More reliable predictors of resectability are essential in classifying disease preoperatively and should be the subject of further study. The second group of patients likely to benefit from NACT are those with reversible risk factors for serious morbidity and mortality: namely malnutrition and poor performance status due to cancer. We and others have published extensively to refine models to predict surgical morbidity and mortality [41], and these should be used to triage patients. During NACT, aggressive attempts to improve nutrition and fitness should be explored. The role and reversibility of frailty and sarcopenia in ovarian cancer are relevant future research topics. Finally, the value of IDS in patients treated with NACT must be explored. Studies have shown that interval debulking surgeries after unsuccessful attempted resection have minimal impact on survival [11, 42]. This brings into question the added value of IDS after NACT when the indication is unresectable disease. Interestingly, patients with CCR had the longest median OS when compared to patients with residual measurable disease after receiving NACT. To further clarify the role of IDS in this subgroup of patients, comparing those with CCR undergoing IDS with other patients achieving CCR not undergoing IDS would be extremely informative and crucial in defining the role of IDS. In conclusion, patients thought to have unresectable disease are likely best served by NACT/IDS as a primary treatment strategy. Conflict of interest statement We have no conflicts of interest.
References [1] K. Bertelsen, Tumor reduction surgery and long-term survival in advanced ovarian cancer: a DACOVA study, Gynecol. Oncol. 38 (1990) 203–209. [2] C.T. Griffiths, Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma, Natl. Cancer Inst. Monogr. 42 (1975) 101–104. [3] C.T. Griffiths, A.F. Fuller, Intensive surgical and chemotherapeutic management of advanced ovarian cancer, Surg. Clin. North Am. 58 (1978) 131–142. [4] N.F. Hacker, J.S. Berek, L.D. Lagasse, R.K. Nieberg, R.M. Elashoff, Primary cytoreductive surgery for epithelial ovarian cancer, Obstet. Gynecol. 61 (1983) 413–420. [5] G. Delgado, D.H. Oram, E.S. Petrilli, Stage III epithelial ovarian cancer: the role of maximal surgical reduction, Gynecol. Oncol. 18 (1984) 293–298. [6] W.J. Hoskins, Epithelial ovarian carcinoma: principles of primary surgery, Gynecol. Oncol. 55 (1994) S91–S96. [7] W.P. McGuire, W.J. Hoskins, M.F. Brady, P.R. Kucera, E.E. Partridge, K.Y. Look, et al., Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer, N. Engl. J. Med. 334 (1996) 1–6. [8] R.E. Bristow, R.S. Tomacruz, D.K. Armstrong, E.L. Trimble, F.J. Montz, Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis, J. Clin. Oncol. 20 (2002) 1248–1259. [9] R.F. Ozols, B.N. Bundy, B.E. Greer, J.M. Fowler, D. Clarke-Pearson, R.A. Burger, et al., 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. 21 (2003) 3194–3200. [10] I. Vergote, C.G. Trope, F. Amant, G.B. Kristensen, T. Ehlen, N. Johnson, et al., Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer, N. Engl. J. Med. 363 (2010) 943–953. [11] M.E.L. Van der Burg, M. Van Lent, M. Buyse, A. Kobierska, N. Colombo, G. Favalli, et al., The effect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer, N. Engl. J. Med. 332 (1995) 629–634.
7
[12] S. Kehoe, J. Hook, M. Nanakivell, G.C. Jayson, H. Kitchener, T. Lopes, et al., Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial, Lancet 386 (9990) (2015) 249–257. [13] D.S. Chi, E.L. Eisenhauer, J. Lang, J. Huh, L. Haddad, N.R. Abu-Rustum, et al., What is the optimal goal of primary cytoreductive surgery for bulky stage IIIC epithelial ovarian carcinoma (EOC)? Gynecol. Oncol. 103 (2006) 559–564. [14] W.E. Winter III, G.L. Maxwell, C. Tian, J.W. Carlson, R.F. Ozols, P.G. Rose, et al., Gynecologic oncology group. Prognostic factors for stage III epithelial ovarian cancer: a Gynecologic Oncology Group study, J. Clin. Oncol. 25 (2007) 3621–3627. [15] D.S. Chi, F. Musa, F. Dao, O. Zivanovic, Y. Sonoda, M.M. Leitao, et al., An analysis of patients with bulky advanced stage ovarian, tubal, and peritoneal carcinoma treated with primary debulking surgery (PDS) during an identical time period as the randomized EORTC-NCIC trial of PDS vs neoadjuvant chemotherapy (NACT), Gynecol. Oncol. 124 (2012) 10–14. [16] I.B. Vergote, I. De Wever, J. Decloedt, W. Tjalma, M. Van Gramberen, P. van Dam, Neoadjuvant chemotherapy versus primary debulking surgery in advanced ovarian cancer, Semin. Oncol. 2 (3) (2000) 31–36. [17] P.E. Schwartz, J.T. Chambers, R. Makuch, Neoadjuvant chemotherapy for advanced ovarian cancer, Gynecol. Oncol. 53 (1994) 33–37. [18] R. Bristow, D.S. Chi, Platinum-based neoadjuvant chemotherapy and interval surgical cytoreduction for advanced ovarian cancer: a meta-analysis, Gynecol. Oncol. 106 (2006) 1070–1076. [19] W. Kuhn, S. Rutke, K. Späthe, B. Schmalfeldt, G. Florack, B. von Hundelshausen, et al., Neoadjuvant chemotherapy followed by tumor debulking prolongs survival for patientswith poor prognosis in International Federation of Gynecology and Obstetrics Stage IIIC ovarian carcinoma, Cancer 92 (2001) 2585–2591 10. [20] J.Y. Hou, M.G. Kelly, H. Yu, J.N. McAlpine, M. Azodi, T.J. Rutherford, et al., Neoadjuvant chemotherapy lessens surgical morbidity in advanced ovarian cancer and leads to improved survival in stage IV disease, Gynecol. Oncol. 105 (2007) 211–217. [21] S. Kang, B.H. Nam, Does neoadjuvant chemotherapy increase optimal cytoreduction rate in advanced ovarian cancer? Meta-analysis of 21 studies, Ann. Surg. Oncol. 16 (2009) 2315–2320. [22] W.J. Hoskins, W.P. McGuire, M.F. Brady, et al., The effect of diameter of largest residual disease on survival after primary cytoreductive surgery in patients with suboptimal residual epithelial ovarian carcinoma, Am. J. Obstet. Gynecol. 170 (4) (1994) 974–979 (discussion 979–80). [23] T. Van Gorp, F. Amant, P. Neven, et al., The role of neoadjuvant chemotherapy versus primary surgery in the management of stage III ovarian cancer, Cancer Treat. Res. 134 (2007) 387–402. [24] I. Vergote, T. van Gorp, F. Amant, et al., Neoadjuvant chemotherapy for ovarian cancer, Oncology (Williston Park) 19 (12) (2005) 1615–1622 (discussion 1623–30). [25] G.C.E. Stuart, H. Kitchener, M. Bacon, A. duBois, M. Friedlander, J. Ledermann, et al., 2010 Gynecologic Cancer InterGroup (GCIG) consensus statement on clinical trials in ovarian cancer: report from the Fourth Ovarian Cancer Consensus Conference, Int. J. Gynecol. Cancer 21 (4) (2011) 750–755. [26] G.D. Aletti, et al., Ovarian cancer surgical resectability: relative impact of disease, patient status, and surgeon, Gynecol. Oncol. 100 (1) (2006) 33–37. [27] G.D. Aletti, et al., Aggressive surgical effort and improved survival in advanced-stage ovarian cancer, Obstet. Gynecol. 107 (1) (2006) 77–85. [28] G.D. Aletti, S.C. Dowdy, K.C. Podratz, W.A. Cliby, Relationship among surgical complexity, short-term morbidity, and overall survival in primary surgery for advanced ovarian cancer, Am. J. Obstet. Gynecol. 197 (2007) 676. [29] S.M. Strasberg, D.C. Linehan, W.G. Hawkins, The accordion severity grading system of surgical complications, Ann. Surg. 250 (2) (2009) 177–186. [30] A. Fagotti, G. Ferrandina, G. Vizzielli, F. Fanfani, V. Gallotta, V. Chiantera, B. Costantini, P.A. Margariti, S. Gueli Alletti, F. Cosentino, L. Tortorella, G. Scambia, Phase III randomised clinical trial comparing primary surgery versus neoadjuvant chemotherapy in advanced epithelial ovarian cancer with high tumour load (SCORPION trial): final analysis of perioperative outcome, Eur. J. Cancer 59 (2016 May) 22–33. [31] N.S. Horowitz, A. Miller, B. Rungruang, S.D. Richard, N. Rodriguez, M.A. Bookman, C.A. Hamilton, T.C. Krivak, G.L. Maxwell, Does aggressive surgery improve outcomes? Interaction between preoperative disease burden and complex surgery in patients with advanced-stage ovarian cancer: an analysis of GOG 182, J. Clin. Oncol. 10; 33 (8) (2015) 937e43. [32] G. Vizzielli, B. Costantini, L. Tortorella, M. Petrillo, F. Fanfani, V. Chiantera, A. Ercoli, R. Iodice, G. Scambia, A. Fagotti, Influence of intraperitoneal dissemination assessed by laparoscopy on prognosis of advanced ovarian cancer: an exploratory analysis of a single-institution experience, Ann. Surg. Oncol. 21 (12) (2014) 3970e7. [33] R.E. Bristow, L.R. Duska, N.C. Lambrou, E.K. Fishman, M.J. O'Neill, E.L. Trimble, et al., A model for predicting surgical outcome in patients with advanced ovarian carcinoma using computed tomography, Cancer 89 (2000) 1532–1540. [34] J.I. Meyer, A.W. Kennedy, R. Friedman, A. Ayoub, R.C. Zepp, Ovarian carcinoma: value of CT in predicting success of debulking surgery, Am. J. Roentgenol. 165 (1995) 875–878. [35] A.-A.K. Tentes, et al., Peritoneal cancer index: a prognostic indicator of survival in advanced ovarian cancer, European Journal of Surgical Oncology (EJSO) 29 (1) (2003) 69–73. [36] A. Fagotti, et al., Prospective validation of a laparoscopic predictive model for optimal cytoreduction in advanced ovarian carcinoma, Am. J. Obstet. Gynecol. 199 (6) (2008) 642-e1. [37] G. Vizzielli, B. Costantini, L. Tortorella, I. Pitruzzella, V. Gallotta, F. Fanfani, S. Gueli Alletti, F. Cosentino, C. Nero, G. Scambia, A. Fagotti, A laparoscopic risk-adjusted model to predict major complications after primary debulking surgery in ovarian cancer: a single-institution assessment, Gynecol. Oncol. 142 (1) (2016) 19–24. [38] A.A. Wright, K. Bohlke, D.K. Armstrong, M.A. Bookman, W.A. Cliby, R.L. Coleman, D.S. Dizon, J.J. Kash, L.A. Meyer, K.N. Moore, A.B. Olawaiye, J. Oldham, R. Salani, D.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021
8
S.E. Gill et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Sparacio, W.P. Tew, I. Vergote, M.I. Edelson, Neoadjuvant chemotherapy for newly diagnosed, advanced ovarian cancer: Society of Gynecologic Oncology and American Society of Clinical Oncology Clinical Practice Guideline, Gynecol. Oncol. 143 (1) (2016 Oct) 3–15. [39] S. Suissa, Immortal time bias in pharmaco-epidemiology, Am. J. Epidemiol. 167 (2008) 492–499. [40] G. Vizzielli, F. Fanfani, V. Chiantera, L. Tortorella, A. Lucidi, M. Petrillo, B. Costantini, G. Scambia, A. Fagotti, Does the diagnosis center influence the prognosis of ovarian
cancer patients submitted to neoadjuvant chemotherapy? Anticancer Res. 35 (5) (2015 May) 3027–3032. [41] A. Kumar, J.N. Bakkum-Gamez, A.L. Weaver, M.E. McGree, W.A. Cliby, Impact of obesity on surgical and oncologic outcomes in ovarian cancer, Gynecol. Oncol. 135 (1) (2014) 19–24. [42] P.G. Rose, S. Nerenstone, M.F. Brady, D. Clarke-Pearson, G. Olt, S.C. Rubin, et al., Secondary surgical cytoreduction for advanced ovarian carcinoma, N. Engl. J. Med. 351 (24) (2004) 2489–2497.
Please cite this article as: S.E. Gill, et al., Optimizing the treatment of ovarian cancer: Neoadjuvant chemotherapy and interval debulking versus primary debulking surgery for e..., Gynecol Oncol (2016), http://dx.doi.org/10.1016/j.ygyno.2016.11.021