Abstract
Objective.
The ideal number of neoadjuvant chemotherapy (NACT) cycles prior to interval tumor-reductive surgery (iTRS) for advanced ovarian cancer is poorly defined. We sought to assess survival stratified by number of NACT cycles and residual disease following iTRS in patients with advanced ovarian cancer with partial response (PR) or stable disease (SD) following 3-4 cycles of NACT.
Methods.
We retrospectively identified patients with advanced high-grade ovarian cancer (diagnosed 2/1/2013 to 2/1/2018) who received at least 3 cycles of NACT and iTRS and had a PR or SD. The population was divided into four groups based on the number of NACT cycles prior to iTRS and residual disease status after (CGR [complete gross residual] or incomplete resection [any amount of residual disease]): 1) 3-4 NACT cycles/CGR, 2) 3-4 NACT cycles/incomplete resection, 3) > 4 cycles/CGR, and 4) >4 cycles/incomplete resection. Overall survival (OS) and progression-free survival (PFS) were estimated using a Kaplan-Meier product-limit estimator and modeled using univariable and multivariable Cox proportional hazards analysis.
Results.
The cohort consisted of 265 patients with advanced high-grade ovarian cancer with a median age at diagnosis of 65 years. Most were White (87%), had serous histology (89%), and stage IV disease (57%), with an overall CGR rate of 81%. In a multivariable analysis receipt of >4 NACT cycles was not associated with worse PFS or OS (adjusted hazard ratio [aHR] 1.02, 95% CI 0.74-1.42; aHR 1.12, 95% CI, 0.73-1.72 respectively) than was receipt of 3-4 cycles. Any amount of residual disease was associated with worse PFS and OS regardless of the number of NACT cycles (aHR 1.56, 95% CI 1.09-2.22; aHR 2.38, 95% CI 1.52-3.72 respectively)
Conclusions.
Residual disease was associated with worse survival outcomes regardless of the number of NACT cycles in patients with PR or SD after NACT for advanced high-grade ovarian cancer. These data suggest that the ability to achieve CGR should take precedence in decision-making regarding the timing of surgery.
Keywords: Ovarian Cancer, Neoadjuvant chemotherapy, Interval tumor reductive surgery
1. Introduction
Ovarian cancer is the most common cause of gynecologic cancer-related deaths in the United States, with an estimated 21,750 new cases and 13,940 deaths in 2020 [1], In 2016, almost half of all patients with stage IIIC or IV epithelial ovarian cancer received neoadjuvant chemotherapy (NACT) [2] based on evidence from multiple phase 3 trials and a meta-analysis demonstrating that the sequence of surgery and chemotherapy does not appreciably affect survival [3–6]. However, randomized trials have not prospectively evaluated the optimal timing of interval tumor-reductive surgery (iTRS) in women receiving NACT. The American Society of Clinical Oncology and the Society of Gynecologic Oncology favor no more than 4 cycles of chemotherapy prior to surgery, but if and when to offer surgery to women with inadequate response to the standard number of cycles is less clear [7].
Observational studies in which researchers investigated the number of chemotherapy cycles prior to interval surgery have produced inconsistent results. Some demonstrated worse overall survival (OS) in patients who received more than 4 cycles of chemotherapy, whereas others found no difference in survival [8–12]. Notably, studies that compared women who received more than 4 cycles because they had poor responses with those who had complete responses after 3-4 cycles are unquestionably biased given their failure to account for disease burden in their survival estimates [9,10,12,13]. Other studies are limited by a lack of control groups [14,15], lack of description of why some patients but not others received more than three or four cycles of NACT [16,17], and use of non-contemporaneous controls [8,18]. Therefore, whether to proceed with surgery or more chemotherapy in a patient who does not have an adequate response after three or four cycles of NACT remains unanswered and clinically relevant. To answer this question, we performed a retrospective study to assess survival stratified by number of chemotherapy cycles and residual disease following iTRS in a population of advanced ovarian cancer patients with a partial response (PR) or stable disease (SD) within 3-4 cycles of NACT.
2. Methods
2.1. Cohort
This study was approved by The University of Texas MD Anderson Cancer Center Institutional Review Board (PA16-1010). Patients diagnosed with advanced ovarian cancer from February 1, 2013, to February 1, 2018, and received at least 3 cycles of NACT (based on laparoscopic triage and/or medical and disease criteria) and underwent iTRS were identified. As described previously [19], per consensus among our institution’s gynecologic oncologists, patients with poor Eastern Cooperative Oncology Group (ECOG) scores, medical co-morbidities precluding primary surgery, newly diagnosed deep venous thrombosis or pulmonary embolus within 6 weeks of presentation, or imaging confirmation of unresectable disease were offered NACT. Patients triaged to receive NACT based on laparoscopic scoring assessment were also included [19]. Those who underwent primary tumor-reductive surgery or did not undergo surgery at all during primary treatment were excluded.
2.2. Exposure
Patients in this study received IV taxane/platinum-based NACT in one of two regimens: every 3 weeks or weekly. To establish the cohort, each patient’s disease status following 3-4 cycles of NACT was examined. Only patients who had a PR or SD based on RECIST 1.1 criteria (Response Evaluation Criteria in Solid Tumors) assessment of imaging following 3-4 cycles were included; those identified as having a complete response based on RECIST 1.1 were excluded. Also excluded were those who received fewer than 3 cycles of NACT and those who had disease progression or unknown RECIST after 3-4 cycles. The number of cycles prior to iTRS was per the discretion of the treating physician based on the burden of disease, response to therapy[19,20]. To simulate the clinical decision a surgeon must make when faced with a patient who does not have a complete response to chemotherapy, the cohort was divided into those who received 3-4 cycles of NACT and those who received more than 4 cycles with interval surgery. Given the prognostic importance of residual disease, the cohort was further divided according to outcome of interval surgery: CGR (complete gross resection) and incomplete resection Therefore, 4 patient groups were examined: 1) 3-4 NACT cycles/CGR, 2) 3-4 NACT cycles/incomplete resection, 3) >4 cycles/CGR, and 4) >4 cycles/incomplete resection. The last patient follow-up visit was on September 12, 2019.
2.3. Covariates
Covariates examined in this study consisted of age (continuous), preoperative serologic CA-125 level (continuous), body mass index (continuous), BRCA mutation status (positive, negative, unknown, or variant of unknown significance), histology (serous or non-serous), disease stage (III or IV according to the 2014 International Federation of Gynecology and Obstetrics classification), race (White, Black, Asian, American Indian/Alaskan Native, or other), ethnicity (Hispanic or non-Hispanic), primary disease site (fallopian tubes, ovary, peritoneum, or Müllerian not otherwise specified, ECOG performance status (at presentation; 0-4), American Society of Anesthesiologists physical status classification (II-IV), Charlson Comorbidity Index, and type of platinum-based NACT regimen (every 3 weeks or weekly). Finally, to account for surgical complexity, the classification system described by Aletti et al. [21] was used to assign a surgical complexity score of low (≤3 points), intermediate (4-7 points), or high (≥8 points) based on the surgical maneuvers performed during iTRS (Supplementary Appendix).
2.4. Outcomes
Progression-free survival (PFS) was measured from the date of NACT start to the date of the last clinic visit, first recurrence or progression, or death. OS was estimated from the date of NACT start to the date of death or last follow-up visit whichever came first. Subjects who were event-free and alive at the date of the last clinic visit were censored.
2.5. Statistical methods
Summary statistics such as means, standard deviations, ranges, frequencies, and percentages were used to describe the study population, and comparisons were performed using the chi-square test, Fisher exact test, analysis of variance, or the Kruskal-Wallis test depending on the underlying distribution of the data. OS and PFS were estimated using a Kaplan-Meier product-limit estimator and modeled via univariable Cox proportional hazards analysis. Multivariable Cox proportional hazards models incorporating age, race, Charlson Comorbidity Index, disease stage, BRCA status, surgical complexity score, type of platinum-based NACT regimen, and residual disease were used to assess the independent association of the number of NACT cycles (3-4 vs >4) with PFS and OS and estimate adjusted hazard ratios (aHRs) with corresponding 95% confidence intervals (CI). CA-125 at iTRS was not included in these models given its collinearity with residual disease status. Interaction terms for CGR and the number of NACT cycles were assessed but were non-significant so were not included in our final adjusted multivariable model. All tests were two-sided, with p levels less than 0.05 and CIs exclusive of the null considered significant. All statistical analyses were performed using Stata/MP software (version 16.0; College Station, TX). Study data were collected by a clinical data coordinator and managed using REDCap electronic data capture tools hosted at MD Anderson [22].
2.6. Sensitivity analyses
Analyses were conducted to evaluate the robustness of the study estimates to potential unmeasured confounding. The “E-value,” a measure developed by VanderWeele and Ding [23] is unlike other sensitivity analyses as it does not use subjective investigator parameters, but utilizes estimates derived from the study to quantify how strong an unmeasured confounder would have to be to explain away an observed treatment-outcome relationship. In this study we used the E-value to calculate the magnitude of the association an unmeasured confounder must have with both the exposure (the number of chemotherapy cycles and CGR) and the outcome (OS), to drive the derived estimate to the null (see Supplementary Appendix for details).
To assess whether survival is impacted by the overall number of chemotherapy cycles (neoadjuvant and adjuvant), we modeled OS and PFS in multivariable Cox proportional hazards models, incorporating the total number of chemotherapy cycles in the primary setting as the exposure.
3. Results
From February 1, 2013, to February 1, 2018, 896 patients with advanced ovarian, fallopian tube, or primary peritoneal cancer presented to our institution, of whom 265 met our study inclusion criteria. The CONSORT flow diagram of the cohort selection process is presented in Fig. 1. We excluded those who underwent primary tumor-reductive surgery (N = 276) or had unclear timing of surgery (N = 20). We also excluded 61 patients who received fewer than three cycles of NACT and 200 who had complete response or unknown response after 3-4 cycles. Furthermore, we excluded 64 patients who did not undergo iTRS, 10 with low-grade disease, and 4 with unknown degree of resection at iTRS. The final cohort consisted of 265 patients of which 4 had missing data for residual disease at iTRS.
Fig. 1.
CONSORT flow diagram of the cohort selection process used in this study.
The demographic characteristics of the cohort stratified into four groups based on number of NACT cycles and residual disease are shown in Table 1. Median age at diagnosis was 64 years and median follow up was 34.3 months. The majority of the cohort was White (87%), non-Hispanic (90%), had stage IV disease (57%), and serous histology (89%). Most of the patients received 3-4 NACT cycles (74%) and had a PR to NACT (96% vs 7% who had SD), with an overall CGR rate of 81%. Patients who received more than 4 cycles of NACT were more likely than those who received 3-4 cycles to present with stage IV disease (p = 0.001). Patients who received more than 4 cycles of NACT were also more likely to have received platinum-based NACT every 3 weeks (vs weekly; p=0.04). ECOG scores differed significantly among the four patient groups (p = 0.007), but American Society of Anesthesiologists classes did not (p = 0.3). The distribution of surgical complexity scores was similar among the four groups (p = 0.7). The mean number of NACT cycles in the entire cohort was 4.04 (Standard deviation [SD] 1.59) and the mean number of adjuvant cycles was 3.34 (SD 1.62). Overall, patients received a mean number of 7.38 (SD 2.17) chemotherapy cycles in the primary treatment setting.
Table 1.
Characteristics of women with partial response or stable disease following neoadjuvant chemotherapy for advanced ovarian cancer
| All Patients | Group* | |||||
|---|---|---|---|---|---|---|
| Characteristic | n=265 | 3-4 cycles/CGR (n= 162) |
3-4 cycles/incomplete resection (n=31) |
>4 cycles/CGR (n=50) |
>4 cycles/incomplete resection(n=18) | P-Value |
| Age at diagnosis, median (range) | 65.0 (27-85) | 62.0 (27.0 - 81.0) | 66.0 (48.0 - 84.0) | 64.5 (37.00 - 85.00) | 71.0 (52.0 - 82.0) | 0.008 |
| CA-125(U/ml, median [range])§ | 30.8 (4-4230) | 29.1 (5.9 - 4230) | 104.4 (10.3 - 2882) | 23.7 (4.0 - 303) | 49.5 (11.7 - 3052) | <0.001 |
| Charlson Comorbidity Index, median (range) | 3 (0-11) | 4 (2 - 11) | 3 (0 - 11) | 3 (0 - 10) | 4 (2 - 7) | 0.01 |
| Body Mass Index (kg/m2) | 27.8 (13.9-49.2) | 27.6 (13.9 - 49.2) | 24.7 (18.3 - 34.4) | 28.3 (18.8 - 44.1) | 25.5 (18.1 - 36.4) | 0.07 |
| BRCA Mutation Status | 0.9 | |||||
| No mutation | 169 (79%) | 103 (77%) | 21 (81%) | 31 (82%) | 11 (92%) | |
| BRCA1 | 23 (11%) | 15 (11%) | 3 (11%) | 4 (10%) | 0 | |
| BRCA2 | 15 (7%) | 12 (9%) | 1 (4%) | 1 (3%) | 1 (8%) | |
| VUS | 7 (3%) | 4 (3%) | 1 (4%) | 2 (5%) | 0 | |
| Unknown | 51 | 28 | 5 | 12 | 6 | |
| Histology | 0.03 | |||||
| Serous | 237 (89%) | 150 (93%) | 27 (87%) | 40 (80%) | 18 (100%) | |
| Non-Serous | 28 (11%) | 12 (7%) | 4 (13%) | 10 (20%) | 0 | |
| Stage | 0.001 | |||||
| 3 | 114 (43%) | 77 (48%) | 18 (58%) | 10 (20%) | 6 (33%) | |
| 4 | 150 (57%) | 85 (52%) | 13 (42%) | 40 (80%) | 12 (67%) | |
| Race | 0.8 | |||||
| White | 219 (87%) | 130 (84%) | 29 (97%) | 42 (90%) | 14 (82%) | |
| Black | 19 (7%) | 14 (9%) | 1 (3%) | 3 (6%) | 1 (6%) | |
| Asian | 13 (5%) | 9 (6%) | 0 | 2 (4%) | 2 (12%) | |
| American Indian or Alaskan Native | 2 (1%) | 2 (1%) | 0 | 0 | 0 | |
| Unknown | 12 | 7 | 1 | 3 | 1 | |
| Ethnicity | 0.7 | |||||
| Hispanic or Latino | 25 (10%) | 16 (10%) | 4 (13%) | 3 (6%) | 1 (5%) | |
| Not Hispanic or Latino | 236 (90%) | 144 (89%) | 26 (84%) | 46 (92%) | 17 (95%) | |
| Unknown | 4 | 2 | 1 | 1 | 0 | |
| Primary Disease Site | 0.9 | |||||
| Fallopian Tube | 12 (4.5%) | 8 (5%) | 1 (3%) | 3 (6%) | 0 | |
| Ovary | 207 (78%) | 129 (80%) | 24 (77%) | 40 (80%) | 14 (78%) | |
| Peritoneum | 41 (15.5%) | 23 (14%) | 5 (16%) | 6 (12%) | 4 (22%) | |
| Mullerian NOS | 5 (2%) | 2 (1%) | 1 (3%) | 1 (2%) | 0 | |
| ECOG Performance status | 0.007 | |||||
| 0 | 105 (43%) | 65 (43%) | 9 (30%) | 25 (58%) | 3 (18%) | |
| 1 | 102 (41%) | 68 (45%) | 12 (40%) | 11 (26%) | 10 (59%) | |
| 2 | 24 (10%) | 10 (7%) | 6 (20%) | 6 (14%) | 2 (12%) | |
| 3 | 12 (5%) | 7 (5%) | 3 (10%) | 0 | 2 (12%) | |
| 4 | 2 (1%) | 1 (1%) | 0 | 1 (2%) | 0 | |
| Unknown | 20 | 11 | 1 | 7 | 1 | |
| ASA Physical Status Class | 0.3 | |||||
| II | 26 (10%) | 19 (13%) | 2 (7%) | 3 (7%) | 2 (12%) | |
| III | 204 (78%) | 126 (85%) | 26 (87%) | 38 (90%) | 13 (76%) | |
| IV | 8 (3%) | 3 (2%) | 2 (7%) | 1 (2%) | 2 (12%) | |
| Unknown | 27 | 14 | 1 | 8 | 1 | |
| Surgical Complexity Score | 0.7 | |||||
| Low | 162 (61%) | 93 (57%) | 19 (61%) | 32 (64%) | 14 (78%) | |
| Intermediate | 85 (32%) | 56 (35%) | 11 (36%) | 15 (30%) | 3 (17%) | |
| High | 18 (7%) | 13 (8%) | 1 (3%) | 3 (6%) | 1 (5%) | |
| Platinum-based NACT Regimen | ||||||
| Every 3 weeks | 145 (57%) | 85 (54%) | 12 (40%) | 31 (65%) | 14 (78%) | 0.04 |
| Weekly | 111 (43%) | 71 (45%) | 18 (60%) | 17 (35%) | 4 (22%) | |
| Unknown | 9 | 6 | 1 | 2 | 0 | |
| Median Follow-up Time (months) | 34.3 | 34.8 | 28.0 | 37.7 | 22.4 | |
Data are no. (%) unless otherwise specified. Percentages may not add up to 100 due to rounding
4 patients missing residual disease status and excluded from group categorizations
P-values derived by Pearson chi-square test or Wilcoxon rank-sum test.
Pre-operative CA-125
Abbreviations: NACT, neoadjuvant chemotherapy; CGR, complete gross residual following interval tumor cytoreduction; ECOG, Eastern Cooperative Oncology Group; ASA, American Society of Anesthesiologists; VUS, variant of unknown significance
3.1. PFS
In a multivariable Cox proportional hazards model, receiving more than 4 cycles of NACT was not associated with worse PFS relative to receiving 3-4 cycles (aHR, 1.02, 95% CI 0.74-1.42; p = 0.9; Table 3). Having any amount of residual disease was associated with worse PFS compared to CGR (aHR 1.56, 95% CI 1.09-2.22). We did not find a significant interaction between the number of cycles and CGR, demonstrating that the number of cycles did not have a differential effect on PFS with a CGR (pinteraction = 0.50). Both harboring a BRCA 1/2 pathogenic variant (aHR 0.57, 95% CI 0.35-0.92) and receiving weekly chemotherapy (vs every 3 weeks; HR 0.63, 95% CI 0.47-0.85) were associated with improved PFS.
Table 3.
Multivariate Cox proportional hazards models
| PFS | OS | |||
|---|---|---|---|---|
| Characteristic | HR (95% CI) | P-value | HR (95% CI) | P-value |
| Age at Diagnosis | 1.02 (1.00-1.04) | 0.1 | 1.02 (0.99-1.04) | 0.2 |
| Charlson Comorbidity Index | 0.96 (0.85-1.10) | 0.6 | 1.12 (0.97-1.28) | 0.1 |
| Race | ||||
| White | Ref | Ref | . | |
| Other | 1.20 (0.79-1.83) | 0.4 | 0.82 (0.44-1.53) | 0.5 |
| Stage | ||||
| 3 | Ref | Ref | . | |
| 4 | 1.08 (0.80-1.44) | 0.6 | 1.10 (0.75-1.63) | 0.6 |
| Histology | ||||
| Serous | Ref | Ref | . | |
| Non-Serous | 1.18 (0.76-1.85) | 0.5 | 1.50 (0.84-2.68) | 0.2 |
| BRCA status | ||||
| None | Ref | Ref | . | |
| BRCA1/2 | 0.57 (0.35-0.92) | 0.02 | 0.55 (0.27-1.13) | 0.1 |
| Complexity Score | ||||
| Low | Ref | Ref | . | |
| Intermediate | 0.99 (0.73-1.35) | 0.9 | 0.76 (0.49-1.17) | 0.2 |
| High | 1.77 (1.03-3.03) | 0.04 | 1.32 (0.70-2.49) | 0.4 |
| NACT Chemotherapy Regimen | ||||
| Every 3 weeks | Ref. | Ref. | ||
| Weekly | 0.63 (0.47-0.85) | 0.002 | 0.73 (0.50-1.07) | 0.1 |
| Number of Cycles* | ||||
| 3-4 cycles | Ref | Ref | . | |
| >4 cycles | 1.02 (0.74-1.42) | 0.9 | 1.12 (0.73-1.72) | 0.6 |
| Resection status | ||||
| CGR | Ref | Ref | . | |
| Incomplete resection | 1.56 (1.09-2.22) | 0.01 | 2.38 (1.52-3.72) | <0.001 |
Neoadjuvant chemotherapy cycles
Abbreviations: NACT, neoadjuvant chemotherapy; OS, overall survival; PFS, progression-free survival; CGR, complete gross residual following interval tumor cytoreduction; ECOG, Eastern Cooperative Oncology Group
3.2. OS
In multivariable analysis, receipt of more than 4 cycles of NACT was not associated with worse OS than was receipt of 3-4 cycles (aHR, 1.12, 95% CI 0.73-1.72; p = 0.6; Table 3). Regardless of the number of NACT cycles, the amount of residual disease was associated with over double the risk of death when compared with no residual disease (aHR, 2.38, 95% CI, 1.52-3.72; p <0.001). We did not observe a significant interaction between the number of NACT cycles and CGR, demonstrating that the number of cycles did not have a differential effect on OS with CGR (pinteraction = 0.45). The median OS estimates per treatment group are shown in Table 4, and a Kaplan-Meier curve of OS by treatment group is shown in Fig. 2.
Table 4.
Summary of the median overall survival estimates per group
| Group | N | Events | Median OS (Months) | HR (95% CI) | P-value |
|---|---|---|---|---|---|
| 3-4 cycles/CGR | 162 | 72 | 47.18 | 1 | 1 |
| 3-4 cycles/incomplete resection | 31 | 20 | 29.27 | 2.16 (1.31-3.56) | 0.003 |
| >4 cycles/CGR | 50 | 23 | 47.44 | 1.08 (0.67-1.73) | 0.7 |
| >4 cycles/incomplete resection | 18 | 11 | 28.06 | 3.28 (1.70-6.32) | 0.001 |
OS, overall survival; HR, hazard ratio; CI, confidence interval; CGR, complete gross residual
Fig 2.
2a. Kaplan-Meier curve of PFS by treatment group; 2b. Kaplan-Meier curve of OS by treatment group CGR, complete gross resection; IR, Incomplete resection
3.4. Results of sensitivity analysis
In assessing the sensitivity of our findings to unmeasured confounding, we found that once adjusting for all measured covariates, an unmeasured confounder associated with both receipt of more than four NACT cycles and OS by a HR of 2.35-fold each could move the HR for the association between the two to 2.0 (a prespecified, clinically meaningful level), but a weaker confounding variable could not. Regarding the association between CGR and OS, we found that the adjusted HR of 2.38 could be explained away by an unmeasured confounder associated with both CGR and OS by a HR of 3.03-fold each, but that weaker confounding could not do so.
In a multivariable Cox proportional Hazards model incorporating overall (neoadjuvant and adjuvant) number of cycles as the exposure, we found that regardless of the number of cycles given in the primary treatment of advanced ovarian cancer in this cohort, and incomplete resection was associated with worse PFS and OS (aHR 1.79, 95% CI 1.21-2.65; aHR 2.5, 95% CI 1.53-4.41, respectively; Full model can be found in the supplementary appendix)
4. Discussion
In this retrospective single-institution study, we found that in patients with PR or SD following 3-4 cycles of NACT for advanced high-grade ovarian cancer, residual disease was associated with worse PFS and OS regardless of the number of chemotherapy cycles prior to surgery. In the presence of iTRS to no gross residual disease, receipt of more than four chemotherapy cycles was not associated with worse PFS or OS than was receipt of 3-4 NACT cycles. Patients with any amount of residual disease following interval surgery had worse OS than did those with no residual disease regardless of the number of NACT cycles received.
These data suggest that the ability to achieve an interval resection to no gross residual disease should take precedence in the decision-making regarding the timing of surgery in patients who do not have complete responses to NACT. The standard timing of iTRS after three cycles of neoadjuvant chemotherapy evolved in Europe in the 1990s based on a mathematical model [24] stipulating a short window for cytoreductive surgery before chemotherapy encourages the formation of chemo-resistant clones [18]. Investigators later adopted this regimen in randomized trials, and it became the standard of care for NACT in advanced ovarian cancer [3].
Observational data initially supported the recommendation to perform interval surgery as early as possible following chemotherapy [25]. For example, in a meta-analysis of 21 studies, each incremental NACT cycle was associated with a 4-month decrease in median survival duration [26]. In a retrospective study, ovarian cancer patients who received more than 4 cycles of NACT had 2.28 times worse odds of survival than did those who received four or fewer [16]. Our study is consistent with more recent studies demonstrating no difference in survival estimates based on number of NACT cycles [8–12,27]. However, many studies often compared women who had complete responses to 3-4 cycles with those who had suboptimal responses and higher disease burdens, likely representing distinct groups with difficult-to-compare tumor biology. [9–13]. By limiting our cohort to only advanced ovarian cancer patients with remaining disease on imaging after three or four NACT cycles, we better simulated the clinical question of whether to proceed with surgery or give more chemotherapy to a patient who has SD or a PR after 3-4 cycles.
Our data also support that iTRS can still benefit patients who require additional chemotherapy in order to be considered completely resectable This is consistent with recent series of patients who received interval surgery after 4 or more cycles of NACT, in which only those who underwent complete cytoreduction to no residual disease had a survival advantage [9,28]. Although cytoreduction to no residual disease has emerged as the goal of primary cytoreduction for advanced ovarian cancer [29–31]. the data on the prognostic importance of complete cytoreduction at interval surgery have been less consistent [27,32]. In our study, residual disease of any amount following interval surgery was associated with more than double the risk of death regardless of the number of NACT cycles. This is consistent with studies demonstrating similar HRs associated with no gross residual disease at interval surgery [33,34], even in patients who received more than the standard number of chemotherapy cycles [11,28]. In terms of NACT regimens, we found that a weekly platinum-based regimen was independently associated with improved PFS (vs every 3 weeks), but not associated with OS.
Our study had limitations, including its retrospective design and sample size generated using strict inclusion criteria. In particular, we included few patients who both received more than 4 cycles of chemotherapy and had residual disease at interval surgery, partially due to the high rate of cytoreduction to no residual disease in this cohort. As in all observational studies, we were unable to control for unmeasured confounding. However, our analysis of unmeasured confounding suggests that the results are moderately robust. The study had potential for selection bias. Although we assembled a cohort of patients with incomplete responses (on imaging) to NACT, inherent survival bias favoring those who underwent surgery after 3-4 cycles was likely, as their surgeon determined that each of them had enough of a response, or that they were healthy enough to merit surgery. This differential selection bias would drive our estimates away from the null. Importantly, based on the upper limit of the 95% CI around our estimate, we cannot exclude an approximate 1.7-fold increase in the risk of death with increased number of cycles. In addition, preoperative CA-125 serum levels are associated with both resectability and survival [35–37], however, given the co-linearity of Ca-125 levels and resection status, we did not include CA-125 in our predictive models. Finally, as this cohort was 87% White and treated at a comprehensive cancer center, the results of this study may not be generalizable to all populations.
Despite these limitations, our findings demonstrate that regardless of the number of NACT cycles, residual disease drives PFS and OS in ovarian cancer patients with incomplete response to 3-4 cycles. These data suggest that the ability to achieve no residual disease should take precedence in decision-making regarding the timing of surgery after NACT and that surgery can still benefit patients who need additional chemotherapy to have completely resectable disease.
Supplementary Material
Table 2.
Univariable Cox proportional hazards models
| PFS | OS | |||
|---|---|---|---|---|
| Characteristic | HR (95% CI) | P-value | HR (95% CI) | P-value |
| Age at Diagnosis | 1.01 (1.00-1.03) | 0.03 | 1.03 (1.01-1.05) | 0.001 |
| Charlson Comorbidity Index | 1.03 (0.95-1.10) | 0.5 | 1.13 (1.03-1.24) | 0.01 |
| Race | ||||
| White | Ref | . | Ref | . |
| Other | 0.97 (0.65-1.43) | 0.9 | 0.73 (0.42-1.28) | 0.3 |
| Stage | ||||
| 3 | Ref | . | Ref | . |
| 4 | 1.02 (0.78-1.33) | 0.9 | 0.99 (0.70-1.1) | 0.9 |
| Histology | ||||
| Serous | Ref | . | Ref | . |
| Non-Serous | 1.31 (0.86-1.99) | 0.2 | 1.53 (0.90-2.59) | 0.1 |
| BRCA status | ||||
| None | Ref | . | Ref | . |
| BRCA1/2 | 0.60 (0.41-0.90) | 0.01 | 0.50 (0.27-0.91) | 0.02 |
| ECOG | ||||
| 0 | Ref | . | Ref | . |
| 1 | 1.15 (0.86-1.55) | 0.3 | 1.29 (0.85-1.98) | 0.2 |
| ≥2 | 1.66 (1.12-2.47) | 0.01 | 2.57 (1.57-4.20) | <0.001 |
| Surgical Complexity Score | ||||
| Low | Ref | . | Ref | . |
| Intermediate | 1.02 (0.76-1.35) | 0.9 | 0.75 (0.50-1.11) | 0.1 |
| High | 1.70 (1.01-2.87) | 0.04 | 1.20 (0.65-2.20) | 0.5 |
| NACT Chemotherapy Regimen | ||||
| Every 3 weeks | Ref. | Ref. | ||
| Weekly | 0.67 (0.51-0.88) | 0.004 | 0.81 (0.57-1.17) | 0.3 |
| Number of Cycles* | ||||
| 3-4 cycles | Ref | . | Ref | . |
| >4 cycles | 1.05 (0.79-1.42) | 0.7 | 1.20 (0.81-1.77) | 0.3 |
| Resection status | ||||
| CGR | Ref | . | Ref | . |
| Incomplete resection | 1.50 (1.08-2.09) | 0.01 | 2.40 (1.58-3.63) | <0.001 |
| Group | ||||
| 3-4 cycles/CGR | Ref | . | Ref | . |
| 3-4 cycles/Incomplete resection | 1.39 (0.93-2.08) | 0.1 | 2.16 (1.31-3.56) | 0.003 |
| >4 cycles/CGR | 1.03 (0.73-1.44) | 0.9 | 1.08 (0.67-1.73) | 0.7 |
| >4 cycles/Incomplete resection | 1.81(1.07-3.04) | 0.03 | 3.28 (1.70-6.32) | 0.001 |
Neoadjuvant chemotherapy cycles
Abbreviations: NACT, neoadjuvant chemotherapy; OS, overall survival; PFS, progression-free survival; CCI, Charlson comorbidity index; CGR, complete gross residual following interval tumor cytoreduction; ECOG, Eastern Cooperative Oncology Group
Highlights.
Regardless of number of neoadjuvant chemotherapy cycles, residual disease drove survival outcomes
The ability to achieve no residual disease may take precedence in decision-making about timing of interval surgery
Surgery may still benefit patients who need additional chemotherapy to achieve a complete gross resection
Acknowledgments
Editorial support was provided by Donald R. Norwood of Editing Services, Research Medical Library, at MD Anderson.
Funding
This work was supported in part by the MD Anderson Ovarian Cancer Moon Shot program, the NIH/NCI under award number P30CA016672 the Ovarian Cancer SPORE at MD Anderson (CA217685), and grants from the NCI (K08 CA234333 [to J.A.R.-H.] and 5T32 CA101642 [to K.H.L.]), and American Cancer Society and Frank McGraw Memorial Chair in Cancer Research (to A.K.S.). The funding sources were not involved in the development of our research hypothesis.
Footnotes
Conflict of Interest Statement
R.N., K.H.L, and J.A.R-H have nothing to disclose. N.D.F reports personal fees from Tesaro, personal fees from BMS/Pfizer, personal fees from Glaxo Smith Kline, outside the submitted work; B.M.F. and L.A.M report grants from NIH, during the conduct of the study. A.K.S reports grants from NIH Grants during the conduct of the study as well as consulting fees from Merck and Kiyatec, shareholder position in Biopath, other from Kiyatec, and research funding from M-Trap, outside the submitted work.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- [1].National Cancer Institute: Surveillance, epidemiology, and end results program. Cancer stat facts: ovarian cancer, (n.d.). https://seer.cancer.gov/statfacts/html/ovary.html (accessed January 25, 2021).
- [2].Knisely AT, Clair CM St., Hou JY, Collado FK, Hershman DL, Wright JD, Melamed A, Trends in Primary Treatment and Median Survival Among Women With Advanced-Stage Epithelial Ovarian Cancer in the US From 2004 to 2016, JAMA Netw. Open 3 (2020) e2017517. doi: 10.1001/jamanetworkopen.2020.17517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Vergote I, Tropé CG, Amant F, Kristensen GB, Ehlen T, Johnson N, Verheijen RHM, van der Burg MEL, Lacave AJ, Panici PB, Kenter GG, Casado A, Mendiola C, Coens C, Verleye L, Stuart GCE, Pecorelli S, Reed NS, Neoadjuvant chemotherapy or primary surgery in stage IIIC or IV ovarian cancer., N. Engl. J. Med 363 (2010) 943–953. doi: 10.1056/NEJMoa0908806. [DOI] [PubMed] [Google Scholar]
- [4].Kehoe S, Hook J, Nankivell M, Jayson GC, Kitchener H, Lopes T, Luesley D, Perren T, Bannoo S, Mascarenhas M, Dobbs S, Essapen S, Twigg J, Herod J, McCluggage G, Parmar M, Swart A-M, Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial, Lancet. 386 (2015) 249–257. doi: 10.1016/S0140-6736(14)62223-6. [DOI] [PubMed] [Google Scholar]
- [5].Onda T, Satoh T, Ogawa G, Saito T, Kasamatsu T, Nakanishi T, Mizutani T, Takehara K, Okamoto A, Ushijima K, Kobayashi H, Kawana K, Yokota H, Takano M, Kanao H, Watanabe Y, Yamamoto K, Yaegashi N, Kamura T, Yoshikawa H, Comparison of survival between primary debulking surgery and neoadjuvant chemotherapy for stage III/IV ovarian, tubal and peritoneal cancers in phase III randomised trial, Eur. J. Cancer 130 (2020) 114–125. doi: 10.1016/j.ejca.2020.02.020. [DOI] [PubMed] [Google Scholar]
- [6].Qin M, Jin Y, Ma L, Zhang Y-Y, Pan L-Y, The role of neoadjuvant chemotherapy followed by interval debulking surgery in advanced ovarian cancer: a systematic review and meta-analysis of randomized controlled trials and observational studies, Oncotarget. 9 (2018) 8614–8628. doi: 10.18632/oncotarget.23808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Wright AA, Bohlke K, Armstrong DK, Bookman MA, Cliby WA, Coleman RL, Dizon DS, Kash JJ, Meyer LA, Moore KN, Olawaiye AB, Oldham J, Salani R, Sparacio D, Tew WP, Vergote I, Edelson MI, Neoadjuvant Chemotherapy for Newly Diagnosed, Advanced Ovarian Cancer: Society of Gynecologic Oncology and American Society of Clinical Oncology Clinical Practice Guideline, J. Clin. Oncol (2016). doi: 10.1200/JC0.2016.68.6907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Akladios C, Baldauf J-J, Marchal F, Hummel M, Rebstock L-E, Kurtz J-E, Petit T, Afors K, Mathelin C, Lecointre L, Schrot-Sanyan S, Does the Number of Neoadjuvant Chemotherapy Cycles before Interval Debulking Surgery Influence Survival in Advanced Ovarian Cancer?, Oncology. 91 (2016) 331–340. doi: 10.1159/000449203. [DOI] [PubMed] [Google Scholar]
- [9].Phillips A, Sundar S, Singh K, Nevin J, Elattar A, Kehoe S, Balega J, Complete cytoreduction after five or more cycles of neo-adjuvant chemotherapy confers a survival benefit in advanced ovarian cancer, Eur. J. Surg. Oncol 44 (2018) 760–765. doi: 10.1016/j.ejso.2018.01.097. [DOI] [PubMed] [Google Scholar]
- [10].Angeles MA, Rychlik A, Cabarrou B, Spagnolo E, Guyon F, Pérez-Benavente A, Gil-Moreno A, Siegrist J, Querleu D, Mery E, Gladieff L, Hernández A, Ferron G, Martinez A, A multivariate analysis of the prognostic impact of tumor burden, surgical timing and complexity after complete cytoreduction for advanced ovarian cancer, Gynecol. Oncol 158 (2020) 614–621. doi: 10.1016/j.ygyno.2020.06.495. [DOI] [PubMed] [Google Scholar]
- [11].Yoneoka Y, Ishikawa M, Uehara T, Shimizu H, Uno M, Murakami T, Kato T, Treatment strategies for patients with advanced ovarian cancer undergoing neoadjuvant chemotherapy: interval debulking surgery or additional chemotherapy?, J. Gynecol. Oncol 30 (2019). doi: 10.3802/jgo.2019.30.e81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Yao S-E, Tripcony L, Sanday K, Robertson J, Perrin L, Chetty N, Land R, Garrett A, Obermair A, Nascimento M, Tang A, Jagasia N, Singh P, Nicklin J, Survival outcomes after delayed cytoreduction surgery following neoadjuvant chemotherapy in advanced epithelial ovarian cancer, Int. J. Gynecol. Cancer 30 (2020) 1935–1942. doi: 10.1136/ijgc-2020-001658. [DOI] [PubMed] [Google Scholar]
- [13].Xu X, Deng F, Lv M, Chen X, The number of cycles of neoadjuvant chemotherapy is associated with prognosis of stage IIIc-IV high-grade serous ovarian cancer, Arch. Gynecol. Obstet 295 (2017) 451–458. doi: 10.1007/s00404-016-4256-x. [DOI] [PubMed] [Google Scholar]
- [14].da Costa Miranda V, de Souza Fêde ÂB, dos Anjos CH, da Silva JR, Sanchez FB, da Silva Bessa LR, de Paula Carvalho J, Filho EA, de Freitas D, Del Pilar Estevez Diz M, Neoadjuvant chemotherapy with six cycles of carboplatin and paclitaxel in advanced ovarian cancer patients unsuitable for primary surgery: Safety and effectiveness, Gynecol. Oncol 132 (2014) 287–291. doi: 10.1016/j.ygyno.2013.12.002. [DOI] [PubMed] [Google Scholar]
- [15].Stoeckle E, Bourdarias L, Guyon F, Croce S, Brouste V, Thomas L, Floquet A, Progress in Survival Outcomes in Patients with Advanced Ovarian Cancer Treated by Neo-Adjuvant Platinum/Taxane-Based Chemotherapy and Late Interval Debulking Surgery, Ann. Surg. Oncol 21 (2014) 629–636. doi: 10.1245/s10434-013-3278-x. [DOI] [PubMed] [Google Scholar]
- [16].Colombo PE, Labaki M, Fabbro M, Bertrand M, Mourregot A, Gutowski M, Saint-Aubert B, Quenet F, Rouanet P, Mollevi C, Impact of neoadjuvant chemotherapy cycles prior to interval surgery in patients with advanced epithelial ovarian cancer, Gynecol. Oncol 135 (2014)223–230. doi: 10.1016/j.ygyno.2014.09.002. [DOI] [PubMed] [Google Scholar]
- [17].Altman AD, McGee J, May T, Lane K, Lu L, Xu W, Ghatage P, Rosen B, Neoadjuvant chemotherapy and chemotherapy cycle number: A national multicentre study, Gynecol. Oncol 147 (2017) 257–261. doi: 10.1016/j.ygyno.2017.08.006. [DOI] [PubMed] [Google Scholar]
- [18].Stoeckle E, Boubli B, Floquet A, Brouste V, Sire M, Croce S, Thomas L, Guyon F, Optimal timing of interval debulking surgery in advanced ovarian cancer: yet to be defined?, Eur. J. Obstet. Gynecol. Reprod. Biol 159 (2011) 407–412. doi: 10.1016/j.ejogrb.2011.07.014. [DOI] [PubMed] [Google Scholar]
- [19].Fleming ND, Nick AM, Coleman RL, Westin SN, Ramirez PT, Soliman PT, Fellman B, Meyer LA, Schmeler KM, Lu KH, Sood AK, Laparoscopic Surgical Algorithm to Triage the Timing of Tumor Reductive Surgery in Advanced Ovarian Cancer, Obstet. Gynecol 132 (2018) 545–554. doi: 10.1097/AOG.0000000000002796. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Patel A, Iyer P, Matsuzaki S, Matsuo K, Sood AK, Fleming ND, Emerging Trends in Neoadjuvant Chemotherapy for Ovarian Cancer, Cancers (Basel). 13 (2021) 626. doi: 10.3390/cancersl3040626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Aletti GD, Podratz KC, Moriarty JP, Cliby WA, Long KH, Aggressive and complex surgery for advanced ovarian cancer: An economic analysis, Gynecol. Oncol 112 (2009) 16–21. doi: 10.1016/j.ygyno.2008.10.008. [DOI] [PubMed] [Google Scholar]
- [22].Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG, Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support, J. Biomed. Inform 42 (2009) 377–381. doi: 10.1016/jJbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].VanderWeele TJ, Ding P, Sensitivity Analysis in Observational Research: Introducing the E-Value, Ann. Intern. Med 167 (2017) 268. doi: 10.7326/M16-2607. [DOI] [PubMed] [Google Scholar]
- [24].Goldie JH, Coldman AJ, A mathematic model for relating the drug sensitivity of tumors to their spontaneous mutation rate., Cancer Treat. Rep 63 (n.d.) 1727–33. [PubMed] [Google Scholar]
- [25].Rauh-Hain JA, Nitschmann CC, Worley MJ, Bradford LS, Berkowitz RS, Schorge JO, Campos SM, del Carmen MG, Horowitz NS, Platinum resistance after neoadjuvant chemotherapy compared to primary surgery in patients with advanced epithelial ovarian carcinoma, Gynecol. Oncol 129 (2013) 63–68. doi: 10.1016/j.ygyno.2013.01.009. [DOI] [PubMed] [Google Scholar]
- [26].Bristow RE, Chi DS, Platinum-based neoadjuvant chemotherapy and interval surgical cytoreduction for advanced ovarian cancer: A meta-analysis, Gynecol. Oncol 103 (2006) 1070–1076. doi: 10.1016/j.ygyno.2006.06.025. [DOI] [PubMed] [Google Scholar]
- [27].Kang S, Nam B-H, Does Neoadjuvant Chemotherapy Increase Optimal Cytoreduction Rate in Advanced Ovarian Cancer? Meta-Analysis of 21 Studies, Ann. Surg. Oncol 16 (2009) 2315–2320. doi: 10.1245/sl0434-009-0558-6. [DOI] [PubMed] [Google Scholar]
- [28].Plett H, Filippova OT, Garbi A, Kommoss S, Rosendahl M, Langstraat C, Phadnis S, Muallem MZ, Baert T, Chi DS, Aletti GD, Taran F-A, Ramspott JP, Zivanovic O, du Bois A, Sonoda Y, Gardner G, Traut A, Roche KL, Harter P, Role of delayed interval debulking for persistent residual disease after more than 5 cycles of chemotherapy for primary advanced ovarian cancer. An international multicenter study, Gynecol. Oncol 159 (2020) 434–441. doi: 10.1016/j.ygyno.2020.08.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Straubhar A, Chi DS, Long Roche K, Update on the role of surgery in the management of advanced epithelial ovarian cancer., Clin. Adv. Hematol. Oncol 18 (2020) 723–731. [PubMed] [Google Scholar]
- [30].Horowitz NS, Miller A, Rungruang B, Richard SD, Rodriguez N, Bookman MA, Hamilton CA, Krivak TC, Maxwell GL, 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 33 (2015) 937–943. doi: 10.1200/JC0.2014.56.3106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Winter WE, Maxwell GL, Tian C, Sundborg MJ, Rose GS, Rose PG, Rubin SC, Muggia F, McGuire WP, Tumor Residual After Surgical Cytoreduction in Prediction of Clinical Outcome in Stage IV Epithelial Ovarian Cancer: A Gynecologic Oncology Group Study, J. Clin. Oncol 26 (2008) 83–89. doi: 10.1200/JC0.2007.13.1953. [DOI] [PubMed] [Google Scholar]
- [32].Mysona DP, Ghamande S, She J-X, Tran L, Tran P, Rungruang BJ, Chan JK, Bae-Jump V, Gehrig PA, Are There Survival Differences Between Women with Equivalent Residual Disease After Interval Cytoreductive Surgery Compared with Primary Cytoreductive Surgery for Advanced Ovarian and Peritoneal Cancer?, Ann. Surg. Oncol (2020). doi: 10.1245/sl0434-020-09304-w. [DOI] [PubMed] [Google Scholar]
- [33].Liang MI, Prendergast EN, Staples JN, Holschneider CH, Cohen JG, Cass I, Prognostic role of pathologic response and cytoreductive status at interval debulking surgery after neoadjuvant chemotherapy for advanced epithelial ovarian cancer, J. Surg. Oncol (2019) jso.25612. doi: 10.1002/jso.25612. [DOI] [PubMed] [Google Scholar]
- [34].Timmermans M, van der Hel O, Sonke GS, Van de Vijver KK, van der Aa MA, Kruitwagen RF, The prognostic value of residual disease after neoadjuvant chemotherapy in advanced ovarian cancer; A systematic review, Gynecol. Oncol 153 (2019) 445–451. doi: 10.1016/j.ygyno.2019.02.019. [DOI] [PubMed] [Google Scholar]
- [35].Kessous R, Wissing MD, Piedimonte S, Abitbol J, Kogan L, Laskov I, Yasmeen A, Salvador S, Lau S, Gotlieb WH, CA-125 reduction during neoadjuvant chemotherapy is associated with success of cytoreductive surgery and outcome of patients with advanced high-grade ovarian cancer, Acta Obstet. Gynecol. Scand 99 (2020) 933–940. doi: 10.1111/aogs.13814. [DOI] [PubMed] [Google Scholar]
- [36].You B, Robelin P, Tod M, Louvet C, Lotz J-P, Abadie-Lacourtoisie S, Fabbro M, Desauw C, Bonichon-Lamichhane N, Kurtz J-E, Follana P, Leheurteur M, Del Piano F, Ferron G, De Rauglaudre G, Ray-Coquard I, Combe P, Chevalier-Place A, Joly F, Leary A, Pujade-Lauraine E, Freyer G, Colomban O, CA-125 ELIMination Rate Constant K (KELIM) Is a Marker of Chemosensitivity in Patients with Ovarian Cancer: Results from the Phase II CHIVA Trial, Clin. Cancer Res 26 (2020) 4625–4632. doi: 10.1158/1078-0432.CCR-20-0054. [DOI] [PubMed] [Google Scholar]
- [37].Mahdi H, Maurer KA, Nutter B, Rose PG, The Impact of Percent Reduction in CA-125 Levels on Prediction of the Extent of Interval Cytoreduction and Outcome in Patients With Advanced-Stage Cancer of Müllerian Origin Treated With Neoadjuvant Chemotherapy, Int. J. Gynecol. Cancer 25 (2015) 823–829. doi: 10.1097/IGC.0000000000000434. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.