Abstract
Background/Aim: This study aimed to identify differences in the pattern and timing of recurrence in patients with advanced ovarian cancer undergoing primary (PDS) or interval debulking surgery after neoadjuvant chemotherapy (NACT). Patients and Methods: Data were prospectively collected on 105 patients from June 2016 to March 2020. Results: There were 30 (50%) recurrences in the PDS group compared to 32 (72.7%) in the NACT group (p=0.020). An intra-abdominal relapse was more common in NACT compared to PDS patients (64.4% vs. 38.3%, p=0.008) and a recurrence in two or more sites (NACT: 44.4% vs. PDS: 23.3%, p=0.010). Among completely cytoreduced patients, a pelvic recurrence was more frequent in NACT patients (NACT: 50% vs. 22.4% in PDS, p=0.011). Progression-free survival (PFS) was longer in PDS compared to NACT (27 vs. 16 months, p=0.039). Conclusion: NACT patients experienced an unfavorable distribution and timing of recurrent disease compared to patient who underwent PDS.
- Recurrence of ovarian cancer
- primary debulking surgery
- interval debulking surgery
- progression-free survival
- metastasis
Ovarian cancer is the second most common malignancy of the female genital tract in the United Kingdom, with 7,495 new cases reported in 2018. Approximately 80% of patients present with advanced disease (FIGO Stage III-IV) (1). A combination of extensive cytoreductive procedures with platinum-based chemotherapy remains the cornerstone treatment for these patients. Extensive debulking surgery is commonly required to achieve complete eradication of tumor, which is associated with improved survival outcomes compared to suboptimal surgical results (2–5). The timing of chemotherapy has been a field of controversy over recent years, despite data from two randomized studies demonstrating that interval debulking surgery (IDS) after 3 or 4 cycles of neoadjuvant chemotherapy (NACT) generates similar survival outcomes with those of primary debulking surgery (PDS) (2, 3, 6, 7). Since the main criticism on both studies was related to low rates of complete cytoreduction in the PDS arm, the Trial of Radical Upfront Surgical Treatment (TRUST) results, which focus on enhanced surgical effort and resection rates, are highly anticipated (8).
Despite the increasing surgical effort and the development of new therapeutic agents, most patients with advanced epithelial ovarian cancer (EOC) will eventually relapse, which is detrimental to their overall prognosis (9). The interval between completing chemotherapy and time of recurrence and the pattern of recurrent disease [site(s) and volume] have been suggested to affect post-recurrence survival (10–12). Patients with a shorter disease-free interval period or with more disseminated recurrent disease are considered to have a worse prognosis. However, similarly to primary ovarian cancer, recurrent disease demonstrates significant heterogeneity in many aspects; hence, further prognostic indicators are required to deliver tailored treatments and improved oncological outcomes.
Our study aimed to describe the anatomical distribution of recurrence in women surgically treated for FIGO Stage III and IV EOC and identify differences between PDS and NACT groups. In addition, we attempted to delineate whether specific surgical procedures were related to site-specific recurrences. We also compared the frequency of relapse in specific anatomical regions in completely cytoreduced patients between the two groups. Finally, we estimated the progression-free survival (PFS) in each treatment group.
Patients and Methods
Our study population included all women undergoing elective surgery for FIGO stage III and IV EOC at Poole Hospital, a tertiary referral center for gynecological cancer in the United Kingdom, from June 2016 to March 2020. Perioperative and surveillance data were collected prospectively on all patients including operation findings, residual disease, and sites of residual disease in optimally cytoreduced patients, and surgical procedures required to achieve complete or optimal cytoreduction. We included in the analysis patients that achieved complete or at least optimal cytoreduction. We excluded patients with non-epithelial or borderline tumors.
Follow-up of the patients included hospital-based physical examination and assessment of new and potentially tumor-related symptoms every three to four months during the first two years and then every six months for the following three years after primary treatment. Serum Ca-125 levels were measured only when indicated, as per the British Gynaecological Cancer Society (BGCS) guidelines, as no improved survival has been demonstrated (13). Patients underwent a computed tomography (CT) of the chest, abdomen, and pelvis when they reported symptoms suspicious for recurrence or when physical examination was abnormal. A recurrence was confirmed either clinically or with imaging, not serologically with an isolated elevated CA125. Histological or cytological confirmation of relapse was not routinely performed if there was strong clinical and radiographic evidence. A positron emission tomography (PET) scan was performed in diagnostic dilemmas or during the work-up pathway of candidates for secondary debulking surgery. All patients with recurrent ovarian cancer had a Multidisciplinary team review where all CT scans were reviewed by an experienced Gynecology radiology Consultant and sites/volume of recurrence were documented on a Multidisciplinary Team (MDT) summary. In addition, all radiological reports were reviewed retrospectively, and all sites of recurrence were documented.
Initially, the sites of recurrences were divided into four groups: pelvis, upper abdomen and hemidiaphragms, distant (including liver and splenic parenchyma, thorax, and other extra-abdominal sites) and lymph nodes. A patient could have recurred in more than one of the above sites. We also documented those patients who presented with carcinomatosis. We then divided the patients into two large groups, those with intra-abdominal and those with extra-abdominal relapse. In addition, we recorded the presence of ascites or pleural effusion.
Furthermore, we categorized the sites of recurrent disease in completely cytoreduced patients into two groups: upper abdomen/diaphragm/intra-abdominal (small bowel mesentery, small bowel surface, paracolic gutters) and pelvis.
Finally, we explored whether specifically modified posterior exenteration or radical procedures in the upper abdomen (diaphragmatic stripping, splenectomy, cholecystectomy, excision of disease from porta hepatis or lesser sac) had an impact on recurrence to the aforementioned areas.
Quantitative variables were presented as median (IQR) values. Qualitative variables were expressed as absolute and relative frequencies (N, %). Normal distribution of data was assessed using the Kolmogorov–Smirnov test. Mann–Whitney U-test was used to determine if there was a statistically significant difference in age, preoperative measurement CA125, BMI, follow-up, and disease-free time between the two groups, when the normality assumption was violated. Pearson Chi-square test or Fischer’s exact test was used to determine the significant association between the treatment groups (PDS and NACT) and various categorical variables.
The endpoint of follow-up information was June 2021. Progression-free survival (PFS) was defined as the interval from the date of surgery for the PDS group and the date of commencement of neoadjuvant chemotherapy for the NACT group until progression or death. A Kaplan–Meier curve was drawn to estimate progression-free survival, and significant differences between the two groups were assessed using the log-rank test.
All statistical analyses were performed by IBM SPSS Statistics version 25.0, Armonk, NY, USA. The aforementioned statistical tests were performed at a 0.05 significance level.
The operations included in this study were performed according to the standard management for advanced ovarian cancer and the patients gave informed written consent in all cases; hence no ethical approval was required.
Results
In the duration of our study, 105 patients who underwent surgery for FIGO Stage III and IV ovarian cancer had a complete or optimal cytoreduction. Fifteen patients (12.5%) that had a suboptimal or open and close laparotomy were excluded from this study. The median age of the patients was 65 years (IQR=55.5-72 years) with a median BMI of 25.3 (range=23.1-28). The majority of the patients had serous histology (91.4%). Overall, 60 (57.1%) patients had PDS, while 45(42.9%) underwent IDS after NACT, including those who had surgery after six cycles of NACT (N=6). Approximately 65.7% had FIGO Stage III disease and 34.3% Stage IV. More patients with Stage III disease underwent PDS than IDS following NACT (80% vs. 46.7%, p<0.001). Table I summarizes the population’s characteristics.
Characteristics of the population.
The median follow-up time was 29 months (range=19.5-38 months). There were 62 (59.6%) recurrences, 30 (50%) in the primary group compared to 32 (72.7%) in the NACT group (p=0.020). The sites of recurrence are shown in Table II.
Sites of recurrences in total sample and according to group.
The pelvis was the most common site of relapse in our study (35.2%) and was detected in a significantly higher percentage in the NACT group than the PDS group (51.1% vs. 23.3%, p=0.003). Similarly, intra-abdominal recurrences in the NACT group were significantly higher compared to those in the PDS group (64.4% vs. 38.3%, p=0.008).
An intra-abdominal relapse was far more common than an extra-abdominal one (49.5% vs. 11.4%), with the same trend noticed in both groups. When analyzing the total number of different sites of relapse (pelvis, upper abdomen/diaphragms, lymphatic and distant, it was higher in the NACT group. 44.4% of women in the NACT group had a recurrence in two or more different sites, compared to 23.3% in the PDS group (p=0.01). Finally, ascites or pleural effusion was observed more frequently in the NACT group (28.9% vs. 10%, p=0.013).
Further comparison of the frequency of upper abdominal/diaphragmatic/intra-abdominal and pelvic recurrences in the PDS and NACT groups, in patients completely cytoreduced following surgery, is shown in Table III.
Recurrences in upper abdomen/diaphragm/intra-abdominal and pelvis in the PDS and NACT group in patients with no residual disease (completely cytoreduced).
Overall, 50% of NACT patients suffered a pelvic recurrence compared to 22.4% of PDS patients (p=0.011). Similarly, 60% of NACT patients experienced an upper abdominal/diaphragmatic/intra-abdominal relapse compared to 36.7% of PDS patients (p=0.044).
Twenty-seven patients (28.4%) had to undergo a modified posterior exenteration (MPE) to achieve at least an optimal surgical outcome, with the vast majority of those in the primary surgical setting (85.2%). Pelvic recurrence was documented in 7 (25.9%) of those, while the corresponding percentage of pelvic recurrence for the rest of the patients that did not undergo MPE had a higher trend (38.5%, p=0.240) (Table IV).
Frequency of pelvic or upper abdominal recurrence in patients who underwent modified posterior exenteration or radical upper abdominal procedure respectively.
Patients who underwent at least a radical procedure in the upper abdomen were less likely to recur in this area than patients who did not undergo upper abdominal debulking (Table IV). However, this difference was not statistically significant (16.7% vs. 29.4%, p=0.120). Patients with upfront surgery were more likely to require upper abdominal radical procedures than NACT patients (68.5% in PDS vs. 31.5% in NACT).
The PFS in the PDS group was significantly higher (median PFS, 27 months, 95%CI=9.3-44.7 months, p=0.039) compared to the NACT group (median PFS, 16 months, 95%CI=11.3-20.7 months) (Figure 1). The median PFS in the whole population was 21 months (95%CI=15.5-26.5 months).
Progression-free survival by group.
Discussion
Most studies focus on comparing primary and interval debulking surgery for advanced ovarian cancer in terms of survival, whereas less is known about the patterns of recurrences in these groups. In this study, we report the differences in the manifestation of recurrent disease. Women in the NACT group experienced an unfavorable recurrence pattern in relation to the number of sites of recurrences, generalized carcinomatosis and the presence of ascites or pleural effusion compared to PDS. One explanation for this discrepancy could be that women who received NACT had more commonly FIGO Stage IV disease than the PDS group. Despite them having more frequently FIGO Stage IV disease, we had demonstrated in our previous study similarly high rates of complete and optimal cytoreduction between the PDS and NACT group, which counters to some extent the FIGO Stage difference (14). In particular, we achieved complete resection rates of 65% and 58% in the PDS and NACT group, with combined rates of complete and optimal cytoreduction of 81.7% and 92%, respectively. The fact that NACT patients experienced more commonly intra-abdominal recurrences compared to PDS patients (64.4% vs. 38.3%, p=0.008) rather than distant recurrences (11.1% vs. 11.7%, p=0.929) underlines that the variation in FIGO Stage distribution between the two groups could not justify the discrepancy in the patterns of recurrence.
Petrillo et al. reported similar results and hypothesized that these differences could be due to NACT leading to earlier chemoresistance (15). The latter has been suggested in relevant studies but has not yet been confirmed (16). On the contrary, the results of the EORTC and CHORUS studies have weakened this argument, although we should consider the criticism on both studies on the low rates of complete cytoreduction. In the CHORUS trial, this was approximately 30% in both primary and interval debulking surgery, while in the EORTC it was lower still (19.4%) (2, 3). The presumed minimum surgical effort in both studies was also indicated from the short operating times in both studies, which was 120 minutes in the CHORUS and 180 minutes in the EORTC trial. Hence, it is hypothesized that this might be why the non-inferiority was achieved in the NACT arm. In our previously published study, the median operating time was approximately 300 min, similar to other published studies, reflecting the higher resection rates achieved (14). On the contrary, Glover et al reported similar survival outcomes for PDS and IDS patients who required complex surgical procedures to achieve complete cytoreduction (17). The results from the TRUST study are highly anticipated. Along with determining the optimal timing for surgery, they could also contribute to clarifying the non-inferiority margin of IDS and whether it is not earlier chemoresistance that can be driven from NACT (8).
In addition, women that underwent more radical procedures and extensive peritonectomies in the pelvis or the upper abdomen were less likely to recur in these areas. Women who underwent a modified posterior exenteration with full pelvic peritonectomy tend to recur less frequently to this area than women who did not require a full pelvic clearance at the time of their surgery. The same trend is noticed in women that underwent extensive diaphragmatic peritonectomies or radical procedures in the upper abdomen compared to those that did not undergo such extensive procedures. In our previously published results, more radical procedures are more often required in the primary setting to achieve complete cytoreduction (14). That would explain why women in the NACT group more commonly experience recurrences in the pelvis or upper abdomen than the PDS group.
The latter findings raise the suspicion of whether the intraoperative estimation of disease extent is equally accurate in interval surgery compared to primary surgery, which some studies have suggested. Hynninen et al. systematically assessed 22 anatomical regions in the abdomen at the beginning of the primary or interval surgery for advanced ovarian cancer, concluding that visual estimation of tumor spread during PDS showed 98% sensitivity, 76% specificity and 95% accuracy compared to histology results. In contrast, the corresponding results in the IDS group were 86%, 76%, and 84%, respectively (p<0.001) (18). Scarring and adhesions because of NACT may interfere with the intra-operative impression of tumor extent and lead to less aggressive surgery. This is particularly evident when it comes to pelvic disease, as in the NACT group far less MPEs are performed compared to PDS (19). Scarring in the pelvis due to chemotherapy could lead to misinterpretation of whether the sigmoid and rectum need excision or not. In addition, Himoto et al. demonstrated that NACT patients suffered a relapse more commonly in the same sites with the original disease sites than the PDS group, suggesting that the absence of visible residual disease at completion of surgery is not necessarily accurate (20). In our study, 50% of completely cytoreduced NACT patients experienced a pelvic recurrence, compared to 22.4% in the PDS group. Similar findings were shown for upper abdominal/diaphragmatic or intra-abdominal relapse.
We acknowledge that our study has certain limitations, mainly selection bias due to its non-randomized nature. On the contrary, its main strength is that our population consists of consecutive patients and all our data were prospectively collected. We believe that patients undergoing NACT should undergo more aggressive surgical management, guided not only by intra-operative findings at the time of interval surgery. It could be suggested that the initial disease distribution on first CT imaging should be taken into consideration as well. We anticipate the results of further trials and studies to support our impression further.
Footnotes
Authors’ Contributions
V. Mitsopoulos: Data collection, Project development, Manuscript writing; I. Biliatis: Project development, Data collection, Manuscript writing; A. Innamaa: Data collection, Manuscript editing; J. Lippiatt: Data collection, Manuscript editing. S. Collins: Data collection.
Conflicts of Interest
The Authors declare that they have no conflicts of interest regarding this study.
- Received January 13, 2022.
- Revision received February 12, 2022.
- Accepted February 14, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
