Critical Analysis of Stage IV Epithelial Ovarian Cancer Patients after Treatment with Neoadjuvant Chemotherapy followed by Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy (CRS/HIPEC)

Background Cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) after neoadjuvant chemotherapy (NACT) showed promise as initial treatment for stage IIIC (SIII) epithelial ovarian cancer (EOC); however, stage IV (SIV) outcomes are rarely reported. We assessed our experience and outcomes treating newly diagnosed SIV EOC with NACT plus CRS/HIPEC compared to SIII patients. Methods Advanced EOC from 2015–2018 managed with NACT (carboplatin/paclitaxel) due to unresectable disease or poor performance status followed by interval CRS/HIPEC were reviewed. Perioperative factors were assessed. Overall survival (OS) and progression-free survival (PFS) were analyzed by stage. Results Twenty-seven FIGO stage IIIC (n = 12) and IV (n = 15) patients were reviewed. Median NACT cycles were 3 and 4, respectively. Post-NACT omental caking, ascites, and pleural effusions decreased/resolved in 91%, 91%, and 100% of SIII and 85%, 92%, and 71% of SIV. SIII/SIV median PCI was 21 and 20 obtaining 92% and 100% complete cytoreduction (≤0.25 cm), respectively. Median organ resections were 6 and 7, respectively. Grade III/IV surgical complications were 0% SIII and 23% SIV, without hospital mortality. Median time to adjuvant chemotherapy was 53 and 74 days, respectively (p=0.007). SIII OS at 1 and 2 years was 100% and 83% and 87% and 76% in SIV (p=0.269). SIII 1-year PFS was 54%; median PFS: 12 months. SIV 1- and 2- year PFS was 47% and 23%; median PFS: 12 months (p=0.944). Conclusion Outcomes in select initially diagnosed and unresectable SIV EOC are similar to SIII after NACT plus CRS/HIPEC. SIV EOC may benefit from CRS/HIPEC, and further studies should explore this treatment approach.


Introduction
Epithelial ovarian, fallopian tube, and primary peritoneal cancers, known as epithelial ovarian cancer (EOC), are heterogeneous diseases staged and treated similarly [1,2]. ese diseases account for the majority of deaths from gynecological cancers in developed countries, due to scarcity of symptoms at early stages and lack of screening methods [3]. Consequently, most EOC patients are diagnosed after peritoneal spread (International Federation of Gynecology and Obstetrics (FIGO) stage III/IV) with 5-year survival of 29% [3].
Primary debulking surgery (PDS) followed by postoperative or adjuvant systemic chemotherapy (ASC) with taxane-platinum combinations is standard for advanced EOC (AEOC) [4]. However, over the past decade, new strategies have been pursued to improve outcomes, including the use of neoadjuvant systemic chemotherapy (NACT) and, more recently, intraoperative hyperthermic intraperitoneal chemotherapy (HIPEC) [5][6][7][8][9][10]. NACT plus interval cytoreductive surgery (CRS) without HIPEC demonstrated improved perioperative outcomes but nonsuperiority in terms of time to recurrence and survivals compared to PDS [5,11,12]. Recently, NACT plus interval CRS with HIPEC showed improved survival for stage III patients, but its' role in stage IV patients, who typically have limited treatment options and high mortality, is unclear [13,14]. We assessed our experience treating newly diagnosed stage IV (SIV) EOC with NACT plus interval CRS/HIPEC and compared findings to the same treatment cohort of stage IIIC (SIII) patients.

Patients and Methods
An institutional CRS/HIPEC database was reviewed, identifying newly diagnosed AEOC patients who received NACT followed by CRS/HIPEC from 2015-2018.
is treatment approach was offered at our institution to AEOC patients deemed ineligible for the randomized clinical trial (NCT 02124421) which assesses the role of CRS/HIPEC as initial treatment in AEOC. NACT criteria included unresectable peritoneal disease due to extensive small bowel or porta hepatis involvement, biliary obstruction, or encasement of common/ external iliac vessels evidenced by imaging and/or laparoscopy, extra-abdominal disease, Eastern Cooperative Oncology Group (ECOG) performance status >2, and/or large volume of ascites or pleural effusion. Staging occurred before NACT and after CRS/HIPEC, assigning the highest staging for analysis. ree cycles of NACT with systemic taxane/platinum regimens were administered. NACT response and surgical eligibility was evaluated with imaging, tumor markers, and performance status. Interval CRS/HIPEC was offered if complete CRS (residual disease <0.25 cm) was feasible, followed by 3 cycles of taxane/platinum ASC, totaling 6 cycles.

Response Evaluation.
Treatment response was based on CT scan of chest/abdomen/pelvis pre-/post-NACT using Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 [15]. To note, the noncomplete response/nonprogressive disease (non-CR/non-PD) label is used over stable disease for patients with only nonmeasurable disease in this classification. [15] CA-125 levels were measured pre-NACT, post-NACT/pre-CRS/HIPEC, post-CRS/HIPEC, and post-ASC. Histopathologic chemotherapy response is not reported due to lack of consensus [16].

Interval Cytoreductive Surgery/Hyperthermic Intraperitoneal
Chemotherapy (CRS/HIPEC) Procedure. Post-NACT, patients were considered CRS/HIPEC candidates if there was no gross extra-abdominal disease, gross resolution of prior pleural effusion, ECOG performance status ≤2, and complete cytoreduction was feasible (<0.25 cm residual disease). Intraoperative disease burden was estimated using Peritoneal Cancer Index (PCI) scores [17]. Multiple peritoneal and visceral resections were performed to reduce tumor to microscopic levels. Completeness of cytoreduction (CC) score quantified residual tumor with CC-0 (no visible residual disease) and CC-1 (residual tumor <0.25 cm) considered complete cytoreductions [18]. Incomplete cytoreduction was defined as tumor nodules 0.25 cm-2.5 cm (CC-2) or >2.5 cm (CC-3). HIPEC agents included carboplatin 800 mg/m 2 (90 minutes, 41-43°C) or melphalan 50 mg/ m 2 (90 minutes, 41-42°C), if platinum resistance was suspected based on NACT response. Bowel anastomoses and chest tube placement were performed following perfusion. All procedures were performed together by an experienced team of gynecologic and surgical oncologists, specializing in peritoneal surface malignancies. Patients were transferred to the intensive care unit (ICU) for 24 hours and then to the inpatient oncology unit when clinically stable. Surgical complications were considered until postoperative day (POD) 90 and graded according to Clavien-Dindo classification with grade III/IV considered major [19].
2.3. Staging. AJCC 8 th edition staging was performed before NACT and at CRS/HIPEC using the highest staging for subgroup designation [20]. Stage IV disease included positive pleural effusion cytology (pM1a or FIGO stage IVA), liver or splenic parenchymal metastases, extra-abdominal metastases, including inguinal and extra-abdominal lymph nodes, and/or transmural intestinal involvement (cM1b/ pM1b or FIGO stage IVB).

Follow-Up.
Postoperative follow-up occurred 3 and 6 weeks after discharge, every 3 months for 2 years, and every 6 months thereafter, for 5 years. After ASC, CT scan of chest/ abdomen/pelvis were performed every 6 months or, as clinically indicated. After complete cytoreduction (CC-0/1), disease recurrence was defined by radiographic/pathologic evidence or disease.

Statistical
Analysis. Categorical variables were analyzed using chi-squared or Fisher's exact test and continuous variables using independent sample Student's t test or Man-nWhitney U test, when not normally distributed. Survival analysis was performed using KaplanMeier method and logrank test. Overall survival (OS) was calculated from CRS/ HIPEC to date of death. Progression-free survival (PFS) was calculated from CRS/HIPEC to date of radiographic/pathologic disease recurrence, or date of death from disease, whichever occurred first. PFS was only calculated with complete cytoreduction (CC-0/1). Median follow-up time was estimated using the reverse KaplanMeier method. All analyses were performed using STATA version 12.0 (StataCorp LLC, College Station, TX, USA) and considered statistically significant if p ≤ 0.05.
2.6. Ethics. IRB approval and preoperative consent were obtained.

Discussion
Newly diagnosed FIGO SIV ovarian cancer patients who initially present with extensive, unresectable disease, or poor performance status are typically managed with a palliative approach. However, NACTfollowed by interval CRS/HIPEC can offer therapeutic benefit for these advanced patients. In our study, SIV had similar outcomes to and a tendency for improved cytoreduction rates (CC-0: 67% vs. 93%) and survival (median OS: 25 vs. 51 months) than SIII. NACT prior to surgical debulking is considered under the premise that reducing tumor burden will improve patient performance and increase complete cytoreduction rates, resulting in lower morbidity, shorter hospital stay, and improved quality of life [10,[21][22][23]. However, no randomized trial has demonstrated superior survival of NACT plus interval debulking to primary debulking surgery (Table 4) [7][8][9][10][11]24]. Moreover, the dominant factor influencing AEOC survival is the quality of cytoreduction, with >2 cm residual disease providing no survival benefit [4,[25][26][27][28].
us, NACT plus interval debulking has been reserved only for patients who are not surgical candidates, commonly due to pleural disease, massive ascites, or extensive small bowel involvement. Interestingly, despite the controversial benefit of NACT in a broad AEOC patient cohort, pooled analysis demonstrates that NACT offers better survival in SIV disease with high tumor burden or poor performance status [5]. In our study, 90% (27/30) who underwent NACT became surgical candidates achieving 96% complete cytoreduction rate and 9-day median hospital stay with encouraging cytoreduction rates and survival outcomes experienced in 15 SIV patients compared to SIII.
Traditionally, NACT is combined with standard debulking surgery; however, the addition of HIPEC during interval debulking surgery is gaining interest [6,29,30]. HIPEC has promising results treating other cancers that commonly present with peritoneal spread [31][32][33]. HIPEC allows for direct contact of high-dose, locoregional chemotherapy, potentiated by heat increasing cytotoxicity and tissue penetration, inhibiting angiogenesis, and inducting apoptosis [34][35][36][37]. Even after complete removal of macroscopic disease, microscopic tumor cells likely remain and could contribute to early recurrence rates [38]. is may especially be true after NACT where both complete and partial response to therapy can appear as scar tissue intraoperatively [39]. erefore, the addition of HIPEC may offer improved outcomes for these patients.
Although taxane and platinum agents are commonly used with remarkable hyperthermic effects, further investigation into the optimal HIPEC agents is needed, especially in platinum/chemo-resistant disease [40,41]. e majority of our patients (88%) received carboplatin. However, those with seemingly platinum-resistant disease requiring 6 NACT cycles (n � 3) were given melphalan, an alternative HIPEC agent for aggressive and recurrent peritoneal malignancies [42,43].
Previous studies report the benefits of HIPEC for recurrent AEOC. Spiliotis et al. conducted the first prospective, randomized phase III study investigating HIPEC as an alternative treatment in recurrent stage IIIC/IV AEOC after PDS and systemic chemotherapy [44]. Patients underwent CRS/HIPEC (n � 60) or CRS alone (n � 60), both followed by ASC. PCI was ≥10 in 48% and 50% and CC-0 was achieved in 65% and 55%, by treatment group, respectively, with the HIPEC having significantly longer OS (median OS: 27 versus 13 months (p � 0.006)). Despite the fact that these results are controversial, subanalysis by stage was not performed, and our cohort relates only to initially diagnosed patients; data from this first trial suggests that HIPEC may be beneficial in recurrent SIV AEOC [45].
Van Driel et al. published the results of their randomized phase III trial in SIII EOC comparing NACTand CRS/HIPEC (n � 122) vs. CRS alone (n � 123), both followed by ASC [6]. Disease burden was measured by the number of abdominal   10 International Journal of Surgical Oncology  [44]. In our cohort, SIII and SIV intraoperative median PCI was 21 and 20, CC-0 was achieved in 67% and 93%, and median OS was 25 months and 51 months, respectively. Our population presented with high disease burden, even after NACT, and this may represent aggressive tumor biology although maximum surgical effort and high complete cytoreduction rates were achieved through combined surgical efforts of gynecologic and surgical oncologists. Regardless, SIV NACT and interval CRS/ HIPEC patients in our cohort demonstrated comparable outcomes to unresectable SIII. e absence of ostomies in our cohort highlights the collaboration between gynecologic and surgical oncologists.
is can be compared to the 72% ostomy rate in the Van Driel et al. trials' CRS/HIPEC group and among other reports ranging from 17-97% after AEOC debulking surgery [6,46,47]. We were able to avoid ostomy creation despite single and double bowel anastomosis in 48% and 33%, respectively. Surgeons should be cognizant that performing bowel resections during CRS/HIPEC does not necessitate ostomy creation. Surgical oncologists are highly skilled and experienced in performing bowel resections with anastomoses. is enhances the collaborative effort between gynecologic and surgical oncology, translating to improved surgical outcomes and patient quality of life.
In our study, overall time from CRS/HIPEC to ASC was significantly longer compared to Van Driel trial (median time of 57 days versus 33 days, respectively). Nevertheless, median length of surgery was considerably longer in our cohort (8 versus 5.6 hours) suggesting extensive resections to achieve complete cytoreductions in patients with extensive disease. Disease burden and quality of cytoreduction play a role in the delay of ASC after CRS/HIPEC due to longer recovery time with extensive procedures and multiple organ resections.
We also found that median hospital stay and time to ASC were significantly longer in SIV than in SIII. Review of cases revealed that more than half of SIV patients had a hospital stay ≥10 days (60% SIV vs. 8% SIII) with median PCI of 24 when hospital stay exceeded 10 days vs. median PCI of 19 when hospital stay was <10 days (p � 0.19). Furthermore, these SIV patients also had the longest time to ASC (>8 weeks) of which delays to adjuvant therapy were often patient driven. us, it seems that SIV patients with high PCI may require longer hospital stay and more time recovering before resuming chemotherapy.
Study limitations include its retrospective design at a single institution, small sample size, and limited follow-up. However, it includes detailed perioperative characteristics and outcomes of SIV disease, a subgroup that has shown to benefit the most from multimodal treatment approaches, such as NACT plus interval debulking surgery and warrant further studies [5]. SIV patients were excluded from the Van Driel et al. trial; nonetheless, these results and other reports are encouraging to further explore the role of NACT and interval CRS/HIPEC in SIV disease [6,14,29].

Conclusions
Promising survival outcomes, similar to stage IIIC, can be achieved for patients with initially unresectable, stage IV EOC treated with NACT and interval CRS/HIPEC. Randomized studies are needed to assess the long-term outcomes of NACT plus CRS/HIPEC in stage IV ovarian cancer. responsible for the conceptualization, methodology, and investigation. MS was involved in the project administration, investigation, data curation, formal analysis, and visualization. FLR was involved with data analysis and provided statistical assistance. VG was involved in the conceptualization and investigation. MCK investigated the study and curated the data. TPDM was responsible for the conceptualization, methodology, and investigation. CAMZ, AS, MS, VG, MCK, CN, FLR, and TPDM wrote, reviewed, and edited the original draft.