Peritoneal metastases from rare ovarian cancer treated with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS/HIPEC)

Patient selection

A retrospective review (1994–2021) of a prospective database of >1,000 patients with peritoneal dissemination from different primaries (appendix, colon, mesothelioma, and ovarian cancer) treated with CRS/HIPEC at a single-, high-volume peritoneal surface malignancy center was performed. Patients with newly diagnosed or recurrent PM (The International Federation of Gynecology and Obstetrics [FIGO] stage III/IV) of primary peritoneal, fallopian tube, or ovarian origin with rare histology who underwent CRS/HIPEC at our institution with ≥3 months follow-up were included. All patients were classified according to the World Health Organization (WHO) and re-staged according to FIGO 2014 criteria [13]. Histopathology reports from previous biopsies or surgeries were reviewed and confirmed by CRS/HIPEC experienced, board certified pathologists. Tumors were defined as rare if recording an incidence of less than six cases per 100,000 annually [3], and included non-high grade serous (HGS) tumors, sex-cord, and germ cell tumors. Carcinosarcomas, complex and rare tumors that involve both cellular components of epithelium and mesenchyme, were excluded from this study and its outcomes were analyzed separately [15, 16]. Patients who had an aborted HIPEC were excluded from the post-HIPEC morbidity/mortality analysis.

CRS/HIPEC and data collection

CRS was performed as previously described [17] by both a gynecological and surgical oncologist. Selection criteria for surgical candidates was extrapolated from standard criteria of common indications for CRS/HIPEC. In brief, patients with an Eastern Cooperative Oncology Group (ECOG) performance status ≤2, adequate bone marrow, renal, hepatic function, and blood coagulation parameters, a sufficient support system, and in whom achieving a complete cytoreduction was feasible were deemed surgical candidates [17]. Certain imaging findings (CT scan or PET-CT) were taken into account for determining the feasibility of a complete cytoreduction. These included disease extension outside the peritoneal cavity, biliary obstruction, foreshortened mesentery, diffuse carcinomatosis on the small intestinal loops and at the diaphragmatic level, and/or liver parenchymal involvement [17]. Of note, no specific PCI cutoff was used to determine surgical eligibility.

Tumor burden, measured with the peritoneal cancer index (PCI), was assessed at exploration, with PCI≥20 considered high tumor burden [18]. Major resections, defined as surgical procedures that independently require an inpatient admission, were performed as needed to achieve a complete cytoreduction [19]. These included splenectomy, diaphragmatic peritonectomies, diaphragmatic resection, portal dissection, liver resection, gastrectomy, omentectomy, small bowel resection, colectomy, low anterior resection, hysterectomy, salpingo-oophorectomy, cystectomy, vaginectomy, and abdominal or retroperitoneal lymphadenectomy. Completeness of cytoreduction (CC) score was recorded after tumor resection as CC-0 (no residual disease), CC-1 (residual tumor <2.5 mm), CC-2 (residual tumor 2.5–25 mm), and CC-3 (residual tumor >25 mm) [20]. Considering the depth of tissue penetration by cytotoxic drugs, tumor deposits of <2.5 mm (CC-0/1) were considered complete cytoreductions [21]. A complete CRS was aimed for every patient. Following CRS, HIPEC was performed with carboplatin (800 mg/m²), cisplatin (50 mg/m²) + doxorubicin (15 mg/m²), cisplatin + paclitaxel (70 mg/m²), melphalan (50 mg/m²), or mitomycin-C (40 mg) for 90 min at a temperature of 41–43 °C. Chemotherapeutic agents were selected by consensus between gynecological, surgical, and medical oncologists, based on histology and previous chemotherapy response. In the case of an incomplete CRS, HIPEC perfusion was performed as a palliative procedure. All anastomoses were performed after perfusion. CRS/HIPEC was performed using both the open technique (coliseum), which was the first method used by our group, and the closed technique. Postoperative complications, including readmissions, were graded according to the Clavien–Dindo classification for 90 days after surgery [22]. Grade III/IV complications were considered major.

The Institutional Review Board approved this study. Consent was obtained from all patients as part of an ongoing observational cohort study. Patients were aware of the experimental nature of the treatment and counseled on the availability of other therapies.

Postoperative care and follow-up

Patients were transferred to the intensive care unit (ICU) during the first 24 h after surgery and then to the inpatient floor when clinically stable. Length of ICU and hospital stay (LOS) was recorded for each patient. Postoperative follow-up occurred at two and four weeks post-discharge, every three months for two years, and then every six months thereafter. Follow-up included physical exam, tumor markers, and CT/PET-CT. Tumor recurrence was diagnosed based on physical exam, rising tumor markers, imaging studies, biopsy results, and/or clinical presentation.

Statistical analysis

Statistical analysis was performed using SPSS V.23.0. Medians and interquartile ranges (IQR) were used to describe continuous variables not normally distributed. Categorical variables are presented as counts and proportions. Progression-free survival (PFS) was defined as the time in months from CRS/HIPEC to the date of tumor recurrence or death, whichever occurred first. Overall survival (OS) was calculated from the date of CRS/HIPEC to date of death or last follow up. PFS and OS were calculated using the Kaplan-Meier method and compared using the Log-rank test.

Patients and tumor types

In total, 44 patients with rare ovarian tumors were identified. Of these, 28 were included for analysis. We excluded six patients in which CRS/HIPEC was aborted due to extensive disease (n=1 granulosa cell, n=3 low-grade serous, n=2 mucinous), four patients who only underwent CRS, and six for other reasons (n=4 stage, n=2 amount of data available) (Figure 1). Seven histologic subtypes were identified, including clear cell (5/28, 17.9 %), endometrioid (5/28, 17.9 %), granulosa cell (3/28, 10.7 %), low-grade serous (6/28, 21.4 %), mesonephric (1/28, 3.6 %), mucinous (6/28, 21.4 %), and small cell (2/28, 7.1 %) carcinomas. The location of tumor origin was tubo-ovarian in 25 (89.3 %) patients and primary peritoneal in three (10.7 %) patients. Twenty-three (82.1 %) patients were classified as FIGO stage III and five (17.9 %) as stage IV. Median age at diagnosis was 54 (IQR: 47–57) years. Eight (28.6 %) patients had primary and 20 (71.4 %) had recurrent disease. Median time from diagnosis to CRS/HIPEC was 23.2 (IQR: 1.7–48.0) months. All patients in the primary group were treated upfront, with no interval cases. In the recurrent group, all (20/20, 100 %) patients had prior surgery and 14/20 (70.0 %) had prior chemotherapy (Table 1).

Figure 1:

Patient selection flow-chart (1994–2021)

Operative outcomes

Median age at CRS/HIPEC was 56 (IQR: 50–62) years. Median PCI score was 21 (IQR: 6–29). Median number of major resections was five (IQR: 2–7) and median estimated blood loss was 500.0 (IQR: 325.0–1,387.5) mL. Lymph nodes were submitted for pathological analysis in 27/28 (96.4 %) patients and 11 (40.7 %) of these patients had lymph node involvement. Median length of surgery was 532 (IQR: 411–596) minutes. Complete cytoreduction (CC-0/1) was achieved in 27/28 (96.4 %) patients who underwent CRS/HIPEC. One (3.5 %) patient had residual disease (CC-3) and received palliative HIPEC. HIPEC agents were carboplatin (n=9, 32.1 %), cisplatin + doxorubicin (n=3, 10.7 %), cisplatin + paclitaxel (n=1, 3.6 %), melphalan (n=9, 32.1 %), and mitomycin-C (n=6, 21.5 %) (Table 1). HIPEC was performed using the open (coliseum) method in two (7.1 %) patients and the closed technique in 26 (92.9 %) patients. In one patient, HIPEC was shortened to 45 min due to persistent hypotension.

Morbidity and mortality after CRS/HIPEC

Median length of hospital stay was nine (IQR: 8–11) days. The readmission rate was 7/28 (25.0 %). At least one complication occurred in 24 (85.7 %) patients. Grade I complications occurred in 12/28 (50.0 %) patients (anemia, thrombocytopenia, diarrhea, and leukopenia). Grade II complications, particularly anemia requiring blood transfusion (13/19, 68.4 %), were the most common type (19/28, 67.9 %). Major complications occurred in 9/28 (32.1 %) patients, with 4/9 (44.4 %) patients presenting with more than one major complication. Major complications included intraoperative ureteral injury (n=1), pyelonephritis (n=1), recurrent ascites (n=2), left upper quadrant sterile collections (n=3), pelvic abscess (n=1), wound dehiscence (n=2), pancreatic leak (n=1), enterocutaneous fistula (n=1), vaginal fistula (n=1), gastrointestinal obstruction (n=1), and septic shock (n=2). One (3.6 %) 90-day mortality occurred on postoperative day 14 due to septic shock secondary to ischemic bowel (Table 2).

Survival and recurrence

After CRS/HIPEC, 15/28 (53.6 %) patients received adjuvant systemic chemotherapy (Table 1). Carboplatin + paclitaxel was the most common regimen with a median of six cycles (range: 6–15). Median follow-up after CRS/HIPEC was 65.8 (95 % confidence interval (CI): 43.6–88.0) months, with a two-, five-, and ten-year follow-up of 85.4 , 54.1, and 19.4 %, respectively. At last follow-up, 12/28 (42.9 %) patients had disease recurrence. Sites of recurrence included intraperitoneal only in 3/12 (25.0 %) patients, extraperitoneal only in 6/12 (50.0 %) patients, and both in 3/12 (25.0 %) patients (Table 1). Median PFS was 75.6 months (95% CI: 16.2–134.7) with a two-, five-, and ten-year PFS of 68.8 , 52.6, and 43.8 %, respectively (Figure 2A). Median OS was not reached, with a two-, five-, and ten-year OS of 88.4 , 68.5, and 52.2 %, respectively (Figure 2A). No differences in OS (p=0.698) or PFS (p=0.899) were observed between primary and recurrent patients (Figure 2B).

Figure 2:

Survival outcomes after CRS/HIPEC.