Implications of peritoneal cancer index distribution on patients undergoing cytoreductive surgery and hyperthermic intraperitoneal chemotherapy

Patient selection and data

This is a retrospective cohort study performed in a single tertiary institution. Data was retrieved from a prospectively maintained database of patients treated with CRS and HIPEC for PM between January 2001 and January 2018. Patients with PM secondary to colorectal, ovarian, and other histological subtypes (mesothelioma and primary peritoneal malignancies) were included. Patients with PMP, multicystic mesothelioma, and incomplete CRS were excluded.

This study was conducted with approval of the Centralized Institutional Review Board of Singapore Health Services. It adheres to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for observational studies.

Definitions

Intraoperatively, the PCI score was tabulated and used to quantify the extent of peritoneal disease. Based on this scoring system, the abdominal cavity was divided into 13 regions and the disease burden in each of these regions were defined by lesion size (LS) scores [1].

Based on the team’s clinical experience with the management of CRS and HIPEC cases, five critical areas of involvement were defined as: (1) the porta hepatitis, (2) caudate lobe of liver, (3) gastric antrum, (4) spleen, and (5) diaphragmatic surfaces. The involvement of these areas has been found to be associated with added surgical morbidity and may have a prognostic impact after CRS and HIPEC [12], [13], [14], [15]. As such, their involvement were recorded and analysed separately for each patient. The completeness of cytoreduction (CC) score was utilized to measure the amount of residual disease [16], with CC-0/1 considered as optimal cytoreduction.

All intra- and postoperative complications were recorded and graded based on the Clavien-Dindo classification [17], with grade three and above complications classified as major events.

Overall survival (OS) was defined as time in months between CRS and HIPEC to date of last follow-up or death. Progression free survival (PFS) was defined as the time in months from the date of CRS and HIPEC to the date of detection of recurrent disease.

CRS and HIPEC and follow-up

The CRS and HIPEC procedure performed at our institution was as previously described [18, 19] and involves the removal of all macroscopic peritoneal disease to achieve optimal cytoreduction with the subsequent administration of HIPEC. A closed technique for HIPEC was adopted with 4 L of peritoneal dialysis solution at 41–42 °C over a duration of 60 min. Mitomycin C was administered for colorectal PM while cisplatin was given for non-gastrointestinal PM. A Belmont^® hyperthermia pump was used during the study duration to deliver the intraperitoneal chemotherapy agent via a single inflow catheter and drainage was via four intra-abdominal drains.

Postoperatively, patients were transferred to the surgical intensive care unit (SICU) or high-dependency unit for monitoring. During the follow-up, patients were reviewed at three monthly intervals during which a full physical examination and markers (Ca 125, CEA and Ca19-9) were taken. A computed tomography (CT) scan of the chest-abdomen and pelvis was performed 6 monthly for the first 2 years post-CRS and HIPEC and then yearly thereafter or when clinically indicated [18]. Details of recurrences, if any, were recorded.

Statistical analysis

The relationship between overall PCI score, lesion size score distribution in each PCI region, and the involvement of any of the five critical organs with postoperative morbidity and survival outcomes were analysed. Wilcoxon rank-sum test was used if the distributions were skewed for numeric variables used. Categorical variables were evaluated using the Pearson chi-square or Fisher’s exact tests. Survival functions were estimated using Kaplan–Meier method and log-rank test was used to evaluate the differences between the two groups. Multivariate Cox proportional hazards (PH) models were built for OS and PFS, and multivariate logistic regression models were built for occurrence of high-grade complications using a forward stepwise variable selection model. PH assumption was verified based on Schoenfeld residuals. A two-sided p-value of <0.05 was taken as significant. All analyses were performed in PASW Statistics 18.0.2.

Patient and tumor characteristics

A total of 205 patients underwent CRS and HIPEC during the study duration with 152 meeting the inclusion criteria. 61 (40%) patients had PM secondary to colorectal primary, 57 (37.5%) ovarian, and 34 (22.5%) others (20 primary peritoneal, 10 mesotheliomas, and four small bowel). Due to the small numbers in ‘other’ histological subtypes, these patients were analysed collectively. Patients with missing data were excluded from the analysis. All patients and tumor characteristics are described in Table 1.

PCI and distribution

Median PCI score was nine (range 0–25). The central pelvis, represented by region six, was the most commonly involved region with the highest incidence of heavy disease burden (i.e. lesion size, LS 3, >5 cm or confluent peritoneal disease).

The lower abdomino-pelvic region, represented by regions 5–7, appeared to be more frequently involved than the mid (regions 0, 4, 8) and upper (regions 1–3) areas. The small bowel (represented by regions 9–12) when involved had predominantly low volume disease, i.e. LS 1, aggregate tumor size <0.5 cm. Figure 1 depicts the frequency of peritoneal disease in each of the 13 defined PCI regions.

Figure 1:

Pattern of PCI distribution.

When comparing colorectal, ovarian, and ‘other’ PM, there was no significant difference in the distribution of peritoneal disease in the 13 defined PCI regions (p=0.112).

Morbidity outcomes

Median CRS and HIPEC duration was 480 min (range 185–960 min) with an average blood loss of 1 L (range 100 mL to 8 L). Average length of hospitalization stay was 14 days (range 7–120 days) with five patients requiring ICU postoperatively and median ICU stay duration of one day.

Postoperative complications occurred in 31.5% (n=48) patients with 10% (n=15) suffering grade three and above major events. Only the involvement of the gastric antrum requiring gastric resection was significantly associated with the occurrence of postoperative morbidity. Diaphragmatic involvement, disease at the porta hepatitis, spleen, and caudate were not associated with increased morbidity rates.

Survival and recurrence outcomes

Overall median OS and PFS was 43 and 17 months, respectively, in all CRS and HIPEC patients. 1, 3, and 5-year OS and PFS for the various primaries are illustrated in Table 2 and Figures 2 and 3. While there was no difference seen in OS, patients with non-colorectal PM have a significantly improved PFS outcome when compared with the colorectal group (p=0.001).

Figure 2:

Overall survival (OS) of patients after CRS and HIPEC.

Figure 3:

Progression free survival (PFS) of patients after CRS and HIPEC.

The relationship between PCI distribution and survival outcomes

In all CRS and HIPEC patients, peritoneal implants in the pelvis, flanks (regions 4–8), and small bowel (regions 10–12) were significantly associated with poorer OS and PFS. Involvement of the porta hepatitis predicted poorer OS but not PFS (Figures 4 and 5).

Figure 4:

PCI distribution/lesion size (LS) score and overall survival (months).

Figure 5:

PCI distribution/lesion size (LS) score and progression free survival (months).

In the colorectal subset, in addition to involvement of the pelvis, flanks, and small bowel, peritoneal disease at the caudate, porta, and bilateral diaphragms also resulted in significant reduction in OS outcomes (p=0.007, 0.001, 0.004, respectively). In contrast, for patients with ovarian and other subtypes, the involvement of the upper abdomen ‘crucial ’organs did not result in a difference in survival nor recurrence outcomes.