Expanding the role of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy: A multicenter study on uncommon peritoneal malignancies

Abstract

BACKGROUND

Cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) has become an established treatment for selected patients with peritoneal metastases (PM) from colorectal, ovarian, and gastric cancers, as well as for certain primary peritoneal tumors such as pseudomyxoma peritonei. However, evidence supporting its role in other uncommon indications remains limited, largely due to the rarity and heterogeneity of these malignancies and the absence of standardized treatment protocols. Understanding the potential survival benefit and safety profile of CRS + HIPEC in these contexts may help refine patient selection and guide clinical decision-making.

AIM

To evaluate surgical and oncologic outcomes, particularly overall survival (OS), in this heterogeneous patient population and assess perioperative morbidity and mortality to better define the safety profile of this aggressive multimodal strategy.

METHODS

We retrospectively reviewed data from five tertiary cancer centers on patients who underwent CRS + HIPEC between January 2004 and December 2021 for PM from uncommon histologies, defined as any primary tumor other than colorectal, gastric, or ovarian carcinomas, pseudomyxoma peritonei, or malignant peritoneal mesothelioma. Baseline characteristics, operative details, complications (graded by the Clavien-Dindo classification), and survival outcomes were analyzed. OS was estimated using Kaplan-Meier analysis. Prognostic factors were evaluated using univariate and multivariate Cox proportional hazards models. The discriminatory ability and overall fit of the final model were assessed by the concordance index (C-index) and likelihood ratio test, respectively.

RESULTS

A total of 60 CRS + HIPEC procedures were performed in 60 patients (mean age = 58.5 years, 78.3% female). The most frequent primary tumors were uterine (35%) and breast cancer (20%). Median operative time was 405 minutes, and 75% of patients required perioperative transfusions. Major complications (Clavien-Dindo ≥ 3) occurred in 21.6% of patients, and 90-day mortality was 1.6%. Median OS for the entire cohort was 28 months. Kaplan-Meier analysis showed that breast cancer patients achieved the longest survival (median OS = 75 months) compared with uterine cancer (32 months) and other primaries (17 months). Multivariate analysis confirmed tumor origin as the strongest independent predictor of OS (C-index = 0.81; likelihood ratio test = 40.07; P < 0.001).

CONCLUSION

Our findings suggest that CRS + HIPEC can be performed safely in highly selected patients with PM from uncommon primary tumors, achieving meaningful long-term survival in subsets such as breast and uterine cancers. Tumor biology, rather than clinical factors, emerged as the key determinant of survival. Given the rarity and heterogeneity of these malignancies, collaborative multicenter efforts and prospective registries are essential to establish standardized selection criteria and optimize outcomes.

Key Words: Cytoreductive surgery; Hyperthermic intraperitoneal chemotherapy; Uncommon peritoneal malignancies; Peritoneal surface malignancy; Oncologic outcomes

Core Tip: This multicenter retrospective study evaluates the safety and efficacy of cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) for peritoneal metastases (PM) from uncommon primary tumors. Among 60 patients treated across five high-volume centers, long-term survival was observed in selected cases, particularly those with breast or uterine cancer. Tumor origin emerged as the strongest independent predictor of overall survival, whereas timing of PM (synchronous vs metachronous) showed a variable prognostic impact. These findings support CRS + HIPEC as a potentially valuable option in rare indications when patients are carefully selected.

INTRODUCTION

Cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) has evolved considerably in recent years. Both randomized controlled trials and non-randomized clinical studies have demonstrated its efficacy in carefully selected patients with peritoneal metastases (PM) from colorectal, ovarian, and gastric cancers, as well as in rare primary peritoneal malignancies such as pseudomyxoma peritonei[1,2]. Given the survival benefit observed in these tumor types, it is important to investigate whether this approach could also be advantageous for carefully selected patients with PM from other primary sites. Here, we present our multicenter experience with CRS and HIPEC performed for uncommon indications—encompassing both rare abdominal malignancies and more common extra-abdominal tumors such as breast cancer—and analyze the associated surgical and oncologic outcomes[3,4]. The primary objective of this study is to evaluate surgical outcomes and overall survival (OS) in this patient population, which would otherwise typically be managed with palliative intent, and to assess whether CRS and HIPEC can be considered both safe and potentially effective in this setting.

MATERIALS AND METHODS

This was a retrospective multicenter study conducted at five tertiary care comprehensive cancer centers. Patients were eligible for inclusion if they underwent CRS and HIPEC for PM arising from uncommon histologies (UH), defined as tumors other than colorectal, gastric or epithelial ovarian carcinomas, as well as pseudomyxoma peritonei, and extra-abdominal tumors characterized by a marked propensity for peritoneal spread, such as breast cancer.

All patients treated between January 2004 and December 2021 were screened for eligibility. Inclusion criteria included age between 18 and 75 years, an Eastern Cooperative Oncology Group Performance Status of 0-2, and histologically confirmed PM (both synchronous and metachronous)[5,6]. Written informed consent was obtained from all patients.

Exclusion criteria were the presence of extraperitoneal disease (except resectable liver metastases not requiring vascular or biliary reconstruction), the presence of another active malignancy, unresectable disease at exploration, and severe comorbid conditions. All patients were evaluated and approved for surgery following multidisciplinary team discussion.

Intraoperatively, peritoneal disease was assessed using the Peritoneal Cancer Index (PCI), as described by Sugarbaker (range: 0-39)[7]. The goal of CRS was complete cytoreduction, defined as no macroscopic residual disease (CC-0) or residual nodules ≤ 2.5 mm (CC-1). Incomplete cytoreduction was defined as residual disease > 2.5 mm (CC-2 or CC-3)[3]. No PCI threshold was applied for patient eligibility. Only patients who achieved complete cytoreduction were considered for HIPEC (and therefore included in the study). Completeness of cytoreduction was collected as a dichotomous variable (complete vs incomplete), and for this reason, it was not possible to perform a further analysis comparing CC-0 vs CC-1 cases.

Following cytoreduction, HIPEC was administered according to the institutional protocol of each participating center. Postoperative complications occurring within 90 days of surgery were recorded and graded using the Clavien-Dindo classification[8]. Operative mortality was defined as any death occurring within the same 90-day period. Additional variables collected included body mass index, length of hospital stay (LOS), perioperative blood transfusion requirements, and planned intensive care unit (ICU) admission.

When deemed appropriate, systemic chemotherapy—administered in the neoadjuvant, adjuvant, or combined perioperative setting—was prescribed according to the recommendations of the local multidisciplinary oncology team. Follow-up was conducted in accordance with international guidelines specific to each primary tumor. Recurrence was diagnosed based on clinical, radiologic, or histopathologic evidence, and all cases were reviewed during multidisciplinary tumor board meetings.

To investigate the prognostic role of tumor origin, survival analyses were pre-specified according to two complementary approaches. First, each histologic subtype represented by more than two cases was considered individually, while less frequent tumors were grouped together as “other”. Second, patients were stratified into three broader categories: Uterine cancer (the most represented histology), breast cancer (the second most represented histology), and all other tumors combined. This classification was applied to allow both granular and simplified assessments of tumor-specific prognostic effects.

Statistical analysis

Descriptive statistics were reported as means ± SD or medians (range) for continuous variables, and as n (%) for categorical variables.

OS was defined as the time from the first HIPEC procedure to the date of death, last follow-up, or study cut-off date, whichever occurred first. Perioperative deaths were included in the survival analysis. Patients who died without evidence of recurrence were censored at the date of death.

Median survival times and survival probabilities were estimated using the Kaplan-Meier method. Hazard ratios (HRs) and corresponding P values were calculated using the Cox proportional hazards model.

All statistical analyses were performed using standard statistical packages. Part of the data processing and graphical outputs were carried out using open-source Python libraries on the Google Colab platform (Google LLC, Mountain View, CA, United States).

RESULTS

A total of 60 CRS and HIPEC procedures for PM from UH were performed across the five participating centers. The mean age of the cohort was 58.5 years, and 78.3% of patients were female. The distribution of primary tumor sites is reported in Table 1, with uterine neoplasms being the most common (35%).

Table 1 Patient characteristics.

	n (%)	PCI median
Females	47 (78.3)
BMI (kg/m2; median)	26
Synchronous PM	26 (43.3)
Neoadjuvant chemotherapy	18 (69.2)
Adjuvant chemotherapy	22 (84.6)
Both	17 (65.4)
Metachronous PM	34 (56.7)
Neoadjuvant chemotherapy	8 (23.5)
Adjuvant chemotherapy	24 (70.6)
Both	3 (8.8)
ASA
1	17 (28.3)
2	34 (56.7)
3	9 (15)
PCI (median)	13
Primary origin of neoplasia
Uterus	21 (35)	13
Breast	8 (13.4)	19
Sarcoma	5 (8.4)	16
Small bowel	6 (10)	14
Liver	4 (6.6)	9
Pancreas	4 (6.6)	8
Biliary tract	3 (5)	2
MPM	3 (5)	12
Miscellaneous1	6 (10)	7.5

¹Miscellaneous includes: Ovarian carcinosarcoma; gastrointestinal stromal tumor (previously treated with TKI, resistant); 2 jejunal NET; desmoplastic small round cell tumor; malignant Pheochromocytoma.

BMI: Body mass index; PM: Peritoneal metastases; PCI: Peritoneal Cancer Index; MPM: Malignant peritoneal mesothelioma.

The median operative time was 405 minutes (range: 180-866). Ten patients (16.6%) underwent total peritonectomy, 40 (66.6%) at least one diaphragmatic peritonectomy, and 47 (68.3%) pelvic peritonectomy. Eighteen patients underwent peritonectomy alone, whereas 42 patients required at least one associated visceral resection (excluding omentectomy and appendectomy). The most frequent combined procedures included colonic resection (n = 27), ileal resection (n = 22), splenectomy (n = 11), hepatic resection (n = 5), hysterectomy with bilateral salpingo-oophorectomy (n = 12, performed in only 7 cases for uterine neoplasms with synchronous PM, while in the remaining cases it was carried out for direct involvement of the uterus and adnexa), and, less commonly, orchiectomy, prosthesis removal, ureteral resection, adrenalectomy, gastric resection, distal pancreatectomy, or nephrectomy (n = 7 in total).

Forty-five patients (75.0%) required perioperative blood transfusions (median, 2 units), and 47 (78.3%) were admitted to the ICU postoperatively. Postoperative complications, classified according to the Clavien-Dindo system, are summarized in Table 2 together with the LOS. One patient died on postoperative day 13 due to complications of hemangiopericytoma. Four patients required urgent reoperation (two for postoperative hemorrhage, one for eventration, and one for abdominal compartment syndrome). Additional interventions included pleural drainage for effusion in two patients (one of whom also underwent reoperation) and percutaneous drainage for intra-abdominal collections in five patients. Other notable complications included one case of acute renal failure and one transient ischemic attack.

Table 2 Post-operative complication according to Clavien-Dindo classification.

Clavien-Dindo	n (%)	LOS day median	LOS day median
0	11 (18.4)	9	14
1	4 (6.6)	10
2	31 (51.7)	16
31	10 (16.7)	23	22
42	3 (5)	15
5	1 (1.6)	13

¹3A = 6 patients; 3B = 4 patients.

²All 4A.

LOS: Length of hospital stay.

Complications were further analyzed according to the type of surgical procedure. Patients were divided into two groups: Group A (peritonectomy alone, n = 18) and group B (peritonectomy plus at least one associated procedure, n = 42), as defined by the additional resections detailed above. Minor complications (Clavien-Dindo grade 0-2) occurred in 16 patients (88.9%) in group A and 30 patients (71.4%) in group B. Major complications (grade ≥ 3) occurred in 2 patients (11.1%) in group A and 11 patients (26.2%) in group B, with one postoperative death in the latter group. The risk of major complications was more than twofold higher in group B compared with group A [risk ratio = 2.36, 95% confidence interval (95%CI): 0.58-9.58]; however, the difference did not reach statistical significance (Fisher’s exact test, P = 0.31), likely due to the limited sample size.

The distribution of CRS and HIPEC procedures across three consecutive time periods is shown in Figure 1, illustrating an increasing trend in the number of procedures performed over time.

Figure 1 Distribution of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy procedures performed for peritoneal metastases from uncommon histologies across three consecutive periods (2004-2010, 2011-2016, 2017-2021), stratified by tumor origin. Female neoplasms (breast and uterine cancer) were more common in earlier years, whereas procedures for other primary tumors increased over time. PM: Peritoneal metastases; UH: Uncommon histologies.

OS analysis was performed in 51 patients, excluding those with less than six months of follow-up at the data cut-off (December 31, 2021). The Kaplan-Meier survival curve for the entire cohort (Figure 2A) showed a steep decline during the first 12-24 months, with survival decreasing from nearly 100% to approximately 60%-65% at 12 months and to 45%-50% at 24-30 months. Beyond 30 months, the slope of the curve flattened, indicating a relative stabilization of mortality among long-term survivors. At 60 months, the estimated OS probability was approximately 30%-35%, suggesting that nearly one-third of patients may derive durable benefit from this aggressive multimodal strategy.

Figure 2 Kaplan-Meier curve showing overall survival. A: The entire cohort of patients undergoing cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) for peritoneal metastases treated under uncommon indications. The median overall survival was 28 months. Shaded areas represent 95% confidence intervals; B: Stratified by tumor origin in patients treated with CRS and HIPEC for peritoneal metastases under uncommon indications. Only patients with ≥ 6 months of follow-up were included; C: Stratified by tumor group: Breast cancer, uterine cancer, and all other uncommon indications. Patients with breast cancer showed the most favorable survival, followed by those with uterine cancer, while patients with other tumor types experienced the poorest outcomes. HCC: Hepatocellular carcinoma; FU: Follow-up.

Survival outcomes differed across histologic subtypes (Figure 2B) and across the three broader categories of uterine, breast, and other tumors (Figure 2C).

Patients with PM from breast cancer (PMBC) had the most favorable outcomes, with a median OS of 75 months (95%CI: 28-not reached), significantly longer than that of uterine cancer (32 months; 95%CI: 21-51; P = 0.015). Both groups had longer OS compared with patients with other primary tumors (median = 17 months; 95%CI: 15-26; P = 0.102). Kaplan-Meier curves stratified by histology are shown in Figure 2B. Breast cancer patients experienced the most favorable survival, with rates exceeding 80% at 40 months and remaining above 60% at 60 months. Sarcoma and mesothelioma showed intermediate outcomes, whereas uterine cancer was associated with a steady decline, with survival falling below 25% by 60 months. Patients with cholangiocarcinoma, pancreatic cancer, and HCC had the poorest prognoses, with most deaths occurring within the first 12-24 months.

As shown in the Kaplan-Meier analysis in Figure 2C, pairwise log-rank tests demonstrated that patients with breast cancer had significantly longer survival compared with both uterine cancer (χ² = 6.29, P = 0.012) and the other primaries (χ² = 8.76, P = 0.003). In contrast, the survival difference between uterine cancer and the other group did not reach statistical significance (χ² = 2.39, P = 0.12).

Multivariate Cox regression including 59 patients confirmed that primary tumor site was the strongest independent predictor of OS. Compared with breast cancer, significantly higher mortality risks were observed for small bowel adenocarcinoma (SBA; HR = 25.21; 95%CI: 3.53-180.14; P < 0.005), cholangiocarcinoma (HR = 32.53; 95%CI: 4.22-250.70; P < 0.005), pancreatic cancer (HR = 24.34; 95%CI: 1.62-365.44; P = 0.021), and uterine cancer (HR = 6.71; 95%CI: 1.72-26.11; P = 0.006; Table 3). Other variables, including age, sex, American Society of Anesthesiologists (ASA) score, PCI, and the use of neoadjuvant or adjuvant chemotherapy, were not independently associated with OS. However, both neoadjuvant (HR = 0.42; P = 0.14) and adjuvant chemotherapy (HR = 0.65; P = 0.28) showed HRs suggestive of a potential protective effect, although not statistically significant. The final model demonstrated good discriminatory ability (concordance index = 0.81) and an overall significant fit (likelihood ratio test = 40.07 on 14 degrees of freedom; P < 0.001).

Table 3 Multivariable Cox regression by primary tumor site.

Primary tumor site	HR	95%CI	P value
Breast cancer (reference)	1.00	-	-
Small bowel	25.21	3.53-180.14	< 0.005
Biliary tract	32.53	4.22-250.70	< 0.005
Pancreas	24.34	1.62-365.44	0.021
Uterus	6.71	1.72-26.11	0.006
Sarcoma	2.38	0.46-12.34	0.30
MPM	1.33	0.30-5.83	0.70
Miscellaneous	1.05	0.32-3.38	0.94

Miscellaneous includes: Ovarian carcinosarcoma; gastrointestinal stromal tumor (previously treated with TKI, resistant); 2 jejunal NET; desmoplastic small round cell tumor; malignant Pheochromocytoma. HR: Hazard ratio; 95%CI: 95% confidence interval; MPM: Malignant peritoneal mesothelioma.

Analysis comparing synchronous and metachronous PM revealed a statistically significant survival advantage in favor of metachronous presentation (log-rank P = 0.015). However, this result was strongly influenced by the inclusion of breast cancer cases, all of which were metachronous and demonstrated excellent prognosis, with no observed deaths during follow-up. When breast cancer was excluded from the analysis, the survival difference between metachronous and synchronous disease was no longer statistically significant (log-rank P = 0.167), although a trend toward improved outcomes persisted in the metachronous group. Stratified analyses by tumor origin confirmed this pattern: Among patients with uterine cancer, metachronous spread was visually associated with longer survival, but the difference did not reach statistical significance (log-rank P = 0.317; Figure 3). In contrast, patients with other tumor types experienced generally poorer outcomes overall, with minimal impact from timing of disease presentation (log-rank P = 0.528).

Figure 3 Overall survival stratified by tumor group (uterus vs other) and timing of peritoneal metastases (synchronous vs metachronous), excluding breast cancer. A survival trend favoring metachronous disease was observed in uterine tumors. OS: Overall survival; Sync: Synchronous; Meta: Metachronous.

In the multivariate Cox model including the entire study population, neoadjuvant chemotherapy demonstrated a borderline protective effect (HR = 0.09, 95%CI: 0.006-1.43, P = 0.088), although not statistically significant. In subgroup analyses, patients with synchronous PM receiving neoadjuvant chemotherapy showed a longer median OS compared with those treated upfront (28.0 months vs 16.0 months), but the difference did not reach statistical significance (log-rank χ² = 2.66, P = 0.103). In the metachronous cohort, median OS was comparable (26.0 months vs 28.0 months; log-rank χ² = 0.25, P = 0.616; Figure 4).

Figure 4 Overall survival by neoadjuvant chemotherapy in patients with peritoneal metastases. A: In synchronous peritoneal metastases (PM), median overall survival (OS) was longer with neoadjuvant treatment (28.0 months vs 16.0 months), though not statistically significant; B: In metachronous PM, OS was similar regardless of neoadjuvant therapy. PM: Peritoneal metastases; OS: Overall survival; CHT: Chemotherapy.

DISCUSSION

Since its introduction, the combination of CRS and HIPEC has undergone progressive refinement, with evolving selection criteria aimed at optimizing patient safety and oncologic outcomes[4,9]. Despite the growing body of evidence supporting its role in specific indications, treatment of peritoneal surface malignancies (PSM) remains confined to specialized centers[3].

This study represents a multicenter experience spanning 17 years across five high-volume institutions dedicated to PSM management. Over time, an increasing number of cases with peritoneal spread from tumors of UH have been considered for CRS and HIPEC. Notably, more than half of the procedures in our cohort were performed within the last five years, reflecting this evolving trend.

A temporal shift in the types of primary tumors was also observed. In the first 12 years, female-specific malignancies (uterine and breast cancer) represented 75.9% of cases, whereas in the last four years, their proportion declined to 22.5%. This change likely reflects both broader indications and increased awareness of potential peritoneal tropism in other tumor types.

In the absence of established guidelines for UH, patient selection criteria were extrapolated from those validated for colorectal cancer and mesothelioma: Age < 76 years, good performance status, low comorbidity burden, absence of extraperitoneal disease, high likelihood of achieving complete cytoreduction, and either chemo-responsive or indolent disease[10,11]. The relatively young age (mean age = 58.5 years) and good preoperative status (85% ASA I-II) in our cohort likely contributed to the decision to offer CRS and HIPEC.

Our findings suggest a potential survival benefit for patients with metachronous PM, particularly in selected tumor types such as uterine cancer. However, this advantage was no longer statistically significant when breast cancer cases-characterized by uniformly metachronous presentation and excellent prognosis-were excluded. While metachronous disease may reflect more favorable tumor biology or improved selection, its prognostic value appears to be context-dependent and should be interpreted with caution in heterogeneous cohorts.

Neoadjuvant therapy also played a role in patient selection, allowing exclusion of those with disease progression during treatment. All patients were evaluated within dedicated multidisciplinary tumor boards, which remains essential in guiding indication and timing.

Complete cytoreduction was a prerequisite for proceeding with HIPEC in all cases. Due to the rarity and heterogeneity of diseases included, our data do not allow for meaningful PCI cut-off values to be defined for these indications.

In the absence of standardized HIPEC protocols for these rare entities, each participating center adopted its own institutional regimen[3].

Despite the complexity of the surgical procedures the perioperative outcomes were acceptable. Grade 0-2 complications occurred in 76.6% of patients, with a reoperation rate of 6.6%, and 90-day mortality was limited to 1.6%[9]. Analysis of postoperative complications according to the extent of surgical procedures suggests that peritonectomy alone is not associated with a significant increase in morbidity. In our series, patients undergoing isolated peritonectomy (group A) experienced mostly minor complications, with only two major events (11.1%) and no perioperative deaths.

The median OS of the cohort was 28 months, suggesting that CRS combined with HIPEC may provide a survival benefit even for these uncommon indications. Most patients also received perioperative systemic chemotherapy, which may have further contributed to the favorable outcomes observed[9].

Our survival analysis highlights that the primary tumor type is the strongest independent predictor of OS in this cohort. Patients with breast and uterine cancer derived the most substantial long-term benefit from CRS and HIPEC, whereas those with pancreatic, biliary, or small bowel primaries experienced markedly reduced survival despite aggressive local-regional treatment. These findings suggest that tumor biology exerts a dominant influence over postoperative outcomes and long-term prognosis. In contrast, clinical variables such as age, ASA score, and PCI did not demonstrate independent prognostic significance, likely due to the limited sample size and potential collinearity with tumor type. Collectively, these results reinforce the importance of careful patient selection based primarily on tumor origin and emphasize the need for multicenter collaborative studies to refine selection criteria for CRS and HIPEC in uncommon PSM.

However, the limited number of patients enrolled in our study, together with the heterogeneity of the primary tumors represented, precluded statistically robust analyses of the behavior of PSM according to the site of origin. This limitation is in line with previous reports that have underscored the challenges of drawing definitive conclusions from small, heterogeneous cohorts[12]. Therefore, while we have discussed our findings in light of these constraints, in the following section we also provide a concise review of the current literature on the management of PSM arising from the most prevalent primary tumors observed in our series. This aims to contextualize our results within the broader body of evidence and to highlight potential directions for future research.

Endometrial cancer (EC) is the most common malignancy of the female reproductive tract, with an estimated 417000 new cases annually worldwide (approximately 4.5% of all cancers)[13,14]. While early-stage EC generally has a favorable prognosis, advanced or recurrent intra-abdominal disease is associated with poor outcomes, with a median OS of around 12 months and a 5-year survival rate below 20%[15].

In selected patients with PM from EC, CRS with or without HIPEC has been investigated as a strategy to improve outcomes. A systematic review by Tempfer et al[16] included eight studies with 68 patients (mean age = 57 years, mean PCI = 16.7, mean interval = 22 months from primary treatment), achieving CC0 in approximately 70% of cases. HIPEC regimens varied but most commonly used cisplatin alone (approximately 57%) or combined with agents such as doxorubicin, paclitaxel, or mitomycin C (approximately 43%). About 73% of patients also received systemic chemotherapy. Median disease-free survival (DFS) and OS ranged from 7-18 months and 12-33 months, respectively.

Subgroup analyses suggested improved outcomes in patients without preoperative chemotherapy, with limited peritoneal disease, treated with cisplatin-based HIPEC, and without lymphatic spread[17]. These findings support CRS ± HIPEC as a potential option for highly selected patients, although prospective comparative trials remain necessary to validate its benefit and refine selection criteria[17].

Breast cancer is one of the most prevalent malignancies in Western countries and the most common extra-abdominal source of PM (PMBC)[18,19]. However, clinical guidelines for PMBC are lacking, and its true incidence is difficult to determine. The association with invasive lobular carcinoma is well recognized, but prospective data remain scarce[20]. A recent multicenter Italian study of 49 PMBC patients reported that 20 underwent CRS, 13 with HIPEC[21]. In the curative-intent group, median OS was 61.5 months, significantly longer than in patients treated with non-curative procedures (36month OS 89.2% vs 6%; P < 0.001). Among CRS patients, 13 experienced recurrence (peritoneal or extraperitoneal) at a median of 54 months. Multivariate analysis identified younger age, hormone receptor-positive subtype, absence of extra-PM, longer interval from primary breast cancer treatment (> 40 months), HER2targeted therapy eligibility, and isolated peritoneal disease as favorable prognostic factors. These findings suggest that curativeintent CRS ± HIPEC may be beneficial for carefully selected PMBC patients—particularly those with luminal A subtype, good performance status, long diseasefree interval, and no systemic spread—and should ideally be considered within a multidisciplinary framework and prospective trials.

Soft tissue sarcomas are rare mesenchymal tumors, about one-third of which arise in the abdomen or retroperitoneum and can spread hematogenously or via peritoneal seeding. Peritoneal sarcomatosis (PS) is associated with poor prognosis, occurring in up to 10% of retroperitoneal sarcomas and with median OS typically 6-15 months. Conventional therapies offer limited benefit[22-24]. In a 2005 randomized trial, Bonvalot et al[25] randomized 38 PS patients to CRS alone or CRS plus early postoperative intraperitoneal chemotherapy (EPIC, not HIPEC); 5year OS was approximately 40% in both arms, with benefit largely attributed to surgery. A 2022 meta-analysis by Wong et al[26] including 320 PS patients treated with CRS + HIPEC reported a pooled median DFS of 12 months and median OS of 29.3 months (95%CI: 23.8-34.8), with 5year OS of 35.3%. Complete cytoreduction (CC0) was associated with improved OS (34.6 months). Tumor burden also impacted outcomes: Berthet et al[27] reported 5year OS of 75% for PCI < 13 vs 12.8% for PCI ≥ 13, and Sugarbaker[28] described OS > 35 months in selected cases. Additional studies confirmed favorable outcomes with bidirectional intraoperative chemotherapy (IV ifosfamide plus HIPEC doxorubicin), showing OS rates of 93%, 89%, 81%, and 73% at 6 months, 12 months, 24 months, and 48 months, respectively[29]. Histologic grade also correlated with prognosis[30,31]. By contrast, a pilot study by Seo et al[32] evaluating prophylactic CRS + HIPEC after RPS resection failed to demonstrate benefit in DFS or PS prevention. Overall, current evidence does not support routine HIPEC for PS of any specific histologic subtype. Instead, HIPEC may be considered as an adjunct to complete cytoreduction in highly selected patients responsive to induction chemotherapy, ideally within experienced centers or clinical trials[26].

Biliary tract tumors account for approximately 1% of all cancers and 10%-15% of primary hepatic malignancies[33]. Peritoneal dissemination occurs in 10%-20% of cholangiocarcinoma cases at diagnosis[34,35]. For unresectable or metastatic disease, standard treatment remains palliative chemotherapy, typically gemcitabine plus cisplatin, with median OS around 15 months[34]. The role of CRS + HIPEC has been explored only in small retrospective series. In 2018, Amblard et al[36] reported on 34 patients treated with CRS + HIPEC compared to 21 receiving chemotherapy alone; median OS was 21.4 months vs 9.3 months, and 3year OS 30% vs 10%, respectively. In 2021, Feng et al[37] similarly found improved survival for CRS + HIPEC compared with CRS alone (median OS 25.5 months vs 11.2 months, P < 0.001). Although these results suggest a potential benefit of adding HIPEC to aggressive locoregional treatment, the evidence is limited by the retrospective design, small sample sizes, and selection bias. Prospective studies are required to validate these findings and better define the patients who may benefit from CRS + HIPEC in biliary tract malignancies.

Hepatocellular carcinoma (HCC) is one of the most common cancers in Asia, with an incidence of 466.1 and a mortality of 422.1 per 100000 in China alone[37]. HCC typically metastasizes to the lungs, bones, lymph nodes, and adrenal glands; PM (HCCPM) are uncommon, occurring in 2%-18% of cases, often due to tumor rupture, which carries up to 50% mortality[38]. Firstline treatment for advanced unresectable or metastatic HCC consists of immunotherapy combined with antiangiogenic agents. Cytotoxic chemotherapy is rarely used due to poor tolerability and limited benefit but may be considered in selected patients who fail standard therapies and maintain good liver function[39]. No standardized approach exists for HCCPM, and median OS with palliative therapy is typically 6-14 months[40]. Lin et al[41] reported significantly longer OS in patients undergoing surgery (12.5 months vs 2.1 months; P = 0.0013), with metachronous PM showing better outcomes than synchronous disease (8.6 months vs 3.8 months; P = 0.0117). Small series have suggested a possible role for CRS ± HIPEC. Tabrizian et al[42] treated 14 patients (7 with HIPEC), achieving CC0 resection in all but one; median OS was 35.6 months for CC-0/1 patients and 42.1 months in the HIPEC subgroup[6]. Berger et al[43] similarly reported median OS of 23.6 months, with a nonsignificant trend toward better outcomes with HIPEC (29.7 months vs 19.5 months). The largest series, by Mehta et al[40], included 21 patients treated with CRS + HIPEC across 10 Peritoneal Surface Oncology Group International (PSOGI) centers, reporting a median PCI of 14, CC0/1 in 16 patients, median OS of 46.7 months, and recurrence-free survival of 26.3 months; CC2/3 patients had a median OS of only 5.9 months. Conversely, a 2020 study on 57 patients with spontaneous HCC rupture found no OS or DFS benefit from adding HIPEC to hepatectomy, but noted longer hospital stays, higher morbidity, and increased costs[44]. Overall, CRS ± HIPEC may benefit highly selected HCCPM patients with limited disease and good performance status, but evidence remains limited. Given the rarity and complexity of these cases, treatment should be restricted to specialized centers within clinical trials or collaborative registries.

Pancreatic ductal adenocarcinoma (PDAC) shows early systemic spread, with 50%-55% of patients presenting with metastases at diagnosis[45]. PM are often underdiagnosed; in a study of > 1000 staging laparoscopies, 12% had PM despite negative imaging, and 42% of patients with positive cytology had no macroscopic disease[46]. The peritoneum is also a frequent site of recurrence after curative resection, especially for tumors of the pancreatic body and tail, and represents the first relapse site in approximately 16.6% of cases[46,47]. Historically, treatment has relied on palliative chemotherapy. Singleagent gemcitabine yielded median OS < 6 months, whereas multiagent regimens such as FOLFIRINOX improved survival to approximately 13 months in patients with PM[48]. Regional strategies including CRS + HIPEC have shown modest benefits, with a review by Frassini et al[49] reporting median OS approximately 17 months. Surgical morbidity is acceptable, but combining HIPEC with pancreatic resection increases complication rates, with distal pancreatectomy + HIPEC associated with pancreatic fistula rates of 33%-43% vs 17% for pancreatectomy alone[50]; adding CRS further increases major complications from 19% to 41%[51]. Recent data suggest potential benefit in selected patients. A 2023 study in 61 PDAC patients with isolated PM who responded to ≥ 6 months of systemic therapy showed median OS 41 months with CRS + HIPEC vs 19 months with chemotherapy alone (P = 0.002); 1, 2, and 3year OS rates were 91%, 66%, and 59%, respectively[52]. Yan et al[53] similarly reported median OS 24.2 months in 10 patients (median PCI 19) treated with CRS + HIPEC. An ongoing phase II trial (NCT04858009) is evaluating HIPEC in PDAC with PM. Preventive strategies are also under investigation: Sugarbaker and Stuart[54] combined HIPEC with postoperative normothermic intraperitoneal chemotherapy in eight patients without evident PM, reporting median OS 29 months and no peritoneal recurrences.

SBA is a rare malignancy, accounting for < 5% of gastrointestinal cancers[55]. About onethird of patients present with metastatic disease, most commonly in the liver, peritoneum, and distant lymph nodes[56]. Prognosis is poor, with 5year OS < 20% and median OS 9-16 months with systemic chemotherapy; supportive care alone yields a median OS of approximately 2.5 months[57,58]. The researcher van Oudheusden et al[59] evaluated 16 patients with SBAPM treated with CRS + HIPEC, reporting a median OS of 31 months, suggesting potential benefit in selected cases. A 2017 study similarly found that CRS + HIPEC may improve outcomes compared with systemic therapy alone, although fluoropyrimidineoxaliplatin regimens remain standard[60]. A review by Chen et al[58] of 164 SBAPM patients treated with CRS + HIPEC reported median OS 9-32 months and 5year OS 25%-38%. Favorable prognostic factors included welldifferentiated tumors, absence of nodal metastases, PCI ≤ 15, complete cytoreduction (CC0/1), age < 70 years, and treatment within 6 months of diagnosis, whereas duodenal primaries and nodal disease were associated with poorer outcomes. Overall, CRS + HIPEC may provide prolonged survival in a highly selected subset of SBAPM patients, but prospective evidence remains limited, and treatment should be restricted to expert centers.

Malignant peritoneal mesothelioma, is an aggressive neoplasm with epithelioid, sarcomatoid, and biphasic subtypes; multicystic mesothelioma, first described by Mennemeyer and Smith[61] in 1979, represents approximately 6% of all peritoneal mesotheliomas, with an estimated incidence of 1 per 1000000 based on PSOGI registry data (1989-2009)[62]. Long-term outcomes are favorable in lowgrade malignant peritoneal mesothelioma and multicystic mesothelioma, with 10year OS approaching 70%-100%, yet recurrence is common, affecting approximately 50% of patients within 10 years of resection[63]. Complete CRS (CC0) is the mainstay of treatment, but the benefit of HIPEC remains unclear. Some centers advocate CRS + HIPEC in recurrent disease or when histology is borderline or ambiguous, to reduce relapse risk. Given the rarity of multicystic mesothelioma, highquality comparative studies are lacking, and management should be individualized in specialized peritoneal malignancy centers, considering histology, recurrence risk, and surgical feasibility[63].

CONCLUSION

CRS combined with HIPEC should not be dismissed a priori for the management of PM from rare or otherwise uncommon primary tumors. In highly selected patients treated at experienced peritoneal malignancy centers, this multimodal strategy has been shown to be both safe and potentially capable of achieving superior palliation and meaningful survival benefits compared with conventional palliative approaches. Our survival analysis underscores that the primary tumor type is the most powerful independent predictor of OS. Patients with breast and uterine cancers derived the greatest long-term benefit from CRS + HIPEC, whereas those with pancreatic, biliary, or small bowel primaries experienced significantly poorer outcomes despite aggressive locoregional treatment. These findings highlight the dominant role of tumor biology in shaping postoperative prognosis and reinforce the need for stringent patient selection primarily based on tumor origin. Clinical variables such as age, ASA score, and PCI did not retain independent prognostic significance, likely due to the limited sample size and collinearity with tumor type. Neoadjuvant chemotherapy showed only a borderline protective effect, with a possible survival advantage in synchronous disease that did not reach statistical significance. Metachronous presentation also suggested a survival benefit, particularly in uterine cancer, but this effect was not consistent across tumor types. Given the rarity and heterogeneity of these malignancies, individual centers are unlikely to accrue sufficient patient numbers to define standardized selection criteria or treatment protocols. Collaborative initiatives—such as national disease-specific protocols and centralized prospective databases, similar to the French RENAPE network for rare peritoneal tumors—are essential to improve patient selection, ensure treatment consistency, and enable robust outcome analyses[64]. Based on current evidence, certain tumor types—particularly intrahepatic and extrahepatic biliary tract cancers and pancreatic adenocarcinoma with synchronous PM—appear to have limited responsiveness to CRS + HIPEC and unfavorable risk-benefit profiles, and may therefore not represent optimal candidates for this approach. Multicenter collaborative studies are warranted to refine patient selection and to further elucidate the role of CRS + HIPEC in the management of uncommon PSM.