Hyperthermic intrathoracic chemotherapy after cytoreductive surgery in malignant pleural mesothelioma: a single-center retrospective cohort study suggesting increased morbidity without survival benefit

Abstract

Background

In the multimodal treatment of pleural mesothelioma (PM), hyperthermic intrathoracic chemotherapy (HITOC) has been investigated as an adjunct to cytoreductive surgery (CRS) to potentially improve local tumor control. The study aimed to evaluate the clinical impact of HITOC as an adjunct to CRS, specifically regarding overall survival (OS), postoperative morbidity and mortality, and operative duration.

Methods

This retrospective study compared patients who underwent CRS with HITOC to those without HITOC at our institution between 2010 and 2022. Primary endpoints included OS, operative time, and perioperative morbidity and mortality. Categorical variables were compared using the Fisher’s exact test. Survival analyses were performed using the Kaplan-Meier method and compared via the log-rank test.

Results

A total of 114 patients underwent CRS for PM, including 14 with HITOC and 100 without. The mean age was 66 years, and 12 patients were female. The mean operative time was longer in the HITOC group (488 vs. 258 minutes). Postoperative morbidity (Clavien-Dindo Grade ≥ III) was significantly higher with HITOC [43 % vs. 10 %, odds ratio (OR) 6.07, 95 % confidence interval (CI): 1.77–20.76, P=0.007], while in-hospital mortality occurred in two patients (1.8 %), one in each group. Median OS was 13 months in the HITOC group vs. 27 months in the non-HITOC group (P=0.03).

Conclusions

In this retrospective single-center cohort, adjunctive HITOC was associated with significantly longer operative time, higher postoperative morbidity, and inferior OS compared with cytoreduction alone. Given the small HITOC sample size and potential residual confounding, these findings should be interpreted cautiously but raise concerns regarding the routine use of HITOC outside clinical trials and highlight the need for prospective comparative evaluation.

Highlight box.

Key findings

• Adjunctive hyperthermic intrathoracic chemotherapy (HITOC) following cytoreductive surgery (CRS) for pleural mesothelioma (PM) was associated with significantly longer operative time, increased major postoperative morbidity (43% vs. 11%), and inferior overall survival compared with surgery alone.

What is known and what is new?

• Previous retrospective studies suggested that HITOC may improve local tumor control and survival in selected patients with PM.

• This study changes this assumption by demonstrating worse survival outcomes and higher perioperative burden associated with HITOC in a real-world cohort.

What is the implication, and what should change now?

• The findings raise concerns regarding the routine use of HITOC in clinical practice. Careful patient selection and critical evaluation of its risk-benefit profile are essential. Prospective, controlled studies are needed before HITOC can be recommended as a standard adjunct to CRS in PM.

Introduction

Background

Pleural mesothelioma (PM) remains an aggressive malignancy with limited therapeutic options and an overall poor prognosis despite advances in multimodal treatment strategies. Surgical cytoreduction is a central component of treatment in selected patients and aims to achieve macroscopic complete resection. However, even after optimal cytoreductive surgery (CRS), microscopic residual disease at the pleural surface is highly likely to occur and is considered a major contributor to local recurrence and disease progression. Consequently, substantial effort has been directed toward the development of adjunct locoregional therapies designed to enhance local tumor control following surgery (1).

Rationale and knowledge gap

Several adjunct locoregional strategies have been explored to improve local disease control after CRS for PM. These include hyperthermic intrathoracic chemotherapy (HITOC), intraoperative photodynamic therapy (PDT), and, in selected series, hyperthermic pleural lavage using providone-iodine or hypotonic solutions (2-5). Each of these approaches is based on the concept of delivering additional cytotoxic or cytotoxic-enhancing treatment directly to the pleural cavity at the time of surgery, thereby maximizing local exposure while minimizing systemic toxicity.

The integration of HITOC into multimodal treatment strategies for malignant PM has been widely discussed over the past decades. HITOC is employed primarily following CRS to target residual microscopic disease that remains after surgical intervention, thereby enhancing local control of the disease (2,6). The rationale for using hyperthermic chemotherapy in the thoracic cavity parallels its application in the peritoneal cavity, where it has demonstrated efficacy against various malignancies, including peritoneal mesothelioma (7,8). Recent studies have underscored the feasibility and safety of combining HITOC with CRS in patients with pleural malignancies, particularly PM, indicating that the combination of CRS and HITOC could be a viable option for selected patients, leading to improved survival rates (9-11).

Objective

Despite encouraging signals from selected retrospective series, robust comparative evidence supporting HITOC in PM remains limited and inconsistent. The aim of this study was to analyze the outcome of patients after CRS with or without HITOC at our institution. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1243/rc).

Methods

This retrospective, single-center cohort study included all consecutive patients who underwent CRS for PM at the University Medicine Essen—Ruhrlandklinik between January 2010 and December 2022. Patients were identified from a prospectively maintained institutional thoracic surgery database, which is routinely updated and used for quality assurance and clinical research purposes. Inclusion of consecutive cases was intended to minimize selection bias and to reflect real-world clinical practice over a prolonged time period encompassing changes in surgical techniques and perioperative management.

According to intraoperative and perioperative treatment strategy, patients were categorized into two groups: those undergoing CRS with additional HITOC and those undergoing CRS without HITOC. The use of HITOC was determined by the institutional treatment strategy during defined time periods, rather than by formalized individual selection criteria. Accordingly, no predefined eligibility criteria for HITOC were applied beyond feasibility and institutional practice at the time of surgery. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of The University Duisburg-Essen (approval No. 24-12313-BO). Individual consent for this retrospective analysis was waived.

Patient selection and data collection

Eligible patients had histologically confirmed PM and were scheduled for CRS with curative or disease-controlling intent, aiming for macroscopic complete resection. Surgical procedures included pleurectomy/decortication (P/D), extended pleurectomy/decortication (eP/D) or extrapleural pneumonectomy (EPP). Patients undergoing macroscopic incomplete pleurectomy performed with purely palliative intent were excluded from the analysis. Clinical, pathological, and perioperative data were extracted from electronic medical records and the institutional database. Collected variables included demographic characteristics (age, sex), tumor-related features [histology subtype, pathological stage according to the applicable tumor-node-metastasis (TNM) classification], details of preoperative systemic therapy, operative parameters, and early postoperative outcomes. Follow-up and survival data were obtained from hospital records and, when necessary, supplemented by information from referring physicians.

Surgical procedure and HITOC technique

All surgical procedures were performed via open thoracotomy by experienced thoracic surgeons specialized in mesothelioma surgery, with the goal of achieving macroscopic complete resection whenever technically feasible. P/D involves complete removal of the parietal and visceral pleura while preserving the lung parenchyma. Extended P/D comprised P/D with additional resection of the pericardium and/or the hemidiaphragm when required to achieve complete macroscopic cytoreduction. In EPP, the pleura is removed en bloc together with the lung, the pericardium, and the mediastinal lymph nodes (12). In patients assigned to the HITOC group, a closed intrathoracic perfusion circuit was established following completion of cytoreduction and hemostasis. Hyperthermic intrathoracic chemoperfusion was administered according to institutional protocol, using cisplatin at a dose of 175 mg/m² body surface area, with a targeted inflow temperature of 41–42 ℃, and a dwell time of 60 minutes. Temperature and perfusion parameters were continuously monitored through the procedure. Patients in the non-HITOC group underwent identical cytoreductive procedure without intrathoracic chemoperfusion. In both groups, additional local measures, such as lavage with hydrogen peroxide or iodine solution, could be applied at the discretion of the operating surgeon, reflecting routine clinical practice and individual intraoperative judgment.

Endpoints

The predefined primary endpoints of the study were:

• ❖ Overall survival (OS), defined as the time from the date of surgery to death from any cause or last follow-up;

• ❖ Operative time, defined as the duration from skin incision to skin closure;

• ❖ Perioperative morbidity and mortality, including 30-day mortality, in-hospital mortality, and postoperative complications graded according to the Clavien-Dindo classification (13).

Statistical analysis

Continuous variables were reported as means or medians, as appropriate, while categorical variables were expressed as absolute numbers and percentages. Group comparisons for categorical variables were performed using the Fisher’s exact test, and continuous variables were compared using the Mann-Whitney U test. OS was estimated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. All statistical tests were two-sided, and a P value <0.05 was considered statistically significant. Due to the limited number of patients in the HITOC group, multivariable regression analyses and propensity score matching were not performed, and the results should be therefore be interpreted as exploratory. Missing values were excluded only from the respective analyses. Statistical analyses were performed using SPSS software (Version 29, IBM Corp.).

Results

Cohort and baseline characteristics

Table 1 summarizes patient demographics, tumor-related characteristics, and operative as well as treatment-related details. During the study period, a total of 114 patients underwent CRS for PM, of whom 14 (12.3 %) received additional HITOC, while 100 patients (87.7 %) were treated without HITOC. Individual demographic, clinicopathological, and survival data for the patients treated with HITOC are provided in Table 2. The mean age of the entire cohort was 66 years (standard deviation 8.9). Patients in the HITOC group were significantly older, with a mean age of 70.9 years, compared with 65.9 years in the non-HITOC group (P=0.03). A total of 12 patients (10.5 %) were female, with no significant difference in sex distribution between groups. With regard to tumor histology, epithelioid mesothelioma was the predominant subtype in both cohorts, accounting for 71.4 % of cases in the HITOC group and 84.0 % in the non-HITOC group. Sarcomatoid and biphasic subtypes were infrequent, and no statistically significant differences in histologic distribution were observed. Pathological staging revealed a broad spectrum of tumor stages in both groups. Notably, nodal involvement differed between cohorts, with pN1 disease occurring more frequently in the HITOC group, whereas pN2 disease was observed exclusively in the non-HITOC group. Preoperative systemic chemotherapy was administered to 81 patients (71.1 %) overall. Importantly, all patients in the HITOC group received neoadjuvant chemotherapy, compared with 67 % of patients in the non-HITOC group, representing a statistically significant difference (P=0.01).

Table 1. Patient demographics, tumor-related characteristics, and operative details of the cohort.

Variables	HITOC group (n=14)	Non-HITOC group (n=100)	P value
Patient characteristics
Mean age (years)	70.9	65.9	0.03
Male sex	12 (85.7)	90 (90.0)	0.64
Tumor characteristics
Histology
Epithelioid	10 (71.4)	84 (84.0)	0.27
Sarcomatoid	3 (21.4)	6 (6.0)	0.08
Biphasic	1 (7.1)	10 (10.0)	>0.99
Stage
ypT0	0 (0.0)	2 (2.0)	>0.99
pT1	6 (42.9)	54 (54.0)	0.39
pT2	5 (35.7)	21 (21.0)	0.14
pT3	1 (7.1)	20 (20.0)	0.46
pT4	2 (14.3)	3 (3.0)	0.11
pN0	9 (64.3)	89 (89.0)	0.03
pN1	5 (35.7)	7 (7.0)	0.006
pN2	0 (0.0)	4 (4.0)	>0.99
Operative details
Preoperative systemic therapy	14 (100.0)	67 (67.0)	0.01
Pleurectomy/decortication	0 (0.0)	57 (57.0)	<0.001
Extended pleurectomy/decortication	8 (57.1)	28 (28.0)	0.03
Extrapleural pneumonectomy	6 (42.9)	15 (15.0)	0.02

Data are presented as number (%). HITOC, hyperthermic intrathoracic chemotherapy; N, node; p, pathologic; T, tumor; y, after neoadjuvant therapy.

Table 2. Individual patient characteristics and clinical outcomes of the HITOC cohort.

Patient	Age	Sex	Histology	Surgery	pN	Neoadjuvant therapy	OS (months)	Status
1	74	M	Epithelioid	EPP	0	Yes	12	Dead
2	69	M	Epithelioid	EPP	1	Yes	20	Dead
3	68	M	Epithelioid	EPP	0	Yes	20	Dead
4	66	F	Epithelioid	eP/D	0	Yes	24	Dead
5	75	M	Sarcomatoid	EPP	0	Yes	6	Dead
6	70	M	Epithelioid	eP/D	1	Yes	13	Dead
7	70	M	Epithelioid	EPP	0	Yes	8	Dead
8	73	M	Epithelioid	eP/D	0	Yes	30	Dead
9	73	M	Sarcomatoid	eP/D	0	Yes	22	Dead
10	73	M	Epithelioid	EPP	1	Yes	11	Dead
11	58	F	Epithelioid	eP/D	0	Yes	42	Alive
12	75	M	Epithelioid	eP/D	1	Yes	7	Dead
13	71	M	Sarcomatoid	eP/D	0	Yes	1	Dead
14	75	M	Biphasic	eP/D	1	Yes	43	Dead

eP/D, extended pleurectomy/decortication; EPP, extrapleural pneumonectomy; F, female; HITOC, hyperthermic intrathoracic chemotherapy; M, male; OS, overall survival; pN, pathologic nodal stage.

Operative details

Regarding the extent of surgery, P/D was performed in 57 patients, eP/D in 36 patients, and EPP in 21 patients. The distribution of surgical procedures differs significantly between groups. In particular, the HITOC cohort contained a substantially higher proportion of EPP, whereas no patient in the HITOC group underwent standard P/D. Conversely, P/D was the most frequently performed procedure in the non-HITOC group. The mean operative time was markedly longer in patients undergoing HITOC, with an average duration of 488 minutes, compared with 258 minutes in the non-HITOC group, reflecting both the additional chemoperfusion procedure and potentially more complex resection. Operative time was missing for 13 patients in the non-HITOC group.

Early postoperative outcomes

Postoperative morbidity (Clavien-Dindo Grade ≥ III) occurred more frequently in patients treated with HITOC than in those without HITOC [43 % vs. 10 %, respectively; odds ratio (OR) 6.07, 95 % confidence interval (CI): 1.77–20.76, P=0.007 Fisher’s exact test]. The distribution of postoperative complications by complication type and Clavien-Dindo grade is summarized in Table 3. In-hospital mortality occurred in two patients (1.8 %), with one death in each treatment group. No HITOC-related deaths were observed.

Table 3. Observed postoperative complications.

Clavien-Dindo grade	Observed postoperative complications	HITOC group (n=14)	Non-HITOC group (n=100)
I	Urinary retention, uncomplicated pleural effusion	1 (7.1)	1 (1.0)
II	Blood transfusion for postoperative anemia, atrial fibrillation/tachyarrhythmia, Clostridium difficile colitis, venous thrombosis, ileus	5 (35.7)	17 (17.0)
IIIa	Pleural drainage for pleural effusion	1 (7.1)	0 (0.0)
IIIb	Re-operation for postoperative hemothorax, revision for persistent air leak, diaphragmatic patch revision	2 (14.3)	4 (4.0)
IVa	Respiratory failure requiring re-intubation, severe pneumonia requiring ICU treatment, sepsis, acute myocardial infarction, hemiparesis	2 (14.3)	4 (4.0)
IVb	Cardiopulmonary resuscitation	0 (0.0)	1 (1.0)
V	Postoperative death	1 (7.1)	1 (1.0)

Data are presented as number (%). HITOC, hyperthermic intrathoracic chemotherapy; ICU, intensive care unit.

Survival

The median follow-up time was 24 months (range 0–177 months), with a cumulative follow-up duration of 3,976 months. No 30-day mortality was observed in the HITOC group, whereas 30-day mortality was 1 % in the non-HITOC group. The 90-day mortality rates were 7.1% in the HITOC group and 4% in the non-HITOC group. Kaplan-Meier survival analysis demonstrated a significantly shorter median OS in patients treated with HITOC compared with those treated without HITOC. Median OS was 13 months (95 % CI: 3.22–22.78) in the HITOC group vs. 27 months (95 % CI: 20.27–33.73) in the non-HITOC group (P=0.03). Kaplan-Meier survival curves for both cohorts are shown in Figure 1.

Figure 1.

Kaplan-Meier survival curves for both cohorts. Cum, cumulative; HITOC, hyperthermic intrathoracic chemotherapy.

Subgroup analyses

An exploratory subgroup analysis stratified by surgical procedure was performed to evaluate whether the observed differences were influenced by the higher proportion of ePP in the HITOC group. Among patients undergoing lung-sparing procedures (P/D or EPP), no significant differences were observed between the HITOC and non-HITOC groups regarding major postoperative complications (Clavien-Dindo Grade ≥ III; Fisher’s exact test, P=0.68). Likewise, OS did not differ significantly (median 22 vs. 28 months; log-rank P=0.37). Similarly, among patients undergoing EPP, the rate of major postoperative complications was comparable between both groups (Fisher’s exact test, P>0.99). Median OS was numerically shorter in the HITOC-group (11 vs. 27 months), although this difference did not reach statistical significance (log-rank P=0.08).

Discussion

Key findings

Published experience with HITOC in PM remains heterogeneous. While nonrandomized series from specialized centers report acceptable perioperative safety, signals regarding long-term survival are inconsistent. Our findings add to growing evidence that theoretical advantages of regional hyperthermic perfusion require careful patient selection, strict protocol standardization, and that routine application outside clinical trials may not improve outcomes.

Strengths and limitations

Strengths of this study include inclusion of all consecutive surgical patients over a prolonged period. The features reflect real-world performance of HITOC when integrated into routine surgical workflows.

The retrospective, non-randomized design is vulnerable to residual confounding. The HITOC group was small (n=14), limiting statistical power, precision of effect estimates, and adjustment for covariates. We did not perform propensity matching or time-dependent analyses of adjuvant therapies: differences in systemic treatment exposure may have influenced survival. Details such as perfusate composition, exact temperature curves, dwell time, and completeness of cytoreduction were not analyzed as effect modifiers. As a single-institution experience, generalizability is limited; outcomes in high-volume centers with standardized HITOC pathways may differ.

Comparison with similar research

Across recent series of lung-sparing CRS for PM, early postoperative mortality is generally low in experienced centers, but overall morbidity remains substantial. Prospective randomized data from the MARS2 trial have reinforced that even under controlled trial conditions, eP/D is associated with a high burden of serious adverse events and non-trivial 30- and 90-day mortality rates of approximately 4% and 9%, respectively (14). Consistent with these findings, other prospective experiences have reported low early mortality but very high complication rates following eP/D, with EORTC 1205 study documenting postoperative complications in 85% of patients despite 30- and 90-day mortality remaining at 1.7% (15). These data underscore the intrinsic physiological impact of extensive pleural surgery, even when lung-sparing approaches are employed.

When HITOC is added following CRS, direct comparative evidence remains limited. Available literature generally supports technical feasibility and early mortality rates comparable to surgery-alone cohorts, but consistently reports high postoperative morbidity (9,16). In our cohort, in-hospital mortality was low overall (1.8%); however, postoperative morbidity was markedly higher in the HITOC group compared with the non-HITOC group (43% vs. 10%). Operative duration was also significantly prolonged in patients undergoing HITOC, reflecting the additional complexity of intrathoracic chemoperfusion. In the present study, OS was significantly worse in patients undergoing CRS with HITOC than in those treated with CRS alone (13 vs. 27 months). This finding contrasts with several retrospective series that have suggested a potential survival benefit of adding HITOC after macroscopic cytoreduction, particularly in lung-sparing approaches (2,10,16).

Explanations of findings

Several factors may explain the observed result. In routine clinical practice, surgeons may preferentially offer HITOC to patients with greater tumor burden or higher risk of residual microscopic disease, which could result in a poorer prognosis. In our study, HITOC was applied during a defined institutional time period rather than based on individualized intraoperative decision-making, making major indication bias less likely, although it cannot be fully excluded. Furthermore, HITOC introduces additional physiological stress related to perfusion setup, hyperthermia, and intrathoracic exposure to cytotoxic agents. These factors plausibly contributed to the higher complication rates and prolonged recovery observed, potentially offsetting any incremental oncologic benefit from regional chemotherapy.

Over the long study period, systemic treatment regimens within multimodal treatment for PM have changed. Even if a survival difference attributable to variations in systemic therapy cannot be fully excluded, it is important to note that, in contrast to the non-HITOC group, all patients undergoing CRS with HITOC received preoperative systemic chemotherapy. This uniform exposure to neoadjuvant treatment in the HITOC cohort should therefore be considered when interpreting survival outcomes and limits the possibility that inferior survival can be explained exclusively by less intensive systemic therapy in this group. An additional limitation of this retrospective study is that detailed information on postoperative systemic therapy was not consistently available. Consequently, differences in systemic treatment could not be accounted for in the survival analyses. As postoperative treatment may have influenced OS, residual confounding cannot be excluded. There were also varying intraoperative local therapies, depending on the surgeon’s preference.

Importantly, the HITOC group in our cohort included a substantially higher proportion of EPP, a procedure associated with increased morbidity and inferior survival compared with lung-sparing surgery. This imbalance represents a major confounder and limits causal interpretation. To explore whether the higher proportion of EPP in the HITOC cohort accounted for the observed findings, an exploratory subgroup analysis stratified by surgical procedure was performed. Within both lung-sparing procedures and EPP, HITOC was not associated with a statistically significant increase in major postoperative morbidity. Likewise, no statistically significant survival benefit of HITOC was observed within either surgical subgroup. Although patients undergoing EPP showed a numerically shorter median OS following HITOC (11 vs. 27 months), the subgroup is small and the analysis was not powered to detect subgroup-specific differences. Therefore, these findings should be considered hypothesis-generating rather than conclusive. In addition, patients in the HITOC group were older on average, which may have further contributed to increased morbidity and reduced survival.

Operative duration was markedly longer in the HITOC group than in the non-HITOC group (488 vs. 258 minutes, respectively). A considerable proportion of this difference—exceeding one hour—can be directly attributed to the HITOC procedure itself, which includes circuit setup, perfusion, and controlled heating. Nevertheless, it cannot be excluded that differences in operative duration were also influenced by intraoperative findings, such as a higher local tumor burden, more extensive pleural involvement, or increased technical complexity of cytoreduction in patients selected for HITOC. These factors may have necessitated more demanding surgical dissection and prolonged operative management independent of the chemoperfusion component.

From a biological standpoint, the therapeutic window of HITOC in PM appears to be largely confined to superficial tissue layers, reinforcing the concept that HITOC should be viewed as an adjunct for microscopic residual disease after macroscopic complete resection rather than a modality capable for bulky tumor. Ex vivo human lung-pleura models exposed to hyperthermic cisplatin perfusion consistently demonstrate a steep concentration gradient with increasing tissue depth: in the frequently cited experimental setting (cisplatin 0.005 mg/mL, 60 min, ∼42 ℃), cisplatin concentrations declined from approximately 2.4 µg/mL at 1 mm to ∼0.5 µg/mL at 4 mm, translating into a median penetration depth of ∼3–4 mm (17). More recent analyses using refined measurement techniques reported a median penetration depth of ∼4.5 mm and a maximum detectable depth up to 7.5 mm (with this upper bound partly method-limited), again showing that the strongest drug exposure occurs with the first few millimeters (18). Importantly, decortication itself may enhance drug access: comparative work summarized in recent reviews indicates greater tumor cell kill at deeper level after decortication than in non-decorticated tissue, suggesting that surgical removal of the pleural rind may improve local drug delivery to the treated surface (19). Collectively, these data support a mechanistic rationale that HITOC is most likely to affect residual cell lining the pleural surface and immediate subsurface layers, whereas residual nodules beyond several millimeters may remain insufficiently exposed—underscoring why the quality of cytoreduction (macroscopic complete resection) remains the central determinant of locoregional efficacy (4,20). However, this proposed biological advantage primarily applies to lung-sparing procedures. In the present cohort, patients treated with HITOC more frequently underwent EPP than P/D. Therefore, our study was not designed to evaluate whether the potentially improved drug penetration following decortication translates into improved clinical outcomes, and this hypothesis warrants further investigation in cohorts undergoing predominantly lung-sparing surgery.

Cisplatin concentration is a pivotal, yet inconsistently reported, parameter of HITOC in PM because it directly determines the intrathoracic exposure achieved at the pleural surface while also influencing systemic absorption and toxicity risk. Across published protocols, cisplatin dosing varies widely between 50 and 250 mg/m², which—depending on the perfusate volume—translates into substantially different in-circuit cisplatin concentrations and thus non-comparable “delivered” locoregional intensity (2,6,21). In addition to cisplatin monotherapy, multi-agent HITOC regimens have been explored in PM with the rationale of increasing cytotoxic breadth and exploiting potential drug synergy at high local exposure. Systematic overviews report that, beyond cisplatin, doxorubicin and mitomycin C are among the most frequently used agents in clinical practice, either alone or in combination with cisplatin (22). In our cohort, cisplatin was used as the chemotherapeutic agent for HITOC. Despite its frequent clinical use, there remains considerable heterogeneity in HITOC protocols with respect to drug dosing, perfusion temperature, and duration, complicating comparisons between studies and underscoring the need for standardized regimens to optimize local tumor control while minimizing systemic toxicity.

Beyond HITOC and PDT, other locoregional measures have been proposed that may exert a cytoreductive effect in PM, although clinical evidence remains limited. In particular, hyperthermic pleural lavage with providone-iodine has been described by Lang-Lazdunski et al., who applied repeated intrapleural washouts using sterile water mixed with 10 % providone-iodine at 40–41 ℃ following completion of P/D. In this approach, both hyperthermia and iodine-based cytotoxicity may have contributed to tumor cell death, and the reported median OS of 32 months in patients achieving macroscopic complete resection suggests a potential oncologic relevance (4). Similarly, hydrogen peroxide has been described by Burk et al. as a commonly used adjunct during cytoreductive procedures, although systematic clinical data supporting its efficacy in mesothelioma are lacking (5). Nevertheless, in vitro studies provide biological plausibility for these approaches: Opitz et al. demonstrated growth inhibition of mesothelioma cells after short exposure to providone-iodine, while experimental work by Kim et al. and Kinnula et al. supports cytotoxic effect of hydrogen peroxide on mesothelioma cells (23-25). Taken together, these observations suggest that simple chemical or thermal adjuncts may contribute to local tumor control, but their true clinical impact remains undefined and warrants formal evaluation.

Our findings should therefore not be interpreted as arguing against the biological rationale of pleural-space-directed therapies in general. As recently emphasized by Levy et al., the pleural cavity represents a unique therapeutic compartment, and pleural-space-directed therapies offer a compelling biological strategy for the treatment of PM, particularly in patients with disease confined to the pleural surfaces (26). Nevertheless, our results suggest that the specific implementation of HITOC in our cohort did not provide an additional clinical benefit beyond macroscopic complete cytoreduction. Future studies should further define those patients and surgical settings in which intrapleural chemotherapy may be most effective.

Implications and actions needed

Prospective, multicenter studies with standardized HITOC protocols, rigorous assessment of cytoreductive completeness, and harmonized systemic therapy are needed to clarify which patients-if any-derive a net benefit. Randomized or well-matched comparative designs should incorporate patient-centered outcomes (quality of life, functional recovery), complication severity, and cost-effectiveness. Until such evidence emerges, the role of HITOC in the surgical management of PM should be considered investigational and confined to specialized centers or clinical trials.

In our setting, the lack of survival benefit compared to other local therapies coupled with higher morbidity and longer operative time argues against routine HITOC during CRS for PM. For centers, considering HITOC, these findings emphasize the need for strict selection criteria, vigilant perioperative management, and transparent discussion of risks and uncertain benefit with patients.

Conclusions

In this retrospective cohort, adjunctive HITOC did not demonstrate a survival advantage and was associated with increased morbidity and operative duration. While these findings argue against routine HITOC use in our institutional setting, definitive conclusions regarding its oncologic value require prospective comparative studies. These data may help inform institutional decision-making and patient counseling until higher-level evidence becomes available.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of The University Duisburg-Essen (approval No. 24-12313-BO). Individual consent for this retrospective analysis was waived.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1243/dss

jtd-18-08-831-dss.pdf ^{(86.9KB, pdf)}

DOI: 10.21037/jtd-2026-1243