Pressurized intraperitoneal aerosol chemotherapy, reasons for interrupting treatment: a systematic review of the literature

Abstract Objectives Pressurized intraperitoneal aerosol chemotherapy (PIPAC) gives encouraging results in the treatment of peritoneal metastasis (PM). The current recommendations require at least 3 sessions of PIPAC. However, some patients do not complete the full treatment course and stop after only 1 or 2 procedures, hence the limited benefit. A literature review was performed, with search terms including “PIPAC” and “pressurised intraperitoneal aerosol chemotherapy.” Content Only articles describing the causes for premature termination of the PIPAC treatment were analysed. The systematic search identified 26 published clinical articles related to PIPAC and reporting causes for stopping PIPAC. Summary The series range from 11 to 144 patients, with a total of 1352 patients treated with PIPAC for various tumours. A total of 3088 PIPAC treatments were performed. The median number of PIPAC treatments per patient was 2.1, the median PCI score at the time of the first PIPAC was 19 and the number of patients who did not complete the recommended 3 sessions of PIPAC was 714 (52.8%). Disease progression was the main reason for early termination of the PIPAC treatment (49.1%). The other causes were death, patients’ wishes, adverse events, conversion to curative cytoreductive surgery and other medical reasons (embolism, pulmonary infection, etc…). Outlook Further investigations are necessary to better understand the causes for interrupting PIPAC treatment and also improving the selection of patients who are most likely to benefit from PIPAC.


Introduction
Peritoneal metastasis (PM) is associated with a poor prognosis in the absence of effective multimodal therapeutic approaches. Pressurized intraperitoneal aerosol chemotherapy (PIPAC) has emerged with encouraging results in the treatment of PM [1][2][3] in the last ten years. The current recommendations require at least 3 PIPAC procedures planned every 4-6 weeks [4,5]. Unfortunately, many patients do not complete the 3 planned procedures and stop PIPAC prematurely. There are many reasons for discontinuing after only 1 or 2 procedures, with the main reason being the progression of the oncological disease [6,7]. We therefore aimed to investigate reasons for stopping PIPAC, to better select patients.
The aim of the present study is to propose a systematic review of the literature in order to evaluate the number and reasons for discontinuing PIPAC procedures.

Materials and methods
A literature review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [8]. A literature search was performed by using the Medline database (via PubMed). This first search step was performed without any language restriction. Search terms included "PIPAC" and "pressurized intraperitoneal aerosol chemotherapy." After removing duplicates, a total of 244 articles were identified up to August 1, 2022. Only articles published in English were analysed in the second steep. The following publications were excluded: case reports of 5 patients or fewer, preclinical experimental studies, including ex vivo and animal models, pharmacological or histological study, ongoing trials, case reports, technique development and safety studies, review articles, editorials and unrelated/erratum (flowchart, Figure 1). We only analysed articles that described the reasons for discontinuing PIPAC treatment.
For each article, when available, the following items were recorded for each study: authors, title, year of publication, primary tumour origin, number of patients, details of patients (median age of patients), previous chemotherapy and number of previous lines of systemic chemotherapy, total number of PIPAC procedures, details of number of PIPAC and protocol (bimodal treatment or not), median number of PIPAC sessions per patient, details on the surgical intervention (rate of non-access, peritoneal carcinomatosis index [PCI], ascites at PIPAC#1 and intraoperative complications). Postoperative outcome measures included postoperative surgical complications (morbidity), mortality, and reasons for stopping PIPAC before three sessions had been completed, which were separated into 6 groups: non-access, conversion to curative cytoreductive surgery, tumour progression intra-or extraperitoneal, patient wishes, postoperative complications/adverse events (compromising other PIPAC), death or other medical reasons (pulmonary embolism, infection, myocardial infarction, protocol violation or other cancer).

Statistical analysis
Small sample size and heterogeneity in the original data did not allow for proper statistical meta-analysis. Microsoft Excel version 16.60 and R version 4.1.3 was used respectively for data collection and statistical analysis. Continuous data are presented using descriptive statistics mean ± standard deviation or median (25th; 75th percentiles). Categorical data are presented using frequencies and percentages. Quantitative parameters were compared between groups using the Student's t-test or Mann-Whitney test when normality was rejected. Qualitative parameters were compared between groups using the Chi-square test or Fisher exact test, as appropriate. A threshold of 5 % was used to define the significance of the statistical tests.

Results
In the literature, a limited number of studies include the reasons for the early discontinuation of PIPAC procedures especially after 1 or 2 sessions. The systematic search identified 26 published clinical articles related to PIPAC with inclusion criteria (Figure 1, Flow chart).
The studies included 11 to 183 patients, with the number of reported PIPAC sessions of
The number of patients who did not complete the 3 planned PIPAC sessions was 714 (52.8 %) and the median failure of 3 planned PIPAC sessions was about 56.1 % (44.9-62.9 %). After analysing the literature, the main reason for discontinuing PIPAC treatment was disease progression (49.1 %). All reasons for PIPAC interruption are described in Figure 2 and Table 3.
Other reasons for the premature cessation of planned PIPAC sessions were bowel obstruction, adverse events (problem with healing, toxicity, etc.), patient wishes, curative intent cytoreductive surgery (CRS) and Hyperthermic IntraPEritoneal Chemotherapy (HIPEC), no secondary peritoneal access, medical reasons (embolism, sepsis, etc.) Mean SD±, per study We performed a subgroup analysis between studies with a failure rate of 3 planned PIPAC sessions ≥50 % (FAILURE Group) and studies with a failure rate <50 % (SUCCESS Group) ( Table 4). The number of studies is small, but notably, in the >50 % FAILURE group, all the treated series include PM from gastric and hepato-biliary tumours. In the SUCCESS group, we observed more patients treated for colorectal cancer or mesothelioma. Statistical analysis showed that patients in the "SUCCESS group" had a higher average mean age (58 vs. 63; p=0.007), but no statistical difference was found in terms of the presence of ascites (p=0.285), bimodal treatment (p=0.263) or previous chemotherapy (p=0.224).
We performed a subgroup analysis by creating a group of studies with >50 % PIPAC discontinuation due to progression and a group with <30 % PIPAC discontinuation due to progression (Table 5). By focusing on cancer types, it appears that colorectal origin causes less progression than cancer of gastric or biliopancreatic origin. Statistical analysis showed that patients in studies stopping PIPAC due to progression in more than 50 % of cases had a higher average PCI (14 ± 4 vs. 20 ± 5 p=0.014), but no statistical difference was found in terms of the presence of ascites (p=0.134), bimodal treatment (p=0.927) or previous chemotherapy (p=0.828). We observed more discontinuation due to progression in cases of gastric cancer.

Discussion
Reasons for discontinuing PIPAC before 3 recommended PIPAC sessions are oncological progression, conversion to CRS/HIPEC, patient wishes, adverse events after PIPAC, medical reasons or death [6]. Fewer studies reported a reason for stopping PIPAC, as our review demonstrated; among all the articles >200, only 26 described the reason for discontinuing PIPAC before 3 cycles.
The results of this review of the available literature suggest that early discontinuation of the 3 planned sessions is a problem that is repeatedly reported, with a median number of 2.3 PIPAC sessions. In 2018, Nowacki et al. reported similar results in a multicenter study [7]. Furthermore, the standard treatment protocol for PIPAC consists of 3 procedures and completion of this treatment model has been shown to lead to improved survival and good tumoural response [4,34,35]. Our analysis of success rate when performing the 3 planned PIPAC sessions showed that failure and the premature cessation of PIPAC is more frequent in young patients with gastric PM.
The main reason for the interruption of treatment before the 3 planned PIPAC sessions in our review was oncological progression. The mean rate was 49.3 ± 24.2 %. In some series, this rate increases to over 70 % [10,21,[24][25][26]29] which can be partially explained by the fact that most patients treated with PIPAC are in palliative situations with advanced, aggressive and refractory disease. Discontinuing before 3 planned PIPAC sessions for progression or clinical deterioration raises questions around the selection of patients. The potential indications for PIPAC are PM from colorectal cancer, gastric cancer, ovarian cancer, pancreatic and biliary tract cancer, peritoneal mesothelioma and appendiceal cancer 1 . In our analysis of discontinuation due to progression, in more than 50 % of cases, we found that young patients with a higher PCI stop the PIPAC protocol earlier.
PIPAC therapy also appears to be legitimate for patients who are not eligible for CRS/HIPEC and do not tolerate or have developed systemic chemotherapy intolerance [1]. For many patients, the absence of proof of efficiency for secondor third-line systemic treatments makes PIPAC a very promising option. It is probably this hope that makes us consider this treatment for our patients and overestimate the potential results.
Other reasons for stopping PIPAC before the 3 planned procedures are abdominal non-access; our review showed a rate of 6.8 %. Some studies have reported rates as high as 30 % [21,31,33]. After at least one PIPAC session, intraperitoneal chemotherapy and repeated PIPAC are well known causes for the induction of peritoneal sclerosis [36]. Moreover, patients who are eligible for PIPAC treatment often have prior abdominal surgery as CRS or HIPEC. This may explain the observed failure rate which is higher due to attempts to access the abdomen to perform PIPAC [37,38].
For the treatment of resectable peritoneal metastasis, cytoreductive surgery and HIPEC could be performed for colorectal, ovarian and gastric peritoneal metastasis as well as for pseudomyxoma peritonei and peritoneal mesothelioma. For unresectable PM, systemic chemotherapy and targeted therapy remain the standard of care [1]. Since 2011, PIPAC has been introduced as a novel treatment for PM, with an alternative for patients who are not eligible for CRS and HIPEC [19,29,39]. The use of PIPAC as a neoadjuvant treatment has been proposed for different types of cancer [40,41], with safe and feasible conclusions. Recently, PIPAC demonstrated encouraging results in patients with unresectable PM and Alyami et al. [41] demonstrated that CRS and HIPEC can be achieved in strictly selected patients with unresectable PM at diagnosis after repeated PIPAC. The reason to stop PIPAC had been identified, but with a limited frequency.
Another reason for the discontinuation of the 3 planned PIPAC sessions was patient wishes. The review showed a mean of 7.9 %, but no further explanation was found in the literature. According to Balmer et al. [6] fear of surgery or general anaesthesia and the refusal for repeated hospitalizations are some of the reasons for discontinuing PIPAC.
Bowel obstruction or perforation is some of the reasons for interrupting the 3 planned PIPAC sessions. Bowel obstruction is often a reflection of oncological disease progression with occlusion on carcinosis. Balmer et al. [6] demonstrated that the absence of a prior history of bowel obstruction before the first PIPAC session was associated with the completion of PIPAC treatment.
Post-operative morbidity is a reason for discontinuing PIPAC. Our review showed that adverse events/postoperative complications are reasons for discontinuing PIPAC and found a mean morbidity rate with CTCAE grade 3 and 4 of 8 %. This is in accordance with other reviews of the literature [2,[42][43][44]. Recently, Alyami et al. summarised 45 clinical studies with 1,810 PIPAC procedures in 838 patients [1]. The review found that repeated PIPAC sessions were feasible, with a 3 % incidence of post-operative surgical complications. Among adverse events, the most common intraoperative complication was an iatrogenic bowel injury (0-3 % of total PIPAC procedures) [45]. In a previous literature review, Winkler et al. [43] found that bowel injury related to a Veress needle or trocar insertion was the most common complication. For this reason, clinical teams continue to search for improved means of access to the peritoneal cavity to avoid complications and the deterioration of therapeutic management [37].
The main limitations of our study were the missing data for all retrospective studies and the lack of case series describing the motivations to stop PIPAC. Therefore, it would be useful to better describe the reasons of stopping PIPAC in the upcoming studies.

Conclusions
Many treatment parameters of PIPAC, such as the minimum number of PIPAC applications to induce maximum tumour regression, are still empirical with limited available evidence. It is assumed by most pioneers of PIPAC therapy that at least 3 PIPAC sessions are probably required to induce a substantial regression of PM. Many patients do not complete the 3 planned PIPAC sessions, especially due to rapid oncological progression. This review seemed to demonstrate that PIPAC could be more successful in the case of colorectal, ovarian or mesothelioma PM. It appears that PIPAC in gastric PM gives poor results, especially in young patients with high PCI. As reported by Balmer et al. investigations are necessary to better understand the causes, in order to improve the selection of patients who are most likely to benefit from PIPAC.
Research funding: None declared. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission. Conceptualization, data curation and original draft writing: AC Ezanno. Writing -Review: B Malgras. Supervision, Writing review: M Pocard.