Treatment of peritoneal disease arising from mucinous vs. non-mucinous appendiceal neoplasms with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy

Study population

This retrospective cohort study is based on a prospectively maintained database of all patients proceeding to CRS/HIPEC from 01/01/2008 to 30/06/2022 at a dual-site, single-service tertiary referral centre in Aotearoa New Zealand. Aotearoa New Zealand is a country of five million. The Waikato region accepts referrals from all regions of the country and provides CRS/HIPEC through Waikato Hospital in the public sector and Braemar Hospital in the private sector. All patients with disseminated peritoneal disease from an appendix primary who underwent surgery with the intention to perform CRS/HIPEC were included in the study.

Patient management

Patient management pathways in Waikato have been described previously and are briefly summarised here [9]. All patients were discussed at a multidisciplinary meeting (MDM) and considered for CRS/HIPEC if they had no major organ dysfunction, good performance status, no unresectable distant metastases, and peritoneal disease amenable to complete cytoreduction on preoperative investigations. No peritoneal cancer index (PCI) restrictions were placed on peritoneal dissemination from an appendiceal primary. CRS/HIPEC was performed according to the standard Sugarbaker technique with mitomycin C administered via an open coliseum technique at 41–42 °C for 90 min when complete cytoreduction was achieved [10]. A PCI and completeness of cytoreduction (CC) score were calculated for each patient [11].

The pathological specimens from the CRS/HIPEC and the index operation at the referring hospital were routinely reviewed at the MDM. Routine postoperative surveillance consisted of four monthly tumour markers (CEA, CA19-9 and CA 125) and a computerised tomography (CT) of the chest, abdomen, and pelvis at 1, 2, 3, and 5 years. Long-term survival and recurrence data were collected from the most recent clinical or radiological follow-up at the referring centre and the Waikato region.

Histology assessment

Primary appendiceal tumours and peritoneal disease were classified according to the Peritoneal Surface Oncology Group International (PSOGI) consensus [2]. Tumours were defined as mucinous when more than 50 % of the cross-sectional area histologically comprised extracellular mucin and as non-mucinous when this was not the case, according to the World Health Organization classification [12].

According to the PSOGI consensus, appendiceal tumours are classified as low-grade appendiceal mucinous neoplasm (LAMN), high-grade appendiceal mucinous neoplasm (HAMN), mucinous adenocarcinoma (MA), intestinal adenocarcinoma of the colorectal type (ITAC), goblet cell adenocarcinoma (GCA), and neuroendocrine tumour [2]. GCA and mixed adenoneuroendocrine carcinoma (MANEC) have features of both adenocarcinoma and neuroendocrine tumours [12]. GCA was previously known as goblet cell carcinoid or adenocarcinoid. GCA is defined by well-differentiated goblet cells with positive epithelial (CK20, CDX2, CD46) and neuroendocrine (synaptophysin and chromogranin) staining [12]. MANECs display high-grade and poorly differentiated neuroendocrine features with ki-67>20 % [12]. GCA and MANEC can be mucinous or non-mucinous. The mucinous disseminated peritoneal disease is characterised by mucinous ascites and peritoneal implants and is classified as (1) acellular mucin (AC), (2) low-grade mucinous carcinoma peritonei (LG), (3) high-grade mucinous carcinoma peritonei (HG), and (4) high-grade mucinous carcinoma peritonei with signet ring cells (HG-S) [2].

Statistical analysis

Descriptive statistics are reported as frequencies and proportions for categorical variables and median and interquartile range for continuous variables. Statistical testing for baseline characteristics was conducted using the Kruskal–Wallis rank sum test, Wilcoxon rank sum test, Chi-square test, and Fisher’s exact test as appropriate. Overall survival (OS) and disease-free survival (DFS) were assessed using Kaplan–Meier survival analysis. Survival and recurrence were determined from the time of procedure to the time of death and recurrence, respectively. Patients were censored at their last follow-up date. Factors affecting survival were identified using univariate and multivariate Cox proportional hazard models. Significant factors in the univariate analysis were entered into the multivariate model. p-values were two-sided, with statistical significance evaluated at the 0.05 alpha level. Statistical analyses were conducted using R Studio version 2023.03.0 + 386 for Mac.

Ethics consideration

This study was approved by the New Zealand Health and Disability Ethics Committee (HDEC 2023 FULL 15566) on 23/05/2023 and was conducted in accordance with the Declaration of Helsinki.

Appendiceal non-mucinous neoplasms

Among 19 patients with NMANs, nine were ITAC, eight were GCA, and two were MANEC. Among the nine ITACs, five cases were moderately differentiated, and four were poorly differentiated. No cases were associated with signet ring cells (SRCs). The adenocarcinoma component of GCAs was moderately differentiated in three patients and poorly differentiated with SRC in four patients. Both cases of MANEC were poorly differentiated with SRC. Fifteen patients had previous surgery, including six appendicectomies and five right hemicolectomies. Lymph nodes were sampled in 10 patients and involved in seven patients. Seven patients received previous systemic chemotherapy.

Compared to MANs, NMANs were associated with a significantly higher tumour grade (p<0.001), more involved lymph nodes (p<0.001), and a higher proportion of positive lymph nodes (p<0.001) and systemic chemotherapy use (p=0.002).

Appendiceal mucinous neoplasms

A total of 209 patients were diagnosed with MAN, including 156 LAMN, 21 HAMN, 29 MA, 1 GCA, and two replaced primaries. A total of 125 patients underwent previous surgery for MAN, including 68 appendicectomies and 46 right hemicolectomies, among other surgeries. Lymph nodes were sampled in 113 patients and involved in nine. Eighteen patients received previous systemic chemotherapy.

Operative outcomes

Table 2 displays operative and early postoperative outcomes. Patients with HG and HG-S mucinous neoplasms were grouped as HG due to small numbers. The median PCI was lower in AC (4) compared to LG (23), HG (19), and NMAN (26) (p<0.001). The complete cytoreduction rate was 100 % (30/30) in AC, 87.7 % (114/130) in LG, 69.4 % (34/49) in HG, and 42.1 % (8/19) in NMAN (p<0.001). There was an associated decrease in HIPEC use in AC compared to NMAN (p<0.001). Eleven patients with NMAN had incomplete cytoreduction. Eight underwent palliative debulking and bypass, and three had open-and-close procedures. Complete cytoreduction was not achievable due to infiltrative small bowel and mesenteric disease in eight patients and extensive upper abdominal disease, including periportal and peri-gastric disease, in three patients.

Stomas were significantly less likely to be formed in AC (0 %, 0/0) compared to LG (33.1 %, 43/130), HG (32.7 %, 16/49), and NMAN (26.3 %, 5/19) (p=0.003). The number of organs resected (p<0.001), the number of peritonectomies (p<0.001), operative duration (p<0.001), and red blood cell transfusions (p=0.002) were the highest in LG, followed by HG and AC, and the lowest in NMAN. A greater proportion of most organs were resected in patients with MAN than NMAN. However, the small bowel was resected in 36.8 % (7/19) of NMAN compared to 9.6 % (20/209) of mucinous disease (p=0.003).

When the analysis was confined to cases with complete cytoreduction, patients with mucinous and non-mucinous disease had a similar PCI (14 vs. 17, p=0.6), lower than the main cohort. There were no longer differences between the two groups in organ resection, peritonectomy, duration, and transfusion.

Short-term outcomes were comparable between the mucinous and non-mucinous groups. Within the NMAN group, there were three cases of major complications (15.8 %, 3/19), no return to theatre (0 %, 0/19), or perioperative mortality (0 %, 0/19).

Survival

Median follow-up was 16 months in NMAN and 53 months in MAN. Seventy-two patients died. The cause of death was disease progression in 62 patients, unrelated causes in six patients, treatment side effects in three patients, and unknown in one patient. NMAN had significantly poorer OS (Figure 1). The median OS was not reached in MAN and 16.0 months in NMAN. Among MAN, the median OS was not reached in AC and LG and 46.6 months in HG (Table 3). 5-year survival was 92.8 % in AC, 81.4 % in LG, 38.0 % in HG, and 6.3 % in NMAN (p<0.001).

Figure 1:

Overall survival in patients with peritoneal dissemination from mucinous and non-mucinous appendiceal neoplasms. AC: acellular mucin; LG: low-grade carcinoma peritonei; HG: high-grade mucinous carcinoma peritonei with and without signet cells; NMAN: non-mucinous appendiceal neoplasm.

Within the NMAN subgroup, ITAC had the poorest survival (Figure 2, p=0.001). Among nine patients with ITAC, only two patients had complete cytoreduction, and both recurred. Eight patients with ITAC died with a median OS of 3.9 months. Among eight patients with GCA, the median OS was 25.1 months. Six had complete cytoreduction; all recurred. Both patients with MANEC had incomplete cytoreduction and died with a median OS of 32.2 months. NMAN patients who had complete cytoreduction had a median OS of 25.1 months.

Figure 2:

Overall survival in patients with peritoneal dissemination from non-mucinous appendiceal neoplasms. GCA: goblet cell adenocarcinoma; ITAC: intestinal adenocarcinoma of the colorectal type; MANEC: mixed adenoneuroendocrine carcinom.

On univariate analysis, poor survival across the entire cohort was associated with advanced age (p=0.008), American Society of Anaesthesiologists score>1 (p=0.025), high European Cooperative Oncology Group (ECOG) performance status (p<0.001), previous systematic chemotherapy (p=0.001), high PCI (p<0.001), high grade (p<0.001), incomplete cytoreduction (p<0.001) and tumour type (p<0.001), the presence of SRC (p<0.001), and lymph node involvement (p<0.001) (Table 4). Lymph node involvement was not included in the multivariate model due to a large number of missing data. Incomplete cytoreduction (p<0.001) and NMAN (p=0.041) remained significant predictors in multivariate analysis. In a separate model for NMAN patients alone, high PCI (p=0.02), ECOG (p=0.004), and high tumour grade (p=0.005) were risk factors for poor survival in NMAN.

Recurrence

Recurrence occurred in 68 out of 186 patients with complete cytoreduction. The site of recurrence was peritoneal in 62 patients, systemic in 13 patients, and pleural in eight patients. Peritoneal recurrence, was more likely to occur in NMAN, HG, and LG, compared to AC (p=0.001) (Table 5). Extraperitoneal recurrence, including pleural and systemic recurrence, was more likely to occur in NMAN than HG, LG, and AC. Among six GCAs that received complete cytoreduction, recurrence occurred in the peritoneal cavity in four patients, the kidney in one patient, and the lung in one patient. Both cases with ITAC recurred in the peritoneal cavity; one also developed pleural and lung metastases. Following complete cytoreduction, DFS was significantly worse in NMAN compared to HG, LG, and AC (p<0.001) (Table S1).