Appendiceal Goblet Cell Carcinoma: Role of Cytoreductive Surgery (CRS) and Hyperthermic Intraperitoneal Chemotherapy (HIPEC)

Abstract

Goblet cell carcinoma (GCC) encompasses a separate entity in appendiceal neoplasms with mixed glandular and neuro-endocrine pathological features. GCC mostly presents as an acute appendicitis duo to luminal obstruction or as an incidental finding on the surgical appendectomy specimen. In case of tumour perforation or presence of other risk factors, guidelines suggest additional treatment with a completing right hemicolectomy or cytoreductive surgery (CRS) with hyperthermic intraperitoneal chemotherapy (HIPEC). We report the case of a 77-year-old male with symptoms of appendicitis for which an appendectomy was performed. The appendix was ruptured during the procedure. There was an incidental finding of GCC on the pathological specimen. Because of possible tumour soiling, the patient received a prophylactic CRS-HIPEC. A literature review was performed to investigate the potential role for CRS-HIPEC as a curative treatment in patients with GCC. GCC of the appendix is an aggressive type of tumour with a high risk of peritoneal and systemic dissemination. CRS and HIPEC is a treatment option: both in a prophylactic setting and in patients with established peritoneal metastases.

Introduction

Goblet cell carcinoma (GCC) is an appendiceal tumour with a prevalence of 0.12–0.16 cases per 1,000,000 [1–3]. Rarely GCC occurs in the stomach, the colorectum or in the ampulla of Vater [2, 4, 5]. The incidence of appendiceal GCC is equal in men and women [6]. Mean age at the time of diagnosis is 52–59 years [3, 7]. GCC can clinically present as an acute appendicitis due to luminal obstruction caused by the tumour. In this case, imaging is nonspecific in differentiating between benign appendicitis and GCC because of their similar appearance characteristics [8, 9]. More often GCC will be an incidental finding on the histopathological examination of an appendectomy specimen, which happens in 0.3–0.9% of appendectomies [10]. GCC has a widely variable course. 51% of patients present with localised disease, 30–73% with regional disease and 15–65% with metastatic disease [6, 11–14]. Incidence of peritoneal involvement is reported in as high as 77% of cases [14]. GCC can appear as a Krukenberg tumour in women, due to the transcoelomic spreading of cells [14, 15]. Guidelines suggest treatment with a laparoscopic appendectomy to be sufficient in the absence of risk factors [16]. A right-sided hemicolectomy is advised in case of adverse tumour features (Tang B/C lesion, pT3/4 tumour, positive margins, mesoappendiceal invasion > 3 mm, angio- or neuroinvasion and Ki67 > 2%) [16–20]. In case of perforated GCC or evidence of peritoneal spread, adjuvant cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS-HIPEC) could be considered [15, 16, 20–29]. Currently, disseminated disease (stage III or IV) is treated with chemotherapy regimens extrapolated from adjuvant therapy in colorectal carcinoma, typically 5-fluorouracil (5-FU), FOLFOX (5-FU, leucovorin, oxaliplatin) and FOLFIRI (5-FU, folic acid, irinotecan) [15, 17, 27, 29]. Perforation during appendectomy reportedly occurs in 22% of cases with GCC [30]. Perforation of the appendix does not seem to affect overall survival (OS) in these patients, but data on this subject are scarce [30]. The reported 5-year disease-free survival (DFS) of all patients with GCC varies between 18 and 86%, depending on the disease stage [1]. Recurrence ensues in 18–50% of patients with a mean time of 51.5 months between initial surgery and recurrence [27].

This paper aims to evaluate CRS-HIPEC as a treatment option for goblet cell carcinoma in a prophylactic or curative setting when risk factors are present, by presenting a unique case and a review of literature.

Case Report

A 77-year-old man consulted our out-patient clinic with recurrence of an epigastric abdominal wall hernia after previous prosthetic abdominal wall repair. Concurrently, he complained of subacute abdominal right fossa pain, irradiating to the lumbar area, without associated systemic complaints. He had no previous oncologic history. The McBurney sign was positive during clinical investigation. Because of these complaints, a computed tomography (CT) scan was performed, which demonstrated discrete infiltration of the mesentery at the level of the right fossa around the appendix (Fig. 1c). The appendix was normal at the base with irregular thickening of the tip (Fig. 1a–b). There was limited contiguity inflammation of the caecum. There were no collections, nor free fluid, nor free air. Based on the CT findings, additional blood works were done, which were negative for inflammatory markers. A repeated biochemical exam, 1 week later, revealed a minimal rise in leucocytosis (11.3 × 10³/µL) and CRP (9.7 mg/L). The patient was subsequently planned for an elective laparoscopic appendectomy, combined with a primary abdominal wall repair. During the procedure, the appendix appeared to have been heavily inflamed and was severely adherent to the caecum and the ileum. The appendix was macroscopically perforated during dissection. The patient was discharged on the first postoperative day.

Fig. 1.

a Computed tomography (CT) with oral and intravenous iodine contrast in the axial plane demonstrates irregular thickening of the tip of the appendix (large arrow).The latter is adhesive with the medial wall of the caecum (small arrow). b CT scan with oral and intravenous iodine contrast in the axial plane: The proximal part of the appendix (small arrows) is normal. c CT with oral and intravenous iodine contrast in the axial plane: Streaky infiltration of the mesentery is seen in the vicinity of the thickened appendix (arrows)

The appendix had a length of 7.1 cm and a diameter of 1.1 cm. Macroscopic examination showed no apparent tumour. On microscopy, there were foci of active inflammation, organising fibrosis and two small diverticula. Additionally, the wall of the appendix contained a tumour growing as tubules composed of goblet-like mucinous cells, variable numbers of endocrine cells and cells with granular eosinophilic cytoplasm (Paneth-like cells) (Fig. 2a–b). Nuclear atypia was mild. There were tumour cells infiltrating as single mucinous or non-mucinous cells or anastomosing tubules and linear arrays (high-grade pattern, comprising 25–50% of the tumour). The tumour invaded the muscularis propria (pT2). There was no perineural or lymphovascular invasion. The base of the appendix (section margin) was free of tumour. The Periodic acid–Schiff diastase demonstrated intracytoplasmic mucin. The tumour cells showed variable immunoreactivity for cytokeratin 20 and caudal-type homeobox transcription factor 2 (CDX2) but were negative for cytokeratin 7. Immunohistochemical staining for chromogranin A and synaptophysin highlighted variable numbers of endocrine cells. A staining with a pan-keratin antibody showed the presence of single infiltrating tumour cells to a greater extent (Fig. 2c). Based on the morphology and the immunohistochemical findings, the tumour was diagnosed as an appendiceal goblet cell adenocarcinoma grade 2 (in accordance with the three-tiered grading system proposed in the WHO classification of tumours of the digestive system, 5th edition) [31].

Fig. 2.

a A haematoxylin–eosin-stained section shows the presence of clusters and nest of cells composed of goblet-like mucinous cells and Paneth-like cells (haematoxylin–eosin, original magnification × 100). b An immunohistochemical staining for synaptophysin demonstrates endocrine cells in the cell clusters (original magnification × 400). c Immunohistochemistry with a pan-keratin antibody shows in addition to cell clusters, single infiltrating tumour cells (high-grade pattern) (original magnification × 100)

The macroscopically perforated appendiceal GCC elevated the risk of developing subsequent peritoneal metastases (PM). In view of these perioperative and pathological findings, the multidisciplinary team meeting discussed the options of either follow-up or completion right hemicolectomy. The additional option of prophylactic CRS, including right hemicolectomy, combined with HIPEC was discussed with and accepted by the patient. In order to achieve a visual margin negative resection, the CRS included an extended right hemicolectomy, a total omentectomy and a peritonectomy of the right lower quadrant. Bidirectional intraoperative chemotherapy (BIC) according to the Sugarbaker Regimen was added to the surgery [32, 33]. HIPEC with mitomycin C 35 mg/m² was applied at 42 °C for 90 min. At the same time, systemic chemotherapy with 5-FU 400 mg/m² and leucovorin 20 mg/m² was administered as intravenous boluses via separate lines to improve the cytotoxic effect [34, 35]. ICU stay was one night. The postoperative course was complicated with a prolonged gastroparesis and a surgical site infection which was treated with wound care and appropriate antibiotic treatment. Pathologic examination exhibited no remnant carcinoma in the colon specimen, nor in the peritonectomy. Fifteen lymph nodes were tumour-free. Hospital discharge was on day 31. The patient was readmitted the same day with an acute evisceration through the laparotomy wound. The wound was revised with a reconstruction of the abdominal wall. The patient developed enterocutaneous fistulas for which he chose, after long counselling, not to seek further treatment but opted for palliative care.

Discussion

Epidemiology

The prevalence of appendiceal GCC is 0.12–0.16 cases per 1 million [1–3]. The incidence is equal for man and woman in some series, while other retrospective trials report a higher female to male ratio [3, 12, 15]. The incidence is higher in Caucasian race (0.182) compared to African American race (0.136) and is more than double of the incidence in Asian (0.077) or American Indian (0.059) race [3]. There does not seem to be an explanation for this given [17]. Mean age at diagnosis is 52–59 years [3, 7]. More rarely, GCC is described in different case reports on the level of the stomach, the colorectum or in the ampulla of Vater [2, 4, 5]. A possible risk factor for development of GCC is schistosomiasis, but the combined presence of GCC and schistosomiases was only reported in three cases [36]. Protein mismatch repair instability is seldom present in patients with GCC [37, 38]. TP53, AIRD1A, SMAD4, KRAS and APC mutations have been described in GCC cases [37–40].

Clinical Presentation and Imaging

Most commonly, patients present with symptoms relatable to acute appendicitis, corresponding to abdominal pain in the right fossa. Other patients present with a small bowel obstruction or a palpable mass (50%), mostly in case of advanced stage disease [12, 15]. Only < 1% of patients receive a preoperative diagnosis of GCC, which demonstrates that it is most often an incidental finding [12]. GCC has a widely variable clinical course [15]. Fifty-one percent of patients present with localised disease, 30% with regional disease (positive lymph nodes) and 15–70% with metastatic disease [6, 11–14]. A total of 44–64% of patients present with PM upon diagnosis, whereas 21% have distant metastases to the ovaries, liver, lung or other sites [14, 41]. Eighty percent of women with disseminated disease present with ovarian metastases caused by trans-coelomic migration of GCC [1, 12, 15].

CT imaging is a nonspecific exam for preoperative diagnosis of GCC. Nonetheless, CT or magnetic resonance imaging (MRI) has an important role in the detection and staging of metastatic disease [9]. A retrospective study on 28 patients with GCC reveals peritoneal involvement in 41%, ascites in 48%, and caecal wall thickening or peri-caecal infiltration in 56% of cases [8]. Involvement of the ovaries is seen in 27% of females [8]. Thirty-four percent of patients demonstrate signs of appendicitis: an abnormally distended appendix (> 6 mm) with hyper-enhancing wall and surrounding inflammatory changes [8, 9]. Presence of abscess or an appendicolith seems to be absent in patients with GCC [8]. Radiological distinction between GCC and cystic mucinous neoplasms can be made by presence of mucoceles in the latter [9]. Fluorodeoxyglucose positron emission tomography (PET) scan or somatostatin-receptor imaging (SRI) can be applied for the detection of PM [17, 42, 43]. Nonetheless, the sensitivity of somatostatin-receptor imaging decreases with the loss of neuro-endocrine differentiation, corresponding to less somatostatin expression and seems to be less useful for staging GCC [44, 45]. Exact data on the sensitivity and specificity of these exams for the detection of GCC are lacking, possibly due to the rare incidence of GCC.

Pathology

GCC has long been considered a subtype of neuroendocrine neoplasm due to its combined neuroendocrine and mucinous appearance on histologic exam [8, 46]. GCC is a subtype of mixed adeno-neuro-endocrine carcinoma (MANEC) and is to be distinguished from neuro-endocrine tumours and appendiceal adenocarcinoma [17, 31]. GCC typically consists of an amphicrine cell type, with exocrine and endocrine characteristics, and thus mixed glandular and neuro-endocrine pathological features. The hallmark feature consists of tight round or oval cluster epithelial cells with a goblet cell morphology in the lamina propria, in combination with scattered neuroendocrine cells [10]. The appendiceal mucosa is spared [10, 12]. Paneth cells (secretory epithelial cells) can be present [1]. Immunohistological stainings for neuroendocrine markers are inconsistent. Chromogranin A and synaptophysine can be positive, as in the case report [1, 10, 29]. Cytokeratin 7 (CK7) and cytokeratin 20 (CK20) immunohistochemical stainings are often positive and suggest the intestinal epithelial origin [1, 7, 10, 29]. In the case report, CK7 is negative and CK20 is only partially positive. The vacuolated cells are positive in the periodic-acid-Schiff diastase stain, marking the presence of mucin [1]. This also corresponds to the case report. GCC rarely presents with protein mismatch repair (MMR) deficiency [29]. Mutations in TP53 (6–24%), ARID1A (0–15%), SMAD4 (0–9.4%) and KRAS (0–7.5%) have been noted in the next-generation sequencing (NGS) panel [10, 37–40]. As in colorectal carcinoma, KRAS-mutated GCC should not receive treatment with anti-EGFR monoclonal antibodies [47]. Application of TP53 (tumour suppressor gene) reactivators proved benefit in TP53-mutated haematological cancer, prostate cancer and oesophageal carcinoma but has not yet been applied in TP53-mutated GCC [48, 49].

Staging Classifications

Several staging systems aim to categorise GCC in accordance with its prognosis (Table 1). The application of these staging systems can be challenging due to their complexity, and the rare incidence of GCC. The first system, developed by Burke, distincts three categories (tubular carcinoid, goblet cell carcinoid and mixed carcinoid-adenocarcinoma) based on histological and clinical descriptive qualities [46]. Successively, the classification system by Taggart is defined by the presence of adenocarcinoma, as described by Burke, as the varying amounts of adenocarcinoma appear to have an implication on prognosis. Taggart debates that a minor component adenocarcinoma (< 50%) withholds a better prognosis in comparison to a major component of adenocarcinoma (> 50%). Nevertheless, this hypothesis cannot reveal a significant difference in OS[27]. The Tang classification seems to be the most reproduced and widely accepted classification system in literature. It is based on morphologic criteria as well as the presence of goblet cells, but without quantitative definition. Tang A is defined as the typical GCC, while Tang B is an adenocarcinoma ex GCC with signet ring cell type and Tang C defines an adenocarcinoma ex GCC, the poorly differentiated carcinoma type [12]. Alas this classification is deemed challenging due to the subjectivity of the applicated criteria [10, 50]. Subsequently, a simplified two-tier grading system (high grade versus low grade), developed by Lee, is based on three histologic criteria. First, cytologic atypia, present in at least 1 focus > 1mm² in size. Second, presence of stromal desmoplasia. Third, solid growth pattern with at least one focus of > 1mm² in size. GCC is marked high grade when two or more of these criteria are present [41]. Lastly, Yozu developed a slightly altered three-grade system which first defines low-grade and high-grade histologic features, which are then categorised by percentual appearance of these low- and high-grade features. Grade 1 has < 25% high-grade features. Grade 2 has 50–75% low-grade features. Grade 3 has < 50% low-grade components. This system is designed to be more applicable and relevant in relation to prognosis. It is also applied in the WHO 5th edition classification system [51].

Table 1.

Comparison table of described pathological classifying systems for GCC

	Tang	Taggart	Yozu/WHO	Lee
Low grade	Tang A: typical GCC - Cohesive clustered architecture	GCC with < 25% adenocarcinoma - GCC in muscularis propria - Signet cells in linear matter - Small mucin pools	> 75% low-grade features +  < 25% high grade features	0–1 point(s) of high grade features
Intermediate grade	Tang B: Adenocarcinoma ex GCC, signet cell type - Loss of goblet cell clusters - Signet cell features	GCC with 25–50% component adenocarcinoma - Infiltrative cords of cells - Complex irregular architecture - Destruction/invasion in 25–50% of the specimen	50–75% low-grade features + any combination of high grade features
High grade	Tang C: Adenocarcinoma ex GCC, poorly differentiated adenocarcinoma type - Minimal 1 focus of undifferentiated adenocarcinoma	GCC with > 50% component adenocarcinoma - Infiltrative cords of cells - Complex irregular architecture - Destruction/invasion in > 50% of the specimen	< 50% low-grade features + any combination of high grade features	2–3 points of high grade features
Low-grade features	- Tubular growth with round to oval discrete tumour clusters (goblet cells, cuboidal cells, Paneth cells) - Simple trabecular growth consistent with longitudinally sectioned tubules - Limited tubule fusion or crowding - Mucin pools with discrete tubules or clusters - Tubular nonmucinous glands, including oncocytic tubules
High-grade features	- Cytologic atypia	- Mucin-poor tumour cells in nests or clusters with high nuclear to cell ratio and jagged outlines - Single cells, including nonmucinous single cells and signet ring like cells, admixed with abortive tubules - Glands lined by cuboidal or columnar cells with high cytologic grade that resembles conventional adenocarcinoma
	- Stromal desmoplasia	- Very large aggregates of goblet cells or drifts of goblet cells in extracellular mucin - Single file growth or sheets of tumour cells, admixed with abortive tubules
	- Solid growth pattern	- Fusion of goblet cell clusters to form anastomosing complex growth of goblet cell clusters or tubules - Glands floating in mucin lined by columnar cells with high cytologic grade

Treatment of Localised GCC (Stage I–II)

Appendectomy alone suffices for the treatment of GCC in localised disease, in absence of risk factors (< 1 cm, Ki67 < 2%, pT1/2, tumour-free margin edges) [16, 52, 53]. This agrees with the Peritoneal Surface Oncology Group (PSOGI) recommendations [16]. The European Neuroendocrine Tumour society (ENETS) and the North American Neuroendocrine Tumour society (NANETS) recommend an additional right hemicolectomy after appendectomy, even in localised disease [17, 20].

Treatment of Regional and Metastatic GCC (Stage III–IV)

Surgical resection is the first-line treatment, even in advanced stage GCC. Appendectomy should be followed by a right hemicolectomy [17]. In case of PM, a CRS-HIPEC should be considered [16, 17].

The use and choice of adjuvant and palliative systemic chemotherapy is extrapolated from treatment of metastatic gastrointestinal adenocarcinoma, and its application is recommended by the ENETS guidelines [17]. The Chicago Consensus Working group advises adjuvant systemic chemotherapy in stage I–III disease in case of high-grade GCC [54]. Yet, there are no specific prospective trials available, evaluating these effects on GCC specifically [17, 29]. Admittance of chemotherapy in stage IV disease is not significantly associated with better survival (HR 0.9; p = 0.86) [55].

Treatment with CRS-HIPEC for PM

The aggressive nature of peritoneally disseminated GCC advocates for a primary profound surgical resection in order to prevent peritoneal or systemic recurrence [22]. All published guidelines concerning treatment of GCC with CRS-HIPEC are eminence based due to the rare incidence of GCC. The European Neuroendocrine Tumour society (ENETS) guidelines suggest CRS-HIPEC in case of advanced PM, based on the publications by Mahteme and Pahlavan [7, 17, 23]. The Peritoneal Surface Oncology Group (PSOGI) developed recommendations based on three Delphi voting rounds with GRADE-based questions amongst a panel of 80 experts with a consensus threshold of 51% [16]. CRS-HIPEC is strongly recommended in case of established peritoneal resectable disease (CC0-1) (consensus 82%) [16]. PSOGI gives a weak positive recommendation for (prophylactic) treatment of perforated GCC without evidence of peritoneal spread with CRS-HIPEC (consensus 76%) [16].

Multiple publications examine the potential beneficial effect on survival of CRS-HIPEC in patients with GCC and PM, listed in Table 2. The applied HIPEC regimens are described in Table 3. Meta-analysis reveals that patients who undergo CRS-HIPEC have an OS of 39 months, in comparison to 7 months for patients who do not receive HIPEC (p = 0.001) [24]. However, the quality of evidence is doubtful since only 15 patients are included in this meta-analysis (11 CRS-HIPEC and 4 CRS).

Table 2.

Overview of literature applying CRS-HIPEC as treatment for GCC

		N	CRS  + HIPEC	CRS¥	Median PCI	LN (%)	CC 0–1 (%)	Median OS	Reported survival
Matheme Sugarbaker	2004	22	20	2	NA	NA	30	18.4 mo	5Y OS 25%
McConnel YJ	2014	45	36	9	24	51	71	NA	3Y OS 68%
Randle RW	2015	31	31	0	NA	61	36	18.4 mo	NA
Lamarca A	2016	56	25	0	1	NA	96	51.2 mo	5Y OS 48%
Radomski M	2016	43	43	0	20	NA	86	22–36 mo	3Y OS 39% 5Y OS 9%
Hsin-Hsien Y	2017	15	11	4	NA	13	27	17 mo	3Y OS 57%
Madsen	2018	48	21	0	NA	24	100	NA	3Y OS 76% 5Y OS 57%
Shyu S	2020	63	63	0	15/22/20*	NA	95	37 mo	5Y OS 31%

P prospective, R retrospective, N number, CRS cytoreductive surgery, HIPEC hyperthermic intraperitoneal chemotherapy, PCI peritoneal carcinomatosis index, LN lymph node involvement, CC completeness of cytoreduction, NA not available. ¥ The reasons for omission of HIPEC were not disclosed, except in 1 patient in Yu HH: HIPEC was omitted due to intolerance. *G1/G2/G3

Table 3.

HIPEC regimens used in available literature on GCC treated with CRS-HIPEC

		HIPEC regimen
Matheme Sugarbaker	2004	Chemotherapy at 42–43 °C NOS
McConnel YJ	2014	Mitomycin C or oxaliplatin NOS
Randle RW	2015	Mitomycin C or oxaliplatin NOS
Lamarca A	2016	Mitomycin C at 42 °C in 3 pulses, during 90 min in total
Radomski M	2016	Chemotherapy at 42 °C NOS
Hsin-Hsien Y	2017	Mitomycin C (12 mg/m2) or oxaliplatin (60 mg/m2) and 5-fluorouracil (300 mg/m2) at 42.5–43.5 °C during 40 min
Madsen	2018	Mitomycin C (35 mg/m2) at 41.0–42.5 °C, during 90 min
Shyu S	2020	30 mg mitomycin C during 60 min with addition of 10 mg mitomycin C during the last 40 min at 42 °C

N number, HIPEC hyperthermic intraperitoneal chemotherapy, NOS not otherwise specified

* Chemotherapy agent not disclosed

Matheme and Sugarbaker are the first authors to publish long-term results on 22 patients with peritoneally disseminated GCC treated with CRS-HIPEC by performing a retrospective analysis of patient data collected between 1981 and 2003. Varying HIPEC regimens are used in this timeframe, with simultaneous evolvement of the perioperative HIPEC protocols [23]. Still, the median OS of 18 months corresponds to other published OS rates (Table 2) [23]. Mahteme and Sugarbaker conclude the peritoneal cancer index (PCI) as well as completeness of cytoreduction to be significant determinants of survival [15]. Patients with a PCI of ≤ 10 have an OS of 62 months, whereas the OS is only 21 months and 6 months in patients with a PCI between 11–20 and > 20 respectively (p = 0,008) [23]. Another study confirms that a high PCI (≥ 27) is associated with a lower OS (16 months), in comparison to a lower PCI (< 27) (OS 30 months) (p = 0.026) [24]. Patients with a complete cytoreduction (CC0-CC1) have an OS of 29 months, opposed to 18 and 6 months in a CC2 or CC3 cytoreduction, respectively (p = 0.007) [23]. The PCI directly indicates tumour load and is thereby a nominator for peritoneal disease burden [56]. Nonetheless, the PCI score should be evaluated in a critical manner. Even a low PCI can have tumour residing in adverse anatomical locations which prohibit complete resection, e.g. in case of diffuse infiltration of the small bowel serosa. Completeness of cytoreduction should therefore be evaluated separately from PCI. Multiple papers confirm that a low PCI is a beneficial factor in determination of survival [22–24]. A low PCI could theoretically indicate timely diagnosis of GCC, when tumour load is still surgically manageable, and thus lead to a better OS after resection. In opposition, complete cytoreduction does not have a significant impact on survival in two publications [24, 57]. This could be due to technical failure, where tumour is missed in hard to reach locations (e.g. the porta hepatis, base of the bladder), or because of the small number of patients in the available series [23]. One study states that a complete cytoreduction corresponds to a longer median OS of 29 months, versus 17 months in case of incomplete resection, without reaching statistical significance [57]. Neither PCI nor complete cytoreduction are considered significant determinants for survival according to one article [58]. Negative lymph nodes relate to a longer OS (29.2 months), compared to patients with positive lymph node involvement (OS 10.2 months) (p = 0.002) [57].

Grading classifications have been but forward in order to predict survival following CRS-HIPEC. Shyu examines whether the WHO 5th edition classification system could predict survival post-CRS-HIPEC for peritoneally disseminated GCC. The WHO classification applies the pathological criteria defined by Yozu (Table 1). Eleven percent of patients are classified grade 1 (G1), 16% grade 2 (G2) and 73% grade 3 (G3). The PCI does not significantly differ between the WHO grades. G1–2 has a median survival of 98 months, and a 5Y OS of 54%, whereas G3 has a median survival of 33 months and a 5Y OS of 23%. Patients with a G3 disease have a 2.81-fold higher risk of death compared to G1–2 disease (p = 0.04) [25]. A total of 100% of G1 patients have a complete cytoreduction, compared to 60% and 50% in the G2 and G3 groups respectively (p = 0.04) [25]. The Tang classification significantly concurs to the respective gradations. Of G1 patients, 57% are classified Tang A and 43% Tang B. A total of 100% of G2 patients are classified Tang B. Of G3 patients, 26% are classified Tang B and 74% Tang C (p < 0.001) [25]. A comparison of the discussed pathological systems is disclosed in Table 1. The criteria between these classification systems indeed overlap to some extent. Where the Tang classification emphasises presence of signet cells and adenocarcinoma, these features are also present in the ‘high grade characteristics’ applied in the WHO classification [12, 25, 51]. The Tang score significantly predicts survival after CRS-HIPEC: survival rates are 59 months, 22 months and 13 months for Tang A, B and C classification respectively (p = 0.005) [26]. This could correspond to the proportion of adenocarcinoma withheld on the specimen, which is classified by the Tang classification, and thus correlate with the metastatic potential of the GCC. Tang B–C classification, as well as WHO G2–G3 classification, could thus predict an ominous prognosis. One study published an exceptionally low median PCI score of 0 (range 0–39) in its population, and the PCI does not differ over the Tang classification groups (12% Tang A, 67% Tang B, 21% Tang C) [26]. Tang A classification is associated with 100% complete cytoreduction, versus only 32% and 29% with complete cytoreduction in Tang B and C classifications, respectively [26]. The median DFS is 21 months for patients with a complete cytoreduction, opposed to 11 months for patients with an incomplete cytoreduction [26]. The authors stratify survival based on the Tang classification, and find Tang A GCC to have better survival outcomes than Tang B and C [26]. Though this data is not significant, it suggests that the Tang classification could predict success of cytoreduction, separate from the PCI.

Prophylactic Treatment of GCC with CRS-HIPEC

There is a lack of strong data for the use prophylactic HIPEC. Madsen e.a. applied CRS-HIPEC as a prophylactic treatment in eight patients with risk factor for developing peritoneal dissemination in a prospective standardised way nationwide [21]. These risk factors consisted of perforation of the appendix, presence of a peri-appendicular abscess or a positive resection margin (R1, < 1 mm) [21]. Patients aged above 75 years were excluded [21]. A total of 100% of patients in both groups had a complete cytoreduction [21]. Interestingly, lymph node involvement was present in one hemicolectomy specimen [21]. The reported morbidity rate was 40% [21]. OS was 100% after 8 years in patients who received prophylactic CRS-HIPEC, compared to 30% in the group with PM [21]. Similarly, other trials for colorectal cancer have examined the adjuvant treatment value of HIPEC in prevention of PM in case of T4 or perforated cancer, but could not prove any benefit in DFS [59, 60].

The patient in our case report underwent prophylactic treatment with CRS-HIPEC for GCC due to intraoperative rupture of the appendix, corresponding to the inclusion criteria used by Madsen, excepting the patient’s age [21]. The patient was 77 years old, without comorbidities, and was deemed fit enough for prophylactic treatment of possible peritoneal disease. Sadly, he encountered grade IV complications (Clavien-Dindo classification) of the procedure for which he chose palliative care [61]. CRS-HIPEC remains a high morbidity surgery with morbidity rates varying between 19 and 49%, and 30-day mortality rates of 2–3% [62, 63]. In case of prophylactic treatment with CRS-HIPEC, the risk for development of PM (tumour perforation, presence of abscess or positive resection margin) should be weighed heavily, considering the high morbidity and mortality of the procedure. Multidisciplinary team meeting treatment options should be discussed at length and in depth with the patient.

Conclusion

GCC of the appendix is an aggressive type of tumour with potential disseminating characteristics. Several classification systems have been put forward in order to predict survival. GCC can be treated with CRS-HIPEC in a peritoneally disseminated setting. Prophylactic treatment with CRS-HIPEC of GCC in presence of risk factors for development of PM remains controversial and published data is scarce. Through demonstration of a case report, we suggest that prophylactic treatment with CRS-HIPEC should only be considered after thorough patient selection and thorough deliberation, weighting the potential benefits on survival against the possible extensive morbidity of this treatment.

Declarations

Conflict of Interest

The authors declare no competing interests.