Surgical Decision-Making in Familial Adenomatous Polyposis

Abstract

Familial adenomatous polyposis (FAP) is an autosomal dominant disorder affecting patients with germline mutations of the adenomatous polyposis coli (APC) tumor suppressor gene. The surgical treatment of colorectal disease in FAP, which has the goal of colorectal cancer prevention, varies based on both patient and disease factors but can include the following: total colectomy with ileorectal anastomosis, proctocolectomy with stapled or hand-sewn ileal pouch-anal anastomosis, or total proctocolectomy with end ileostomy. The operative options and extent of resection, as well as the use of endoscopy and chemoprevention for the management of polyposis, will be discussed in detail in this article. In addition, commonly debated management decisions related to the treatment of patients with FAP, including the timing of prophylactic colorectal resections for patients with FAP and management of the polyp burden in the rectum, will be discussed. Finally, genotype considerations and the impact of desmoid disease on operative decisions in the setting of FAP will also be reviewed.

Addressing the Polyp Burden in the Rectum

The surgical treatment of colorectal disease in familial adenomatous polyposis (FAP) varies depending on several patient and disease factors but include the following: total colectomy with ileorectal anastomosis (TAC with IRA), proctocolectomy with stapled or hand-sewn ileal pouch-anal anastomosis (TPC with IPAA), or total proctocolectomy with end ileostomy (TPC with EI). When considering the operative approach, a surgeon must consider the degree of involvement of the rectum with polyposis, in addition to other patient factors discussed below. The current American Society for Colon and Rectal Surgeons (ASCRS) guidelines from 2017 propose the inclusion of a proctectomy when a patient has more than 20 rectal adenomas (strong recommendation based on moderate-quality evidence, 1B). ¹

Above and beyond the guidelines, other studies have assessed what might guide surgeons as to the merits of each operative approach noted above, including a 2006 meta-analysis of 12 nonrandomized studies. Factors that favored TAC and leaving the rectum in place included relative rectal sparing from polyposis and the desire to avoid a pelvic dissection (obligatory for the proctectomy portion of a TPC) and the possible resulting infertility or sexual dysfunction from this dissection. On the other hand, factors that support TPC with IPAA or EI in the setting of FAP include the presence of rectal cancer, greater than 20 rectal adenomas, high-grade dysplasia, a large (greater than 3 cm) adenoma, or a severe familial phenotype (greater than 1,000 adenomas). ² When considering the operative approach using decision analysis, Melnitchouk et al found that, overall, an IRA is less technically challenging compared with an IPAA and is associated with better functional outcomes and improved quality of life. Preservation of the rectal reservoir allows for better function with the tradeoff of the cancer risk that remains in the retained rectum. On the other hand, TPC with IPAA remains a more technically demanding operation that can lead to more frequent bowel movements, nocturnal bowel movements, incontinence/nighttime fecal soiling, and a higher potential for sexual dysfunction and/or infertility. ³ Pasquer et al note that patients who underwent an IRA have a significantly lower mean number of stools per day and less frequent fecal incontinence compared with patients who underwent an IPAA. ¹² In summary, TAC with IRA is a technically easier operation to perform with the benefit of improved fecal urgency, urinary continence, and sexual function compared with IPAA. This option should be strongly considered in patients with a limited rectal polyp burden (< 20 polyps), a low-risk genotype, and are able to comply with subsequent surveillance of the remaining rectum after TAC. ⁴

Other factors that can be considered when determining operative approach and rectal preservation in FAP include the patient's general status and comorbidities (e.g., age, functional status, obesity, nutritional status), preoperative fecal continence, presence or risk of desmoid disease, childbearing plan, and the patient's educational, intellectual, and emotional development, as this may affect their ability to follow-up reliably for endoscopic surveillance of the remaining rectum and pouch. ¹ The surveillance of the rectum or the pouch postoperatively includes annual endoscopy and remains mandatory to mitigate rectal cancer risk. In a cohort study by Church et al, the incidence of rectal cancer was found to be 1.6% in patients with less than 20 rectal adenomas compared with 10.8% in patients with greater than 20 adenomas. ⁵ Several other studies showed that if the rectum was preserved, there is a rectal cancer risk of 10 to 25% after 15 to 25 years and increasing to 30% by the age of 60 years old. ^{6 7 8 9}

Recently, it has been suggested that the management of the polyp burden in the rectum in the setting of FAP may also include endoscopic control. The Cleveland Clinic recently reported their experience instituting an aggressive, multistage, snare polypectomy regimen. They report that with this approach they have effectively managed patients with a significant rectal polyp burden and allowed patients to postpone proctectomy for several years in both patients with prior TAC and IRA and those with minimal colonic disease and more involved rectal disease. ¹⁰ Although the follow-up data remains short, on the scale of a few years, this approach has shown sustained control without functional complications for the patient. Another similar study in Japan of just 34 patients concluded that intensive endoscopic surveillance can be used to prevent the development of secondary/metachronous cancer after resection in patients with FAP with either a rectal remnant or ileal pouch. ¹¹ What remains unknown is whether or not the removal of all polyps does, in fact, change the risk of cancer in the retained rectum and does the scarring from the polypectomies increase the chances of the short, flat, subtle cancers like those seen in chronic ulcerative colitis. ¹⁰ As a result, the combination of aggressive endoscopic treatment and extended rectal preservation appears to be a safe alternative to IPAA, but conclusions would benefit from longer-term follow-up. ¹²

Finally, patients with a retained rectum postoperatively after TAC with IRA for FAP might benefit from chemoprevention with either sulindac or celecoxib (see below for potential concerns related to their use) after an individual risk/benefit assessment with the patient. While this is recommended in the ASCRS guidelines, this is noted to be a weak recommendation based on high-quality evidence (2A). ¹ The three main roles of chemoprevention in patients with FAP that have been described are: to delay prophylactic colectomy, to prevent cancer development in the retained rectum of patients after colectomy with IRA, and to prevent cancer development in the upper gastrointestinal tract, especially the duodenum. Nonsteroidal anti-inflammatory drugs (NSAIDs) inhibit cyclooxygenase, one enzyme needed for the conversion of arachidonic acid to prostaglandins and other eicosanoids. This is important in the adenoma-carcinoma sequence because prostaglandins play a key role in cell adhesion, apoptosis inhibition, and promotion of angiogenesis. ¹³ Elevated prostaglandin levels are found in many premalignant and malignant lesions including colorectal adenomas and adenocarcinomas. ¹⁴ Therefore, by inhibiting prostaglandins, this progression is hindered and polyp regression ensues. Three randomized, double-blind, placebo-controlled studies have shown this effect of sulindac, but importantly, all note that this effect is transient and reversible. ^{15 16 17} Therefore, sulindac therapy cannot replace a prophylactic resection as the primary treatment for FAP but may be useful in postponing resection, especially for young patients or those with attenuated FAP (AFAP). ¹⁸

Notably, while there are some benefits in FAP, there are also risks related to the use of these medications. Sulindac has a higher risk of gastritis in patients who are greater than 65 years old, have a history of peptic ulcer disease, or require concurrent use of aspirin, corticosteroids, or anticoagulants for other medical conditions. In these settings, the patient should be given a proton-pump inhibitor prophylactically. ¹ The risk of celecoxib has been noted to be a dose-dependent increase in deaths from cardiovascular causes, nonfatal myocardial infarction, stroke, and heart failure, ¹³ and therefore the use in patients with FAP has been limited. The ASCRS guidelines stress the importance that even with chemoprevention, patients with a retained rectum must continue to undergo endoscopic surveillance for dysplasia or cancer. ¹ Overall, while there is data to suggest that sulindac is the most potent polyp regression medication, if FAP patients or their providers are interested in chemoprevention, referral to an expert center is recommended. ^{16 19}

Timing of Surgery in Classic FAP Presentation and in Attenuated Phenotype

Classic FAP is defined as developing hundreds to thousands of adenomatous colorectal polyps starting in adolescence, which is in comparison to AFAP defined as developing 100 or fewer polyps through the patient's life, often occurring one to two decades later in life than classic FAP. ²⁰ Despite accepted differences in the phenotypes of these two entities, there is not a consensus on the precise timing of prophylactic resection in patients with FAP and AFAP.

There are four organizations that have published guidelines regarding colorectal resections for patients with FAP. The American College of Gastroenterology guidelines in 2015 do not delineate the exact patient age recommended for resection, but they do state that the absolute indications for immediate colectomy in FAP and AFAP include documented or suspected cancer or significant symptoms, which could occur at any age depending on the phenotype. Relative operative indications include the presence of multiple adenomas greater than 6 mm, significant increase in adenoma number between surveillance endoscopies, and inability to adequately survey the colon because of numerous diminutive polyps, regardless of the age that these issues occur. ²¹ The second group, the International Society for Gastrointestinal Hereditary Tumors (InSIGHT), proposed guidelines in 2016 ( Table 1 ) which included a detailed staging system for polyposis of the colon that they hoped would be used in surgical decision-making. The InSIGHT system was developed through expert providers who were used to test the concordance of scoring endoscopy videos and assigning a polyposis stage and selecting an intervention for that stage. There was a statistically significant degree of agreement in both the staging of polyp burden and the selected intervention. Nearly all reviewers agreed that the development of a staging system for colorectal polyposis is necessary to standardize FAP management across institutions. They also agreed that based on the number and size of polyps, there are four progressive polyposis stages which in turn impact recommended interventions, including the timing of operation. For those patients with stage three and stage four polyposis, providers should consider short-interval repeat endoscopy but only when there is a clear indication to delay resection. Otherwise, TAC or TPC are recommended within 3 months to 1 year of the classification of this advanced stage. ²² The third organization, the ASCRS, recommended in 2017 that in FAP patients with fewer than 20 rectal adenomas, the operation of choice would be a TAC with IRA, while those with greater than 20 rectal adenomas should undergo TPC with IPAA. Despite these cutoffs for extent of resection, the timing, per these recommendations, should be individualized to the patient. ¹ Finally, other national guidelines suggest that prophylactic TPC in patients less than 18 years old is not routinely recommended because cancer prior to this age is very rare. However, extensive resection should be considered in patients less than 18 years old with severe polyposis or a family history of early cancer or severe genotype, and patients with AFAP with fewer than 20 nonadvanced adenomas can delay resection until 21 years old or older given the delayed nature and lower degree of presentation versus classic FAP.

Table 1. 2016 InSIGHT staging system for lower GI tract polyposis ²² .

Stage

Intervention Options

0 = < 20 polyps, all < 5 mm 1 a  = 20–200 polyps, most < 5 mm, none > 1 cm 2 a  = 200–500 polyps, < 10 that are > 1 cm 3 a  = 500–1,000 polyps, or any number if there are 10–50 that are > 1 cm and amenable to complete polypectomy 4 = > 1,000 polyps and/or any polyps grown to confluence and not amenable to simple polypectomy, ANY invasive cancer

A = repeat colonoscopy in 2 years, need baseline biopsy to confirm histology, polyp removal is discretionary (not clearly indicated) B = repeat colonoscopy in 1 year, some would consider colectomy, especially when polyp count is high C = repeat colonoscopy in 1 year, polypectomy preferred, removal of large polyps clearly necessary when done to postpone surgery, alternative would be to consider surgery D = repeat colonoscopy 6–12 months or consider colectomy, removal of large number of large polyps defensible, but only when clear indication to delay surgery E = colectomy, proctocolectomy clearly indicated within 3 months–1 year, any decision to delay surgery must be highly individualized and based on compelling circumstances

Abbreviations: GI, gastrointestinal; InSIGHT, International Society for Gastrointestinal Hereditary Tumors.

^aPresence of high-grade dysplasia upstages the patient to stage 4.

Outside of the above guidelines, several studies have investigated the timing of investigation and prophylactic operation in patients with FAP and AFAP. An analysis of the Finnish FAP registry concluded that very early diagnosis at less than 15 years old is not necessary for asymptomatic patients due to the very low risk of colon cancer in this patient age group. ²³ In this study, the authors note that increasing polyp count is the most important factor determining the timing of operation. ²⁴ Knudsen et al suggest that while prophylactic colectomy in classic FAP is well established and usually recommended at 20 to 25 years of age, patients with AFAP may be safely managed with yearly endoscopy and polypectomy (thus delaying need for surgical intervention) depending on the size and number of polyps as well as evidence of dysplasia on surveillance endoscopy. ²⁵ Sokic-Milutinovic also concluded that for patients with AFAP, colectomy may never be needed, and asymptomatic patients can be managed with annual endoscopic surveillance and subsequent polypectomy of detected lesions unless high-grade dysplasia or cancer is detected. ²⁶ Overall, even if chemoprevention or endoscopic interventions are utilized to attempt to delay resection in FAP, interventions must be on a case-by-case basis; and although definitive ages for prophylactic TAC or TPC in the setting of classic FAP and AFAP are not clearly defined, indications that remain noncontested are in patients with high-grade dysplasia or cancer at the time of surveillance endoscopy or severe symptoms.

Genotype Considerations in Surgery

Germline mutations of the adenomatous polyposis coli ( APC ) tumor suppressor gene located on chromosome 5q21 are found in 80% if patients with classic FAP. ^{1 27} Further investigation into the genotype differences in FAP and AFAP have found that mutations in three specific regions of the APC gene are associated with AFAP (in addition to the known autosomal recessive, biallelic MUTYH gene mutations found in a subset of patients with AFAP ²⁷ ). Anele et al report that variants in the 5′ region of codon 233 of APC are the most common and that this mutation results in the production of a truncated protein, which is unable to dimerize with wild-type APC proteins and therefore cannot interfere with its function, thereby producing an attenuated phenotype. Another variant is the 3′ variant of codon 1595 which results in the production of an almost intact and functional protein. Finally, variants in the alternative splice region of exon 9 suggest that the allele with this pathogenic variant produces an APC protein with normal tumor suppressor activity. Because these three variants have a more limited, if not any, effect on the function of the APC gene, they produce an attenuated phenotype. ²⁰ Lal and Gallinger also studied the genotype of FAP and found that mutations proximal to codon 1249 of APC are associated with less than 1,000 polyps, those between codon 1250 and 1330 are associated with greater than 5,000 polyps, and those distal to codon 1465 are associated with fewer polyps but remain in the FAP phenotype. Additionally, they found that mutations located at the extreme 5′ end of the gene (exons 3 and 4) and distal to codon 1578 at the 3′ end of the gene are associated with a mild and variable AFAP phenotype. ²⁸ Regardless of the genotype, however, the phenotype and patient characteristics, especially in AFAP, have the greatest impact on surgical decision-making, especially since 80 to 90% of these patients may have no gene mutation found (with < 100 adenomas found). ²⁷

Some studies have underscored that genotype-phenotype correlations in FAP may impact treatment choices. Two studies by Sinha et al and Church et al showed that genotype and phenotype have been documented as independent predictors of advanced adenomas and the need for proctectomy following TAC and IRA in FAP. Sinha et al note that the independent predictors of need for proctectomy include the following: colonic adenoma count greater than 500, APC pathogenic variant in the mutation cluster region or approximately codons 1250 to 1450, and age under 25 at the time of resection. ^{5 29} Another study looking at patients in the Hereditary Colorectal Tumor Registry from the National Cancer Institute in Milan noted the risk of rectal cancer with FAP. Patients with APC mutations beyond codon 1275 were found to have a high risk of developing rectal cancer in the retained rectum while other studies have shown that patients with a mutation at codon 1309 should undergo TPC (as opposed to TAC) at the time of diagnosis because they have more extensive rectal disease. ^{30 31} Finally, a study of the Manchester Polyposis Registry grouped patients into two groups: those with an APC mutation between codons 1249 and 1549 and those with a mutation between codons 0 and 178 or 312 and 412. The former group was found to have polyposis at an early age and a worse survival while those in the latter group developed polyposis later in life and had improved survival, a difference that had not been previously documented. ³² These studies suggest that genotype may be helpful when managing FAP patients.

Despite the genotype/phenotype correlations in some studies, other reports suggest that the location of the APC gene mutation is not the sole determinant of phenotype and other genetic and environmental factors play a role. Therefore, genotype should be used as an adjunct , just one “piece of the puzzle,” and not the sole decision-driver when determining the timing and extent of resection in FAP patients. ^{28 32} Some centers have instituted a selective approach to operative intervention using colonic and rectal polyp burden (in addition to other factors, as noted above), together with genotype, to select patients with severe disease who would benefit from upfront TPC with EI or IPAA instead of TAC with IRA or nonoperative endoscopic surveillance, but this is institutionally and patient-dependent and not yet a standardized approach. ³³

Desmoid Disease and the Need for Proctectomy

Desmoid disease or desmoid tumors (DTs) can cause considerable morbidity and mortality for patients with FAP. Desmoids are slow-growing, nonmetastasizing, but highly invasive tumors that occur in both the intra- and extra-abdominal compartments. ³⁴ The incidence of DT in FAP is estimated to be 10 to 25%. ³⁵ The majority of FAP DTs are intra-abdominal and usually located in the mesentery or retroperitoneal tissue. They originate from the fibro-aponeurotic tissue as plaque-like desmoid precursor lesions or mesenteric fibrosis, and they can develop into multifocal, infiltrative masses. ³⁶ DTs represent a significant challenge in FAP. Bowel obstruction, bowel perforation, fistula formation, and ureteral compression are known sequelae of DT, among others. ^{37 38} Moreover, they remain the principal cause of death following preventative proctocolectomy in FAP patients. ³⁴ As of now, there is a relative paucity of safe and reliably effective treatments. Management is typically multimodal to include NSAIDs (such as sulindac), hormonal therapy with tamoxifen, and Adriamycin-based chemotherapy. Surgical intervention for DTs carries high rates of recurrence and often necessitates substantial small bowel resections leading to considerable morbidity and, as such, is avoided if possible. ³⁴

Risk factors for clinically significant desmoid disease include female gender, positive family history, certain genetic mutations such as a germline APC mutation near 3′ of codon 1400, and surgical trauma. ^{39 40} Given the potential for de novo desmoid formation and high rates of dense adhesions from existing DT, much attention has been paid to initial choice of prophylactic operation for FAP, with the decision significantly predicated on the degree and severity of rectal involvement with the disease, as discussed above. Performing TPC with IPAA for FAP at the index operation not only has the benefit of removing nearly all of the high-risk colorectal epithelium, but it also avoids a situation that may occur with rectal-sparing operations in which the rectum might become unresectable in the future due to desmoid burden in patients at high risk of desmoid disease. ³⁹ However, Church et al showed that for FAP patients who had a prior TAC with IRA and subsequently needed a proctectomy, it was always surgically possible to remove the rectum despite the presence of intra-abdominal desmoid disease. ³⁹ They emphasized that while continuity could not be restored in 8% of patients due to disease in the ileal mesentery, rectal entrapment prohibiting future proctectomy was not an issue. This suggests that the presence of desmoid disease alone should not preclude TAC and IRA. In fact, recent data has shown regression of rectal polyp burden after IRA ¹³ suggesting that expectant management might perhaps benefit younger patients for whom personal and professional goals might otherwise be affected by pouch dysfunction or an unintended stoma, even in the setting of desmoid disease. ³⁹

With regard to the approach to surgical treatment of FAP, many studies have examined the effects of type of procedure, approach (i.e., open vs. minimally invasive), and timing of operation on desmoid development. A systematic review by Aelvoet et al showed no significant difference in desmoid disease after TAC with IRA versus TPC with IPAA. ⁴⁰ This is similar to other reports with comparable rates of postoperative de novo desmoid formation at 12.6% after TPC with IPAA, 13% after TAC with IRA, and 13.5% after partial colectomy, respectively. ⁴¹ Aelvoet et al also observed no difference in desmoid disease between the open and laparoscopic approaches, nor an impact of age at the time of operation on desmoid development. This was recapitulated in a study by Campos et al who demonstrated no association between extent of resection (TPC vs. TAC) and approach (laparoscopic vs. open) with risk of DT formation. ⁴² Importantly, a meta-analysis of publications between 1989 and 2019 by Xie et al with 1,072 patients found similar DT incidence after IPAA versus IRA (11.8% vs. 9.5%, odds ratio 0.95, p  = 0.85). ³⁸ Ultimately, the extent of operation, approach, and timing of colectomy for FAP patients should be based on neoplastic and other patient factors and not desmoid burden (although the presence of DTs may require a more experienced surgeon or treatment center) ( Table 2 ).

Table 2. Guidelines for the colorectal surgical management of FAP.

	TAC + IRA	TPC + IPAA	TPC + EI
Indications	- To eradicate colon polyposis and colon cancer risk while preserving the rectum and maintaining intestinal continuity (i.e., avoiding an ostomy)	- To eradicate colon and rectal polyposis or severe rectal disease after TAC + IRA - High-grade dysplasia or rectal cancer may be an indication in some cases	- Unable to undergo IPAA due to patient factors (e.g., incontinent, morbidly obese) or low rectal cancer needing APR - Patients with IPAA failure or severe pouch polyposis requiring pouch excision - Unable to surveil the rectum/pouch (e.g., due to patient access, preference)
Contraindications	- Rectal polyposis (see text, can consider chemoprevention) - Unable to surveil the rectum (e.g., due to patient access, preference)	- Patient factors (e.g., perianal/small bowel Crohn's, baseline incontinence, morbid obesity, etc.) - Received XRT for rectal cancer (or other pelvic malignancies) - Patients with desmoid disease preventing the surgeon from performing IPAA - Unable to surveil the pouch/rectal cuff (e.g., due to patient access, preference)	Essentially none
Advantages	- Avoids an ostomy - Low morbidity and mortality, especially versus IPAA (less sexual/bladder dysfunction, infertility risk, etc.) by avoiding pelvic dissection - Able to perform in the setting of desmoid disease	- Avoid permanent ostomy (often requires temporary one, however) - Decreased rectal cancer risk	- Removes all risk of colon and rectal cancer with one operation - Avoid anastomotic leak complication risk
Disadvantages	- Rectal cancer risk persists, requires surveillance - Quality of life can be affected by number of bowel movements per day	- Technically challenging operation - Often requires temporary stoma - Highly variable functional outcomes - Increased morbidity risk (infertility and/or fecundity, sexual/bladder dysfunction, incontinence risk) - Anastomotic leak risk - Does not eliminate rectal cancer risk, requires surveillance	- Increased morbidity risk (infertility and/or fecundity, sexual/bladder dysfunction) - Permanent stoma

Abbreviations: APR, abdominoperineal resection; FAP, familial adenomatous polyposis; TAC + IRA, total abdominal colectomy with ileorectal anastomosis; TPC + EI, total proctocolectomy with end ileostomy; TPC + IPAA, total proctocolectomy with ileal pouch anal anastomosis; XRT, radiation therapy.

Polyp Regression after Total Colectomy in Patients with IRA

Spontaneous regression or resolution of rectal polyposis after TAC with IRA has been reported in the literature. In 1983, Watne et al were the first to show a reduction in polyp count in 15 out of 17 patients (88.2%) 3 months after rectum-sparing resection. ⁴³ In 1988, Feinberg et al reported a spontaneous resolution of rectal polyposis in 64% of FAP patients following TAC with IRA within a median of 6 months postoperatively. ⁴⁴ While 38% of patients had complete resolution and 25% of patients had a partial response, a total of 55% of patients redeveloped polyps at a median of 6.8 years. Theories to explain the regression of rectal polyps after TAC include that the removal of the colon induces changes in fecal microbiota and bile acid content which affects rectal mucosal proliferation and polyp development. Additionally, it is possible that the increased alkalinity in the ileal content “destroys” polyps in the rectum. On the other hand, other studies were unable to find convincing evidence of rectal polyp regression after TAC with IRA, and 97% of patients were instead found to have polyp progression over time, albeit at a slow rate. ⁴⁵ Overall, patients may have spontaneous regression of their rectal polyp burden, but this may be transient, and therefore all patients must continue to undergo endoscopic surveillance after TAC with IRA.