Hereditary GI Cancer Syndromes: Hamartomatous Polyposis and Diffuse Gastric Cancer

0:00Welcome to Board Pearls. This is episode four of five of the Hereditary GI Cancer Syndromes chapter, in the Special Populations and Acute or Supportive Care module. In this episode we cover the hamartomatous polyposis syndromes and hereditary diffuse gastric cancer: Peutz-Jeghers syndrome with STK11 and its mucocutaneous pigmentation, the PTEN hamartoma spectrum including Cowden, and CDH1-driven hereditary diffuse gastric cancer where prophylactic total gastrectomy is the standard.

0:30The last episode ran the adenomatous polyposis syndromes, where the histology is adenoma and the dominant risk is colorectal cancer. These syndromes shift the histology to hamartoma and the dominant risk somewhere else. Peutz-Jeghers shifts the gene, the histology, and the dominant complication. The gene is STK11 on chromosome nineteen-p, also called LKB1. It encodes a serine-threonine kinase that suppresses mTOR signaling. When it goes, hamartomatous polyps grow throughout the GI tract and the carrier accumulates multi-organ cancer risk that runs into the third and fourth decades. Penetrance exceeds ninety percent by age thirty. Roughly eighty percent of clinically defined patients have a positive family history; the remaining twenty percent are de novo mutations. Recognition is on the mucocutaneous pigmentation. Melanin macules develop in early childhood across the lips, perioral skin, buccal mucosa, fingertips, palms, soles, and perianal region. Cutaneous pigmentation tends to fade at puberty but the buccal pigmentation persists, which makes the buccal site the most reliable adult finding. Three features separate Peutz-Jeghers pigmentation from common freckles. Freckles are absent at birth. Freckles almost never appear on the buccal mucosa. And freckles are rare in the perinasal and perioral distribution. So a twenty-two-year-old with bowel obstruction whose buccal mucosa is pigmented is Peutz-Jeghers regardless of how the family history reads.

2:04The polyps themselves are hamartomas with a characteristic arborizing smooth-muscle pattern. Smooth-muscle bands extend from the muscularis mucosae up into the lamina propria, and that arborizing pattern is what distinguishes Peutz-Jeghers polyps from juvenile polyps, which show cystically dilated mucus-filled crypts, and from PTEN-associated hamartomas. The polyps concentrate in the small bowel. The jejunum and ileum together account for sixty to ninety percent of polyps, followed by stomach and colon. And the small bowel is where the trouble is, because small bowel polyps become intussusception lead points. The first presentation is often in childhood or early adulthood with bowel obstruction or hematochezia, frequently requiring emergent laparotomy with segmental small bowel resection. Recurrent emergent resection is the dominant cause of late short bowel syndrome in this population. So the entire surveillance program exists to remove polyps before they reach intussusception size.

3:02This is the most testable rule in the syndrome. Resect any small bowel polyp greater than one centimeter. Resect any symptomatic polyp regardless of size. And resect any polyp showing rapid growth between studies, because growth velocity predicts impending intussusception better than absolute size. Resection is by double-balloon enteroscopy, intraoperative enteroscopy, or surgical polypectomy depending on accessibility. Stomach and colon polyps come out when they exceed five millimeters, because endoscopic access is straightforward and the cancer risk per polyp is similar. Lifetime small bowel adenocarcinoma risk is dramatically elevated in relative terms, around a five-hundred-fold relative risk, but the absolute lifetime risk is around thirteen percent. So cancer prevention is real but secondary. Obstruction prevention is primary, and that is what the resection threshold is calibrated to.

3:55Surveillance starts young. Baseline EGD and colonoscopy at age eight to ten, or earlier if the child is symptomatic. If polyps are found, the studies repeat every two to three years. If no polyps are found at baseline, surveillance pauses and resumes at age eighteen at the same interval. Small bowel surveillance runs in parallel from age eight to ten with capsule endoscopy or MR enterography every one to three years. Both modalities avoid radiation, which matters across decades of repeated imaging in a pediatric and adolescent population. CT enterography is the wrong primary surveillance modality precisely because of that cumulative radiation burden.

4:36The multi-organ cancer surveillance layers on from the mid-twenties. The starting ages are not arbitrary. They are calibrated to organ-specific tumor onset under STK11 loss. Annual breast MRI plus mammography starts at age twenty-five, because STK11 loss accelerates breast cancer onset to the mid-twenties on a timeline similar to BRCA-one and BRCA-two carriers. Lifetime female breast cancer risk in Peutz-Jeghers reaches forty-five to fifty percent by age seventy, which is in the BRCA range. Pelvic examination plus Pap plus transvaginal ultrasound starts in late adolescence to capture the ovarian and cervical tumors. Pancreatic surveillance with annual EUS or pancreatic-protocol MRI starts at age thirty to thirty-five, because pancreatic tumors emerge in the thirties and forties. Testicular examination runs from infancy in boys, because Sertoli cell tumors present in early childhood with precocious puberty through tumor estrogen secretion.

5:33Three Peutz-Jeghers cancers are worth memorizing because each is a recognition stem on its own. Sex cord tumor with annular tubules, SCTAT, of the ovary is essentially pathognomonic for Peutz-Jeghers. It produces estrogen, so the presentation is endometrial hyperplasia, irregular menses, breast tenderness, and a complex adnexal mass on imaging, typically with normal CA one-twenty-five because SCTAT is not epithelial. The Peutz-Jeghers SCTAT is typically multifocal and bilateral, which distinguishes it from sporadic SCTAT. Minimal-deviation adenocarcinoma of the cervix, also called adenoma malignum, is the cervical counterpart. And Sertoli cell tumor of the testis secretes estrogen and presents with gynecomastia or precocious puberty in prepubertal boys. Identification of any of these tumors in an unrelated workup should trigger the question of underlying Peutz-Jeghers. One more rule worth holding. Pancreatic surveillance in Peutz-Jeghers qualifies on the gene alone, unlike BRCA, ATM, PALB2, and Lynch, where surveillance requires an additional family-history modifier. The mechanism is that STK11 lifetime pancreatic cancer risk is high enough by itself to clear the absolute-risk threshold.

6:56The PTEN hamartoma tumor syndrome spectrum changes the shape of the problem again. Cowden is the adult presentation. Bannayan-Riley-Ruvalcaba is the pediatric presentation. Both come from germline mutations in PTEN on chromosome ten-q, and PTEN regulates the PI3K-AKT-mTOR proliferative pathway. Loss of PTEN removes the constitutive brake on PI3K signaling. The clinically important consequence is that the cancer burden tracks PI3K activity in each tissue. Breast epithelium, thyroid follicular epithelium, endometrium, and renal tubular epithelium all have high baseline PI3K signaling. They are the tissues that accumulate cancer when the brake is lost. Colon and lung have lower baseline PI3K activity. They are at lower risk despite carrying the same germline defect. So Cowden is dominantly a breast and thyroid syndrome, with significant endometrial and renal risk, and the GI piece is recognition-driven rather than cancer-burden-driven.

7:57Recognition lives on physical examination before any imaging is ordered, which is unusual in this chapter. Macrocephaly with dolichocephaly, head circumference well above the ninety-seventh percentile. Trichilemmomas of the face, small skin-colored papillomatous papules clustered around the mouth, nose, and eyes, present in over ninety percent of carriers. Cobblestone papillomatosis of the gingiva, buccal mucosa, and tongue. Acral keratoses of the palms and soles. And Lhermitte-Duclos disease, a cerebellar dysplastic gangliocytoma that is pathognomonic in adults and produces ataxia and a striated cerebellar pattern on MRI.

8:38The GI signature is two endoscopic findings. On EGD, diffuse glycogenic acanthosis presents as small whitish plaques throughout the esophagus, and biopsy shows benign squamous epithelium loaded with cytoplasmic glycogen. Glycogenic acanthosis occurs sporadically too, but a dense and diffuse pattern in a young patient with macrocephaly is a Cowden marker. On colonoscopy, Cowden produces mixed-histology polyposis. Hamartomatous, juvenile-type, hyperplastic, ganglioneuromatous, lipomatous, and inflammatory polyps coexist in the same patient. That mixture is itself the clue. When the histology of a single patient's polyps does not fit one category, the question becomes which underlying syndrome. PTEN is high on the answer list, especially with macrocephaly and trichilemmomas on the same exam.

9:29The surveillance program follows the cancer-burden distribution. Annual thyroid ultrasound starts at the time of diagnosis, and from age seven in pediatric carriers, because thyroid nodules emerge in childhood under unrestrained PI3K signaling. Thyroid cancers in Cowden are predominantly follicular and papillary, rarely medullary, so calcitonin is not the surveillance test and prophylactic thyroidectomy is not the strategy. Annual mammography plus tomosynthesis plus breast MRI starts in the early thirties, with clinical breast exam from age twenty-five, because breast epithelial transformation begins early under PTEN loss. Endometrial sampling plus transvaginal ultrasound in postmenopausal carriers starts at age thirty-five. Renal imaging, preferably MRI to avoid annual CT radiation, starts at age forty, because renal cell carcinoma rarely presents before then. Colonoscopy every three to five years starts at age thirty-five, with shorter intervals when polyps are found, and upper endoscopy on the same schedule. The contiguous PTEN-BMPR1A deletion on chromosome ten-q is testable as the explanation for why an occasional pediatric juvenile polyposis patient also satisfies Cowden criteria. The deletion takes out both genes and produces an unusual juvenile polyposis of infancy with overlapping Cowden features.

10:56That brings us to hereditary diffuse gastric cancer, where the algorithm flips harder than anywhere else in this chapter. The gene is CDH1, which encodes E-cadherin, on chromosome sixteen-q. CDH1 accounts for thirty to fifty percent of clinically defined HDGC. CTNNA1, which encodes the alpha-E-catenin that binds E-cadherin, accounts for a smaller subset. The mechanism explains the management. E-cadherin loss disrupts cell-cell adhesion in gastric epithelium. Poorly cohesive signet-ring cells start growing in the gastric wall without forming glands. The carrier accumulates multifocal microscopic foci beneath grossly intact mucosa over years to decades.

11:42Why this matters is that endoscopic surveillance fails. Standard biopsies sample surface epithelium. The disease lives deeper, scattered as multiple small foci, invisible on white-light endoscopy until the disease is advanced. Even meticulous random biopsy at high count misses cases. The validation comes from the surgical specimens. When confirmed CDH1 carriers undergo prophylactic total gastrectomy and the entire stomach is sectioned at thin intervals, the majority of those stomachs contain foci of intramucosal signet-ring cell carcinoma. The preoperative EGD was unremarkable in many of those cases. The cancer was there. Surveillance could not see it. So the recommendation for confirmed CDH1 carriers is prophylactic total gastrectomy with Roux-en-Y reconstruction, typically between ages twenty and thirty, after multidisciplinary counseling on the nutritional, psychological, and quality-of-life consequences. Lifelong B-twelve and iron supplementation, dumping risk, weight loss, and altered eating patterns are all part of the conversation. This is the central testable decision in the syndrome, and it is one of the few places in gastroenterology where prophylactic removal of a normal-appearing organ is the standard of care.

12:59The patients who decline surgery or who are waiting for it get the Cambridge protocol. Annual EGD with at least thirty to fifty random biopsies from all gastric anatomic regions, antrum, body, fundus, cardia, and the transitional zones. It is the surveillance fallback, not a substitute for surgery. Patients are counseled explicitly that surveillance failure is documented even with rigorous protocol adherence. Median age at gastric cancer diagnosis under surveillance is around thirty-eight years, which sets the timing pressure for the gastrectomy decision.

13:34The female carrier program adds a second organ that surveillance can address. Lifetime risk of lobular breast cancer in female CDH1 carriers runs around forty to fifty-five percent. Lobular breast cancer is histologically distinct from ductal cancer, with a diffuse architecture that mammography detects less reliably, which is why the program is annual breast MRI plus mammography starting at age thirty. Risk-reducing bilateral mastectomy is an option after counseling. CTNNA1 carriers get similar counseling, though the gastric cancer penetrance estimate is lower and the breast cancer association is less well established. The testing trigger for CDH1 itself runs through the IGCLC criteria. Testing is offered for two or more gastric cancers in a family with at least one diffuse type. Testing is offered for one diffuse gastric cancer under age forty. Testing is offered for a family history of diffuse gastric plus lobular breast cancer with one diagnosis under fifty. Testing is offered for bilateral lobular breast cancer under seventy. Testing is also offered for early-onset signet-ring gastric cancer in any patient and for in situ signet-ring carcinoma on biopsy. Contemporary panels include CDH1 on most multi-gene cancer panels regardless of pre-test probability, because the result has direct surgical implications.

14:55One differential earns its place in the HDGC vignette. Gastric Adenocarcinoma and Proximal Polyposis of the Stomach, GAPPS. GAPPS is a separate autosomal dominant syndrome caused by point mutations in the promoter one-B region of APC. That is the same gene as FAP, but a different regulatory region. The phenotype is fundic gland polyposis carpeting the proximal stomach with low-grade and high-grade dysplasia, antral sparing, and no colorectal or duodenal polyposis. So GAPPS is anatomically segmental and stays in the proximal stomach. HDGC produces multifocal submucosal disease without polyposis. FAP carries duodenal and colorectal polyposis that GAPPS does not. The board distinction is anatomic. Where is the polyposis, and what other GI sites are involved. Management of GAPPS resembles HDGC. Surveillance endoscopy with biopsy for dysplasia, with prophylactic total gastrectomy considered for high polyp burden, multifocal dysplasia, or progression.

15:59So hold the hamartomatous and diffuse gastric syndromes by gene and by dominant risk. STK11 drives Peutz-Jeghers, with arborizing smooth-muscle hamartomas predominantly in the small bowel, persistent buccal pigmentation, and a resection threshold of one centimeter for small bowel polyps because intussusception, not cancer, is the dominant near-term risk. PTEN drives Cowden, with the cancer burden concentrated in tissues with high baseline PI3K signaling, breast and thyroid above all, and the GI role is recognition through mixed-histology polyposis and diffuse glycogenic acanthosis. CDH1 drives hereditary diffuse gastric cancer, where multifocal submucosal signet-ring disease beneath intact mucosa makes endoscopic surveillance unreliable, so prophylactic total gastrectomy between ages twenty and thirty is the standard, with annual breast MRI plus mammography from age thirty for the female carrier. The thread across all of them is the same: the gene dictates the histology, the histology dictates the natural history, and the natural history dictates whether the right answer is surveillance, polyp-by-polyp resection, or removal of the whole organ.

17:07The next episode closes the chapter with hereditary pancreatic cancer and the CAPS protocol, plus the multi-organ coordination of cascade testing and genetic counseling, including the absolute-risk gating rule, the GINA limitation, and the VUS actionability rule.

17:22For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode four of five of chapter thirty six, and I'll see you in the next one.