Human Molecular Genetics, 2021, Vol. 30, No. 14 1273–1282 doi: 10.1093/hmg/ddab094 Advance Access Publication Date: 2 April 2021 General Article GENERAL ARTICLE mTOR inhibitors reduce enteropathy, intestinal bleeding and colectomy rate in patients with juvenile polyposis of infancy with PTEN-BMPR1A deletion Henry Taylor1,† , Dilay Yerlioglu2,† , Claudia Phen3,† , Antje Ballauff4 , Natalia Nedelkopoulou5 , Isabel Spier6,7 , Inés Loverdos8 , Veronica B. Busoni9 , Jürgen Heise4 , Peter Dale10 , Tim de Meij11 , Kevin Sweet12 , Marta C. Cohen13 , Victor L. Fox14 , Emmanuel Mas15 , Stefan Aretz6,7 , Charis Eng16,17 , Stephan Buderus18 , Mike Thomson5 , Isabel Rojas3 and Holm H. Uhlig19,20,21,* 1 Department of Surgery and Cancer, Imperial College London, London SW7 2BX, UK, 2 Faculty of Medicine, Istanbul University, Istanbul, Fatih 34093, Turkey, 3 Division of Pediatric Gastroenterology, Hepatology, and Nutrition, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA, 4 Zentrum für Kinder- und Jugendmedizin Gastroenterology, HELIOS Klinikum Krefeld, Krefeld, Nordrhein-Westfalen 47805, Germany, 5 Pediatric Gastroenterology, Sheffield Children’s Hospital NHS Foundation Trust, Sheffield, Yorkshire S10 2TH, UK, 6 Institute of Human Genetics, Medical Faculty, University of Bonn, Bonn, Nordrhein-Westfalen 53012, Germany, 7 National Centre for Hereditary Tumor Syndromes, University Hospital Bonn, Bonn, Nordrhein-Westfalen 53126, Germany, 8 Pediatric Gastroenterology Hepatology and Nutrition Unit, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Catalonia comunidad 08208, Spain, 9 Pediatric Gastroenterology, Hepatology and Liver-Intestine Transplantation Division, Hospital Italiano de Buenos Aires, Ciudad Autónoma de Buenos Aires, Argentina, 10 Royal Gwent Hospital, Newport NP20 2UB, UK, 11 VU University Medical Center, Amsterdam 1081, The Netherlands, 12 Division of Human Genetics, Ohio State University Wexner Medical Center, Columbus, OH 43210, USA, 13 Histopathology Department, Sheffield Children’s Hospital NHS Foundation Trust, Sheffield, Yorkshire S10 2TH, UK, 14 Division of Gastroenterology, Hepatology and Nutrition, Boston Children’s Hospital, Boston, MA 02115, USA, 15 Unité de Gastroentérologie, Hépatologie, Nutrition, Diabétologie et Maladies Héréditaires du Métabolisme, Hôpital des Enfants, CHU de Toulouse, and IRSD, Université de Toulouse, INSERM, INRA, ENVT, UPS, Toulouse, Occitanie 31300, France, 16 Genomic Medicine Institute, Lerner Research Institute, and Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH 44195, USA, 17 Department of Genetics and Genome Sciences, and CASE Comprehensive Cancer Center, Case Western Reserve University School of Medicine, Cleveland, OH 44106, USA, 18 GFO-Kliniken Bonn, St. Marien-Hospital, Bonn, Nordrhein-Westfalen 53115, Germany, 19 Translational Gastroenterology Unit, University of Oxford, Oxford, Oxfordshire OX3 9DU, UK, 20 Department of Pediatrics, University of Oxford, Oxford, Oxfordshire OX3 9DU, UK and 21 Biomedical Research Centre, University of Oxford, Oxford, Oxfordshire OX4 2PG, UK *To whom correspondence should be addressed. Tel: +44 (0)1865220137; Fax: +44 (0)1865222901; Email: holm.uhlig@ndm.ox.ac.uk † The authors wish it to be known that, in their opinion, the first three authors should be regarded as joint First Authors. Received: January 13, 2021. Revised: March 28, 2021. Accepted: March 30, 2021 © The Author(s) 2021. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 1273 1274 Human Molecular Genetics, 2021, Vol. 30, No. 14 Abstract Ultra-rare genetic disorders can provide proof of concept for efficacy of targeted therapeutics and reveal pathogenic mechanisms relevant to more common conditions. Juvenile polyposis of infancy (JPI) is caused by microdeletions in chromosome 10 that result in haploinsufficiency of two tumor suppressor genes: phosphatase and tensin homolog deleted on chromosome 10 (PTEN) and bone morphogenetic protein receptor type IA (BMPR1A). Loss of PTEN and BMPR1A results in a much more severe phenotype than deletion of either gene alone, with infantile onset pan-enteric polyposis and a high mortality rate. No effective pharmacological therapy exists. A multi-center cohort analysis was performed to characterize phenotype and investigate the therapeutic effect of mammalian target of rapamycin (mTOR) inhibition (adverse events, disease progression, time to colectomy and mortality) in patients with JPI. Among 25 JPI patients identified (mean age of onset 13 months), seven received mTOR inhibitors (everolimus, n = 2; or sirolimus, n = 5). Treatment with an mTOR inhibitor reduced the risk of colectomy (hazard ratio = 0.27, 95% confidence interval = 0.07–0.954, P = 0.042) and resulted in significant improvements in the serum albumin level (mean increase = 16.3 g/l, P = 0.0003) and hemoglobin (mean increase = 2.68 g/dl, P = 0.0077). Long-term mTOR inhibitor treatment was well tolerated over an accumulated follow-up time of 29.8 patient years. No serious adverse events were reported. Early therapy with mTOR inhibitors offers effective, pathway-specific and personalized treatment for patients with JPI. Inhibition of the phosphoinositol-3-kinase–AKT–mTOR pathway mitigates the detrimental synergistic effects of combined PTEN–BMPR1A deletion. This is the first effective pharmacological treatment identified for a hamartomatous polyposis syndrome. Graphical Abstract BMP/SMAD signaling pathway, are a cause of juvenile polyposis Introduction syndrome (JPS). The study of rare genetic disorders can give insight into PHTS encompasses individuals with Cowden syndrome and pathogenic mechanisms relevant to more common conditions, Bannayan–Riley–Ruvalcaba syndrome carrying germline PTEN and targeting these mechanisms can provide proof of concept loss of function mutations (1–3). PHTS is associated with a range for personalized medicine. Dysregulation of signaling path- of dermatological, neurological and vascular features and leads ways controlling epithelial homeostasis is implicated in the to increased cancer susceptibility (4–6). Patients with PHTS typ- pathogenesis of hereditary polyposis syndromes and sporadic ically develop gastrointestinal symptoms secondary to polyp colorectal cancer. Targeting those mechanisms and preventing formation during young adulthood (1,7–9). Current recommen- tumor progression has been a research priority. dations are to initiate endoscopic surveillance for cancer at Phosphoinositol-3-kinase (PI3K)–AKT signaling and the bone 35 years of age (10). morphogenetic protein (BMP) pathway control epithelial and JPS is caused by germline defects in BMPR1A or SMAD4 mesenchymal crosstalk and maintain epithelial homeostasis. (11–14). Hamartomatous polyps typically develop in JPS by Hereditary hamartomatous polyposis syndromes result from 14–20 years of age (11,15,16). Endoscopic surveillance is recom- dysregulation of these pathways. Germline loss of function mended starting at age 12 or earlier if there are gastrointestinal mutations in phosphatase and tensin homolog deleted on symptoms (14,17). chromosome 10 (PTEN), the master negative regulator of PI3K– Since the PTEN and BMPR1A genes are located in close prox- AKT signaling, result in PTEN hamartoma tumor syndrome imity on chromosome 10, de novo 10q23.3 microdeletions can (PHTS) (1). Similarly, germline mutations of bone morphogenetic result in haploinsufficiency of both genes. Combined loss of protein receptor type IA (BMPR1A), a key receptor in the PTEN and BMPR1A has a synergistic effect, resulting in juvenile Human Molecular Genetics, 2021, Vol. 30, No. 14 1275 polyposis of infancy (JPI), a very severe hereditary polyposis syn- Patients presenting aged 12 months or under drome. It presents in the first 2 years of life and is characterized have a more severe phenotype by recurrent gastrointestinal bleeding, diarrhea, protein-losing Fourteen patients presented at 12 months of age or younger. enteropathy and intestinal obstruction secondary to polyp for- These patients had a higher polyp burden than patients present- mation (18–20). It often requires early surgery and significantly ing between 1 and 2 years of age (RR of identifying >100 polyps reduces the lifespan of affected children. All published cases = 6.46, 95% CI = 1.32–37.8, P = 0.015). This was associated with of PTEN–BMPR1A deletion syndrome have presented as sporadic significantly greater risk of having protein-losing enteropathy cases, caused by a de novo microdeletion. (RR = 3.93, 95% CI = 1.34–14.32, P = 0.0082) or requiring colectomy Polyposis syndromes are associated with a high risk of (RR = 2.16, 95% CI = 1.02–5.34, P = 0.0325). gastrointestinal and extraintestinal malignancy. Gastrointesti- nal symptoms, and the requirement for endoscopic surveillance and surgical management, result in significant morbidity. There mTOR inhibitor therapy was initiated in patients are currently no established pharmacological treatment options with a severe phenotype for JPI or any of the hamartomatous polyposis syndromes (21,22). Seven JPI patients received mTOR inhibitor treatment (mTOR In familial adenomatous polyposis (FAP), chemoprevention inhibitor group), and 18 patients received standard of care strategies using combined cyclo-oxygenase and epidermal without mTOR inhibitor (conventional treatment group). The growth factor inhibitors show promise in clinical trials, but median age of onset of gastrointestinal symptoms in the mTOR they are not in routine clinical use (22–24). Case reports have inhibitor group was 0.67 years (range = 0.42–1.50) compared with described clinical benefit from mammalian target of rapamycin 1.38 years (range = 0.08–2.0) in the conventional treatment group. (mTOR) inhibitors in JPI (25,26), PHTS (27) and FAP (23), but their There was panenteric polyposis in both groups. Endoscopic efficacy has not previously been demonstrated in controlled images from previously unpublished cases in the mTOR inhibitor studies. group demonstrate severe polyposis throughout the GI tract This study summarizes phenotypic features of patients (Supplementary Material, Fig. S1). Prior to initiating therapy, all with JPI and shows that treatment of JPI patients with patients in the mTOR inhibitor group had anemia, and all but mTOR inhibitors reduces disease progression, reduces need one had hypoalbuminemia. for colectomy and corrects hypoalbuminemia and anemia as surrogate markers of epithelial barrier dysfunction and bleeding. Treatment with mTOR inhibitors reduces risk of colectomy Four patients initiated mTOR inhibitor therapy prior to colec- Results tomy. None of these patients have required colectomy during follow-up. The hazard ratio for colectomy was significantly lower Clinical features and natural history of JPI in the mTOR inhibitor group than the conventional treatment A total of 25 patients with a definitive diagnosis of JPI result- group (hazard ratio = 0.27, 95% CI = 0.07–0.954, P = 0.042) (Fig. 2A). ing from confirmed microdeletions encompassing PTEN and Since disease severity and outcome differ depending on BMPR1A genes were included in the study (Fig. 1A). All developed the age at presentation, we performed a stratified analysis. multiple gastrointestinal polyps at 0.08–2 years of age (Fig. 1B). The colectomy rate among 11 conventional treatment patients Seven cases have not been previously reported (Supplementary presenting at age 12 months or under was 100% (median age of Material, Table S1). The JPI patients presented with typical PHTS- colectomy = 1.42 years, range = 0.83–10 years). Three patients in associated features such as macrocephaly in 80% [95% confi- the mTOR inhibitor group presented in this age range (median dence interval (CI) = 61–91], developmental delay in 48% (95% CI = age of commencing treatment = 1.50 years, range = 1.08– 30–67), muscular hypotonia in 36% (95% CI = 20–55), hemangioma 2.08 years), and none required colectomy (hazard ratio = 0.23, and/or lipoma in 20% (95% CI = 9–39) and penile freckling in 12% 95% CI = 0.056–0.91, P = 0.037) (Fig. 2B). (95% CI = 4–30) (Fig. 1C). Among patients who presented at over 1 year of age, the Median age of onset of gastrointestinal symptoms was conventional treatment patients had a colectomy rate of 36%, 1.0 year (range = 0.08–2.0 years). Symptoms included bleeding while the single patient who presented in this age range in the and anemia in 80% (95% CI = 61–91), protein-losing enteropathy mTOR inhibitor group has not required colectomy (hazard ratio or hypoalbuminemia in 52% (95% CI = 33–70), diarrhea in 36% = 0.33, 95% CI = 0.014–7.8, P = 0.49) (Fig. 2C). (95% CI = 20–55), intestinal obstruction or intussusception in Four patients in the conventional treatment group died (mor- 36% (95% CI = 20–55) and rectal prolapse in 24% (95% CI = 11–43) tality rate 22%) at a mean of 3.29 years after onset of gastroin- (Fig. 1C). testinal symptoms. None of the seven patients in the mTOR Other than mTOR inhibitors, pharmacological treatment was inhibitor group have died. There is no significant difference in limited to one patient who received a COX2 inhibitor (celecoxib) survival distribution between groups (P = 0.26). for 8 months. Blood transfusions, albumin infusions and iron supplementation were used as part of conventional treatment. Non-pharmacological interventions such as herbal and food Treatment with mTOR inhibitors improve protein-losing supplements were not systematically reported. enteropathy and chronic gastrointestinal bleeding Subtotal or total colectomy was performed in 60% of patients Hypoalbuminemia, resulting from protein-losing enteropathy, (mean age at colectomy = 3.2 ± 2.9 years, range = 0.83–10 years). was observed in all patients in the mTOR inhibitor therapy Mortality rate was 16% (mean age at death = 4.2 ± 1.9 years; range group prior to initiating therapy (Fig. 3A). mTOR inhibitor therapy = 3–7 years) and was from persistent rectal bleeding, nutritional resulted in significant increases in mean serum albumin within failure and complications owing to surgery. 3 months of starting treatment, and these were sustained over 1276 Human Molecular Genetics, 2021, Vol. 30, No. 14 Figure 1. Case identification and characterization of JPI patients. (A) Strategies used to identify JPI patients with PTEN and BMPR1A deletions. (B) Age of onset of gastrointestinal symptoms for each included case (black dots). Patients presenting at 12 months of age or younger (below gray dotted line) had more severe phenotype. (C) Percentage of included cases with selected phenotypic features of PHTS and gastrointestinal complications (error bars show 95% confidence interval). the follow-up period (Fig. 3B). A significant increase in mean increased significantly following initiation of mTOR inhibitors serum albumin was recorded in every treated patient (Fig. 3C). (Fig. 3E). After 1 year of treatment, the mean increase in After 1 year of treatment, the mean increase in serum albumin hemoglobin was 2.68 g/dl (95% CI = 1.08–4.28, P = 0.0077), and the concentration was 16.3 g/l (95% CI = 11.6–21.1, P = 0.0003). increase in hemoglobin was significant in every patient (Fig. 3F). Anemia, secondary to chronic gastrointestinal bleeding, was After 3 months of treatment, the minimum serum albu- present in all patients in the mTOR inhibitor group before min and hemoglobin concentrations did not fall below com- starting treatment (Fig. 3D). Mean hemoglobin concentration monly used thresholds for correction (albumin concentration Human Molecular Genetics, 2021, Vol. 30, No. 14 1277 Figure 2. The effect of mTOR inhibitor therapy on time to colectomy in patients with JPI resulting from PTEN and BMPR1A deletion. Survival analyses showing time to colectomy from first onset of gastrointestinal symptoms in patients who received mTOR inhibitor (‘mTOR inhibitor’) and those who did not (‘conventional treatment’). Ticks indicate end point of patients who did not have colectomy. Dots indicate point mTOR inhibitor initiated. P-values calculated by Mantel Cox Log rank analysis. (A) All patients included. (B) Limited to patients presenting at 1 year of age or under. (C) Limited to patients presenting over 1 year of age. of 20 g/l and hemoglobin concentration of 8 g/dl) in any of study to compare a case series of patients managed with mTOR the treated patients (Fig. 3B and E). The reduced requirement for inhibitors against a control group and is the first to confirm blood transfusion and albumin infusion allowed the removal of that the therapeutic inhibition of the PI3K–AKT–mTOR signaling a central venous catheter and facilitated discharge of one of pathway has a significant clinical benefit in JPI. the treated patients from hospital (Supplementary information: The retrospective cohort analysis showed that mTOR Clinical Narratives—Patient 4, Supplementary Material, Fig. S2). inhibitors reduced the risk of colectomy in JPI. The reduction in colectomy risk was greatest in patients presenting aged 12 months or under, who had a more severe phenotype with Effects of mTOR inhibitor treatment on health-care aggressive polyp burden. Intraindividual quantitative analysis utilization, growth and development identified significant improvements in clinically relevant surrogate markers of gastrointestinal bleeding and protein Individual case reports of patients treated using mTOR inhibitors losing enteropathy in all seven treated patients, and individual provide supportive evidence of their efficacy (Supplementary patient narratives identified other beneficial treatment effects. information: Clinical Narratives). A marked decrease in polyp Everolimus and sirolimus were both well tolerated with no burden following treatment was noted in patients 1–4 (Fig. 4), serious adverse events. allowing the reduction in frequency of surveillance endoscopy. Clinical experience in ultra-rare genetic conditions is limited Nutritional status was noted to improve in four patients. by the low number of patients worldwide and the lack of for- Patients 4 and 7 stopped total parenteral nutrition following malized trials to assess treatments. Different trial strategies are initiation of treatment, and patients 4–7 gained several centiles required to assess the treatment efficacy in these disorders. Our in weight and height on a standardized growth chart. study is an observational, open-label and retrospective cohort Developmental delay is a feature of PHTS and is aggravated study. It has several limitations: the number of patients who in patients with JPI by nutritional insufficiency and chronic underwent treatment is small, the follow-up time per patient illness. Patient 4 was noted to meet several new developmental is limited and there are insufficient data to define the optimal milestones as their medical condition improved. choice or dosing schedule of mTOR inhibitor. However, the treat- ment effect observed is large, highly significant and transfor- mative for patients. Publication bias was reduced by including mTOR inhibitors were well tolerated with no serious all JPI patients who received mTOR inhibitors as well as all adverse events patients who have not as a comparator group. The international The total cumulative time of mTOR inhibitor treatment was multicenter nature of the study reduced institutional bias and 29.8 years with 17.8 years cumulative follow-up on sirolimus showed that the therapeutic effect was observed independently and 12.0 years follow up on everolimus. Different dosing in seven centers. regimens were used, and the most common aimed for a trough This is the first controlled case series to demonstrate signif- sirolimus level of 5 ng/ml (Supplementary Material, Table S1). icant benefit from the pharmacological treatment of a hamar- Both treatments were well tolerated. During the total of tomatous polyposis syndrome and illustrates the potential ben- 29.8 years treatment, no serious adverse events have been efits of mechanistically informed treatments in these conditions noted (i.e. no hypertriglyceridemia, hypercholesterolemia or (22). Treatment benefit is clearer in JPI than in PHTS where severe infections). Aphthous ulcers were noted in patient 1 on everolimus and sirolimus have previously been used off-label everolimus treatment. Treatment with mTOR inhibitors has been (28). In PHTS, case reports describe the beneficial effects on vas- continued in all patients. cular malformation (29), thymic hyperplasia (30) and cerebellar function scores (27), but significant effects on gastrointestinal polyp progression have not yet been demonstrated (31). These Discussion reports do demonstrate that long-term mTOR inhibitors are This study is the largest published cohort of cases with genet- generally well tolerated, and the most frequent adverse events ically confirmed JPI. It provides a summary of the phenotypic were mild-to-moderate headache, fatigue, hypercholesterolemia features, the severe clinical course and the pharmacological and mucositis (27–30). However, long-term side effects such as and surgical treatments used in this condition. It is the first muscle wasting may be under-reported (32). 1278 Human Molecular Genetics, 2021, Vol. 30, No. 14 Figure 3. The effect of mTOR inhibitor therapy on hemoglobin and serum albumin concentration. (A, D) variation of serum albumin (A) or hemoglobin (D) concentration by time. Each participant initiated mTOR inhibitor therapy at t = 0. (B, E) A scatter plot comparing minimum serum albumin (B) and hemoglobin (E) concentration before and 3 months and onward on mTOR inhibitor treatment and comparing mean concentrations before and at three time points after initiation of mTOR inhibitor. Paired T-test performed to determine P-values. (C, F) Mean serum albumin (C) and hemoglobin (F) concentrations before and after initiating mTOR inhibitor therapy for each treated patient (error bars show standard deviation). Welch’s T-test performed to determine P-value. MTI = mTOR-inhibitor, # = patient number, ∗ P < 0.05, ∗∗ P < 0.01, ∗∗∗ P < 0.001. Translational research in ultra-rare genetic disorders offers may be relevant in the treatment of sporadic colon cancers, 20% exceptional insight into key pathogenic pathways and identifies of which have acquired 10q deletion (33). potentially generalizable, targeted treatment strategies for more The BMP, PTEN and Wnt/ß-catenin pathways control the common disorders. This study highlights a therapeutic axis that regeneration and development of the gut epithelium (34,35). Human Molecular Genetics, 2021, Vol. 30, No. 14 1279 Figure 4. Endoscopic images showing decrease in polyp burden in patient with JPI treated with mTOR inhibitor. Endoscopic images of small intestine showing polyp progression from pre-sirolimus (A), 2 weeks of sirolimus (B) and 4 weeks of sirolimus (C). They are dysregulated in hereditary polyposis syndromes and has over 2700 subscribers, including pediatric gastroenterology sporadic colorectal cancers. BMPs influence cell differentiation fellows and consultants from at least 40 countries. Cases of and tumorigenesis through their canonical role in SMAD infantile polyposis syndrome were requested and selected on signaling (Fig. 5) (36–38). BMP activation also blocks Wnt/ß- a voluntary basis by the responding physicians. All participat- catenin signaling (39), and correspondingly, inhibition of ing institutions were advised to consult with their institutional BMP activates the Wnt/ß-catenin pathway in intestinal cells review board prior to transfer of information. Data were obtained (36). Loss of BMPR1A reduces transcription of SMAD target via a retrospective chart review by the responding physician and genes and increases Wnt/ß-catenin signaling, which promotes were anonymized prior to transfer via secure email. Reports were cell proliferation. Similarly, loss of PTEN upregulates the reviewed to identify and exclude duplicated patient information PI3K–AKT–mTOR pathway, promoting cell proliferation and and to update follow-up data. survival (Fig. 5) (40–42). In addition, a structured search of the medical literature PTEN–BMPR1A deletion dysregulates both the BMP–SMAD was performed to identify published reports of patients with and PI3K–AKT–mTOR signaling pathways. These pathways JPI caused by PTEN–BMPR1A deletion syndrome. The PubMed are largely independent, but there are relevant intersections database was searched using the terms ‘infantile polyposis’, which are informative when considering the efficacy of mTOR ‘PTEN BMPR1A’, or ‘polyposis PTEN BMPR1A’. Results were lim- inhibitors. BMP activates PTEN, and BMP2 stabilizes and ited to publications since 1997 (i.e. when PTEN was identified). increases the expression of PTEN (43). PTEN, in reducing the References were screened to identify additional relevant reports. activity of PI3K–AKT signaling, also decreases WNT signaling For each patient, age at onset of gastrointestinal symptoms, by reducing the AKT-mediated inhibition of GSK3ß and by age at diagnosis and last follow-up, presence of chromosomal preventing ß-catenin accumulation (36). mTOR inhibition is deletion and extent, clinical features, gastrointestinal symp- expected to partially correct disturbed PI3K signaling and toms, number and distribution of polyps, gastrointestinal pro- re-establish some regulation of the Wnt signaling pathway, cedures and operations (whether a colectomy was performed potentially explaining the significant clinical benefits observed and at what age) and whether patients received pharmacolog- in this study. More specific therapeutic targets that may be used ical treatment for the polyps were recorded. For patients who in the future include PI3K inhibitors, dual inhibitors of mTOR received treatment with an mTOR inhibitor, hemoglobin and and PI3K (40,41,44), or increasing PTEN expression via targeting serum albumin concentration values were collected. Responding the WW domain-containing ubiquitin E3 ligase 1-dependent physicians and authors of published reports were contacted to degradation pathway (45) (Fig. 5). acquire the missing data and for additional follow-up data. There is currently neither a standard of care nor a pharma- All patients in the final analysis fulfilled the following three cological treatment for patients with JPI owing to PTEN–BMPR1A inclusion criteria: (A) the presence of a confirmed germline deletion syndrome (17). This study shows mTOR inhibitors can deletion encompassing the PTEN and BMPR1A genes, (B) the modify the severe clinical course of this condition and indicates presence of intestinal polyposis consistent with JPI with age that they should be considered as a primary treatment option, of onset between 0 and 2 years and (C) clinical follow-up data particularly in those presenting under 1 year of age or with a available for at least 18 months after the onset of polyposis severe phenotype. A prospective multi-center study should be (Fig. 1). performed to confirm these findings. However, the observed treatment effect in this study questions whether a placebo- controlled trial of everolimus or sirolimus would be warranted. Retrospective cohort analysis For the retrospective cohort analysis, outcomes were compared Materials and Methods between patients who received treatment with everolimus or sirolimus in addition to normal treatment (‘mTOR inhibitor Case identification and data collection group’) and those who did not receive mTOR inhibitors but Previously unreported cases of JPI were identified by two strate- otherwise received standard of care (‘conventional treat- gies. First, genetic data from four large cohorts of PHTS and JPS ment group’). Longitudinal clinical data and details of the patients were searched for chromosome 10 deletions encom- treatment are reported for the mTOR inhibitor group. Serum passing PTEN and BMPR1A genes. The cohorts were compiled albumin concentration and hemoglobin concentration were by specialist treatment centers in Oxford, Bonn, Cleveland and compared before and after treatment to assess individual Southampton. Second, cases were identified by email survey outcomes, and surgical interventions including colectomy were on an international pediatric gastroenterology list server which recorded. 1280 Human Molecular Genetics, 2021, Vol. 30, No. 14 Figure 5. Signaling of the PTEN–PI3K and the BMPR1A pathway. A diagram of the PI3K and BMP/SMAD signaling pathways. BMP binding to BMPR complex causes BMPR1A activation and SMAD phosphorylation, nuclear translocation and transcription of SMAD target genes. BMPR1A signaling intersects with Wnt signaling by blocking ß- catenin activation and preventing Wnt target gene expression. PTEN controls the levels of PIP3 , a key mediator of the PI3K–AKT–mTOR signaling pathway that regulates cell proliferation and survival. The complex intersection of signaling pathways at multiple levels of cell cycle control explains the synergistic defect caused by loss of BMPR1A and PTEN in chromosome 10 microdeletions. Therapeutic interference with this dysregulated signaling can be achieved via inhibition of several kinases that mediate the PI3K signaling pathway (including mTOR, AKT or PI3K inhibitors). Ethical considerations differences between groups were compared using Chi-squared test for dichotomous variables and Welch’s T-test for continuous For all previously unreported cases, patients or their guardians variables. Changes in mean and minimum hemoglobin and provided written informed consent for research as approved serum albumin concentration were assessed using paired T-test. by the institutional review boards (Oxford GI biobank study CIs were calculated by the Wilson/Brown method for proportions and investigation of rare diseases sub-project; University of and by the Koopman asymptotic score for relative risks. Survival Oxford and University of Bonn). Each case met criteria for de- distributions were compared using the Mantel-Cox log-rank identified data in accordance with the HIPAA standards. The test. Signaling diagrams were created in PowerPoint (Microsoft, UT-Southwestern Human Research and Protection Program has Redmond, USA) based on the graphical elements of Servier determined that storage and analysis of de-identified data Medical Art (https://smart.servier.com/) and were edited using for this series does not constitute human subjects research the GNU Image Manipulation Program (www.gimp.org/). as defined under federal regulations [45 CFR 46.102] and does not require institutional review board approval. All parameters investigated reflect the standard of clinical care. Supplementary Material Supplementary Material is available at HMG online. Reporting and statistics Data were reported based on the STrengthening the Reporting Acknowledgements of OBservational studies in Epidemiology (STROBE) guide- We thank the patients and their families who participated in the lines (46). Statistical analysis was performed using Graph- study and Dr Katherine Lachlan for stimulating Pad Prism (GraphPad Software, San Diego, USA). Statistical discussions. We acknowledge the contribution of the BRC Human Molecular Genetics, 2021, Vol. 30, No. 14 1281 Gastrointestinal biobank (11/YH/0020, 16/YH/0247). The research 10. Tischkowitz, M., Colas, C., Pouwels, S. and Hoogerbrugge, was supported by the National Institute for Health Research N. (2020) Cancer surveillance guideline for individuals with (NIHR) Oxford Biomedical Research Centre (BRC). C.E. is an ACS PTEN hamartoma tumour syndrome. Eur. J. Hum. Genet., 28, Clinical Research Professor and is the Sondra J. and Stephen 1387–1393. R. Hardis Endowed Chair in Cancer Genomic Medicine at the 11. Aretz, S., Stienen, D., Uhlhaas, S. and Stolte, M. (2007) High Cleveland Clinic. Two authors (I.S., S.A.) of this publication proportion of large genomic deletions and a genotype phe- are members of the European Reference Network on Genetic notype update in 80 unrelated families with juvenile poly- Tumour Risk Syndromes (ERN GENTURIS)—Project ID No 739547. posis syndrome. J. Med. Genet., 44, 702–709. This project is supported by the NIHR Oxford Biomedical 12. Byrne, R.M. and Tsikitis, V.L. (2018) Colorectal polyposis and Research Centre. The views expressed are those of the authors inherited colorectal cancer syndromes. Ann. Gastroenterol., and not necessarily those of the NIHR or the Department of 31, 24–34. Health and Social Care. 13. Blatter, R.H.E., Plasilova, M., Wenzel, F., Gokaslan, S.T., Ter- racciano, L., Ashfaq, R. and Heinimann, K. (2015) Somatic Conflict of Interest statement. The authors declare no potential alterations in juvenile polyps from BMPR1A and SMAD4 conflicts of interest related to this article. H.H.U. has received mutation carriers. Genes. Chromosomes Cancer, 54, 575–582. research support or consultancy fees from Janssen, UCB Pharma, 14. Blatter, R., Tschupp, B., Aretz, S., Bernstein, I., Colas, C., Evans, Eli Lilly, Boehringer Ingelheim, Pfizer, Celgene, MiroBio, Regen- D.G., Genuardi, M., Hes, F.J., Hüneburg, R. and Järvinen, H. eron and AbbVie. (2020) Disease expression in juvenile polyposis syndrome: a retrospective survey on a cohort of 221 European patients and comparison with a literature-derived cohort of 473 Funding SMAD4/BMPR1A pathogenic variant carriers. Genet. Med., 22, This project is funded by NIHR Oxford Biomedical Research 1524–1532. Centre. 15. Gammon, A., Jasperson, K., Kohlmann, W. and Burt, R.W. (2009) Hamartomatous polyposis syndromes. Best Pract. Res. Clin. Gastroenterol., 23, 219–231. References 16. Levi, Z., Baris, H.N., Kedar, I., Niv, Y., Geller, A., Gal, E., Gingold, 1. Yehia, L., Ngeow, J. and Eng, C. (2019) PTEN-opathies: from R., Morgenstern, S., Baruch, Y., Leach, B.H., Bronner, M.P. biological insights to evidence-based precision medicine. J. and Eng, C. (2011) Upper and lower gastrointestinal find- Clin. Invest., 129, 452–464. ings in PTEN mutation—positive Cowden syndrome patients 2. Heald, B., Mester, J., Rybicki, L., Orloff, M.S., Burke, C.A. and participating in an active surveillance program. Am. Coll. Eng, C. (2010) Frequent gastrointestinal polyps and colorectal Gastroenterol., 1, e5–e6. adenocarcinomas in a prospective series of PTEN mutation 17. Cohen, S., Hyer, W., Mas, E., Auth, M., Attard, T.M., Spalinger, carriers. Gastroenterology, 139, 1927–1933. J., Latchford, A. and Durno, C. (2019) Management of juvenile 3. Marsh, D.J., Coulon, V., Lunetta, K.L., Rocca-Serra, P., Dahia, polyposis syndrome in children and adolescents: a position P.L., Zheng, Z., Liaw, D., Caron, S., Duboué, B., Lin, A.Y. et al. paper from the ESPGHAN polyposis working group. J. Pediatr. (1998) Mutation spectrum and genotype-phenotype analy- Gastroenterol. Nutr., 68, 453–462. ses in Cowden disease and Bannayan-Zonana syndrome, 18. Chandan, O.C., Reyes-Santiago, E., Vargas, L., Mercer, D., two hamartoma syndromes with germline PTEN mutation. Grant, W., Langnas, A., Quiros-Tejeira, R.E. and Vo, H. (2019) Hum. Mol. Genet., 7, 507–515. Multiorgan transplantation as a viable treatment option in 4. Perry, J.M., He, X.C., Sugimura, R., Grindley, J.C., Haug, infantile juvenile polyposis syndrome. J. Pediatr. Gastroenterol. J.S., Ding, S. and Li, L. (2011) Cooperation between both Nutr., 68, e86–e87. Wnt/{beta}-catenin and PTEN/PI3K/Akt signaling promotes 19. Vargas-Gonzalez, R., de la Torre-Mondragon, L., Aparicio- primitive hematopoietic stem cell self-renewal and expan- Rodriguez, J.M., Paniagua-Morgan, F., Lopez-Hernandez, G., sion. Genes Dev., 25, 1928–1942. Garrido-Hernandez, M.A. and Nunez-Barrera, S. (2010) Juve- 5. Mester, J. and Eng, C. (2015) PTEN hamartoma tumor syn- nile polyposis of infancy associated with paracentric inver- drome. Handb. Clin. Neurol., 132, 129–137. sion and deletion of chromosome 10 in a Hispanic patient: a 6. Tan, M.-H., Mester, J.L., Ngeow, J., Rybicki, L.A., Orloff, case report. Pediatr. Dev. Pathol., 13, 486–491. M.S. and Eng, C. (2012) Lifetime cancer risks in individ- 20. Waisbourd-Zinman, O., Mamula, P. and Piccoli, D.A. (2016) uals with germline PTEN mutations. Clin. Cancer Res., 18, Chromosome 10q23 deletion syndrome: an overlap of Ban- 400–407. nayan–Riley–Ruvalcaba syndrome and juvenile polyposis 7. Stanich, P.P., Pilarski, R., Rock, J., Frankel, W.L., El-Dika, S. and syndrome. Chromosome 10q23 deletion syndrome: an over- Meyer, M.M. (2014) Colonic manifestations of PTEN hamar- lap of Bannayan–Riley–Ruvalcaba syndrome and juvenile toma tumor syndrome: case series and systematic review. polyposis syndrome. J. Paediatr. Child Health, 52, 852. World J. Gastroenterol., 20, 1833–1838. 21. Patel, R. and Hyer, W. (2019) Practical management of poly- 8. Lachlan, K.L., Lucassen, A.M., Bunyan, D. and Temple, posis syndromes. Frontline Gastroenterol., 10, 379–387. I.K. (2007) Cowden syndrome and Bannayan Riley Ruval- 22. Samadder, N.J., Baffy, N., Giridhar, K.V., Couch, F.J. and caba syndrome represent one condition with variable Riegert-Johnson, D. (2019) Hereditary cancer syndromes—a expression and age-related penetrance: results of a clin- primer on diagnosis and management, part 2: gastrointesti- ical study of PTEN mutation carriers. J. Med. Genet., 44, nal cancer syndromes. Mayo Clin. Proc., 94, 1099–1116. 579–585. 23. Kemp Bohan, P.M., Mankaney, G., Vreeland, T.J., Chick, R.C., 9. Pilarski, R., Stephens, J.A., Noss, R., Fisher, J.L. and Prior, T.W. Hale, D.F., Cindass, J.L., Hickerson, A.T., Ensley, D.C., Sohn, (2011) Predicting PTEN mutations: an evaluation of Cowden V., Clifton, G.T., Peoples, G.E. and Burke, C.A. (2020) Chemo- syndrome and Bannayan-Riley-Ruvalcaba syndrome clinical prevention in familial adenomatous polyposis: past, present features. J. Med. Genet., 48, 505–512. and future. Familial Cancer, 20, 23–33. 1282 Human Molecular Genetics, 2021, Vol. 30, No. 14 24. Burke, C.A., Dekker, E., Lynch, P., Samadder, N.J., Balaguer, 35. Guaraldi, F., Di Nardo, G., Tarani, L., Bertelli, L., Susca, F.C., F., Hüneburg, R., Burn, J., Castells, A., Gallinger, S., Lim, R. Bagnulo, R. and Resta, N. (2017) Association of autoimmune et al. (2020) Eflornithine plus sulindac for prevention of thyroiditis and celiac disease with juvenile polyposis due to progression in familial adenomatous polyposis. N. Engl. J. 10q23.1q23.31 deletion: potential role of PI3K/Akt pathway Med., 383, 1028–1039. dysregulation. Eur. J. Med. Genet., 60, 380–384. 25. Busoni, V.B., Orsi, M., Lobos, P.A., D’Agostino, D., Wagener, M., 36. Freeman, T.J., Smith, J.J., Chen, X., Washington, M.K., De la Iglesia, P. and Fox, V.L. (2019) Successful treatment of Roland, J.T., Means, A.L., Eschrich, S.A., Yeatman, juvenile polyposis of infancy with sirolimus. Pediatrics, 144, T.J., Deane, N.G. and Beauchamp, R.D. (2012) Smad4- 2922. mediated signaling inhibits intestinal neoplasia by 26. Quaranta, M., Laborde, N., Ferrand, A., Danjoux, M., Vergnolle, inhibiting expression of beta-catenin. Gastroenterology, 142, N., Barreau, F., Racaud-Sultan, C. and Mas, E. (2019) Sustain- 562–571.e2. able positive response to sirolimus in juvenile polyposis of 37. Cichy, W., Klincewicz, B. and Plawski, A. (2014) Juvenile infancy. J. Pediatr. Gastroenterol. Nutr., 68, e38–e40. polyposis syndrome. Arch. Med. Sci., 10, 570–577. 27. Komiya, T., Blumenthal, G.M., DeChowdhury, R., Fioravanti, 38. Howe, J.R., Dahdaleh, F.S., Carr, J.C., Wang, D., Sherman, S.K. S., Ballas, M.S., Morris, J., Hornyak, T.J., Wank, S., Hewitt, and Howe, J.R. (2013) BMPR1A mutations in juvenile polypo- S.M., Morrow, B. et al. (2019) A pilot study of sirolimus in sis affect cellular localization. J. Surg. Res., 184, 739–745. subjects with Cowden syndrome or other syndromes char- 39. Yuan, G., Yang, G., Zheng, Y., Zhu, X., Chen, Z., Zhang, Z. acterized by germline mutations in PTEN. Oncologist, 24, and Chen, Y. (2015) The non-canonical BMP and Wnt/beta- 1510–e1265. catenin signaling pathways orchestrate early tooth develop- 28. Schmid, G.L., Kassner, F., Uhlig, H.H., Korner, A., Kratzsch, ment. Development, 142, 128–139. J., Handel, N., Zepp, F.-P., Kowalzik, F., Laner, A., Starke, S. 40. Pitts, T.M., Newton, T.P., Bradshaw-Pierce, E.L., Addison, R., et al. (2014) Sirolimus treatment of severe PTEN hamartoma Arcaroli, J.J., Klauck, P.J., Bagby, S.M., Hyatt, S.L., Purkey, A., tumor syndrome: case report and in vitro studies. Pediatr. Tentler, J.J. et al. (2014) Dual pharmacological targeting of the Res., 75, 527–534. MAP kinase and PI3K/mTOR pathway in preclinical models 29. Hammer, J., Seront, E., Duez, S., Dupont, S., Van Damme, A., of colorectal cancer. PLoS One, 9, e113037. Schmitz, S., Hoyoux, C., Chopinet, C., Clapuyt, P., Hammer, F. 41. Janku, F., Yap, T.A. and Meric-Bernstam, F. (2018) Targeting et al. (2018) Sirolimus is efficacious in treatment for exten- the PI3K pathway in cancer: are we making headway? Nat. sive and/or complex slow-flow vascular malformations: a Rev. Clin. Oncol., 15, 273–291. monocentric prospective phase II study. Orphanet J. Rare Dis., 42. Laplante, M. and Sabatini, D.M. (2012) mTOR signaling in 13, 191. growth control and disease. Cell, 149, 274–293. 30. Heindl, M., Handel, N., Ngeow, J., Kionke, J., Wittekind, C., 43. Zhou, X.-P., Waite, K.A., Pilarski, R., Hampel, H., Fernandez, Kamprad, M., Rensing-Ehl, A., Ehl, S., Reifenberger, J., Lod- M.J., Bos, C., Dasouki, M., Feldman, G.L., Greenberg, L.A., denkemper, C. et al. (2012) Autoimmunity, intestinal lym- Ivanovich, J. et al. (2003) Germline PTEN promoter muta- phoid hyperplasia, and defects in mucosal B-cell home- tions and deletions in Cowden/Bannayan-Riley-Ruvalcaba ostasis in patients with PTEN hamartoma tumor syndrome. syndrome result in aberrant PTEN protein and dysregulation Gastroenterology, 142, 1093–1096.e6. of the phosphoinositol-3-kinase/Akt pathway. Am. J. Hum. 31. Kim, S.T., Lee, J., Park, S.H., Park, J.O., Park, Y.S., Kang, W.K. Genet., 73, 404–411. and Lim, H.Y. (2017) Prospective phase II trial of everolimus 44. Kawata, T., Tada, K., Kobayashi, M., Sakamoto, T., Taki- in PIK3CA amplification/mutation and/or PTEN loss patients uchi, Y., Iwai, F., Sakurada, M., Hishizawa, M., Shirakawa, with advanced solid tumors refractory to standard therapy. K., Shindo, K. et al. (2018) Dual inhibition of the mTORC1 BMC Cancer, 17, 211. and mTORC2 signaling pathways is a promising thera- 32. Gyawali, B., Shimokata, T., Honda, K., Kondoh, C., Hayashi, peutic target for adult T-cell leukemia. Cancer Sci., 109, N., Yoshino, Y., Sassa, N., Nakano, Y., Gotoh, M. and Ando, Y. 103–111. (2016) Muscle wasting associated with the long-term use of 45. Lee, Y.-R., Chen, M., Lee, J.D., Zhang, J., Lin, S.-Y., Fu, T.-M., mTOR inhibitors. Mol. Clin. Oncol., 5, 641–646. Chen, H., Ishikawa, T., Chiang, S.-Y., Katon, J. et al. (2019) 33. Karoui, M., Tresallet, C., Julie, C., Zimmermann, U., Staroz, Reactivation of PTEN tumor suppressor for cancer treatment F., Brams, A., Muti, C., Boulard, C., Robreau, A.-M., Puy, H. through inhibition of a MYC-WWP1 inhibitory pathway. et al. (2004) Loss of heterozygosity on 10q and mutational Science, 364, eaau0159. status of PTEN and BMPR1A in colorectal primary tumours 46. Vandenbroucke, J.P., von Elm, E., Altman, D.G., Gotzsche, P.C., and metastases. Br. J. Cancer, 90, 1230–1234. Mulrow, C.D., Pocock, S.J., Poole, C., Schlesselman, J.J. and 34. Nusse, R. and Clevers, H. (2017) Wnt/beta-catenin signal- Egger, M. (2014) Strengthening the reporting of observa- ing, disease, and emerging therapeutic modalities. Cell, 169, tional studies in epidemiology (STROBE): explanation and 985–999. elaboration. Int. J. Surg., 12, 1500–1524.