Pattern and staging of gastrointestinal graft-versus-host disease: A clinicopathologic correlational study. (2023)

INTRODUCTION

Graft-versus-host disease (GVHD) is a serious condition that can occur after allogeneic bone marrow transplantation or peripheral hematopoietic stem cell transplantation (HSCT). The underlying pathogenetic mechanism involves the alloimmune response induced by transplanted donor immune cells, leading to epithelial cell apoptosis, inflammation and tissue damage. (1)

Acute GVHD usually occurs within the first 100 days after transplantation, while chronic GVHD occurs later. Due to the presence of late-onset acute GVHD, the distinction between acute and chronic GVHD should not depend solely on the time of GVHD onset, as clinical manifestations are more important indicators to distinguish between acute and chronic GVHD. (2) In acute GVHD, the liver, skin, gastrointestinal tract (GIT), bone marrow, thymus and lungs can be affected. The gastrointestinal tract is the second most frequently affected body part in acute GVHD (3) and manifests mainly in non-specific symptoms such as diarrhoea, nausea and vomiting, which can also occur with infectious gastroenteritis (e.g. cytomegalovirus) or drug toxicity. Therefore, histopathological examination of GI biopsy is essential for accurate diagnosis. (4-6)

Acute GVHD is an urgent diagnosis for pathologists, but the assessment of GVHD in the gastrointestinal tract is not always straightforward. Given that apoptosis, the main histopathological feature of GVHD, can be observed in various situations, the diagnosis can become a nightmare, especially when there is a superimposed GI infection with CMV. Therefore, unless the biopsy reveals florid abnormalities and sufficient clinical and endoscopic information, the pathologist should exercise caution in diagnosing acute GVHD. (4)

Histopathological assessment of GVHD requires grading of tissue changes that reflect disease severity. The grading of GI-GVHD with increasing severity consists of apoptosis, crypt damage, crypt loss, and mucosal detachment originally proposed by Lerner et al. was defined. (7) Based on this scoring system, it has been reported that the lower gastrointestinal tract, mainly the rectum and left colon, are the sites most severely and frequently affected by GVHD (8-13), while the duodenum is the most frequently affected share. in the upper gastrointestinal tract is the GIT. (13) The 2014 consensus conference on the histopathology of GVHD revised the histopathological criteria for acute and chronic GVHD without recommending a scoring scheme or a specific biopsy site in the gastrointestinal tract. However, it has been suggested that if an institution prefers to use a rating system, the site with the greatest damage and grade should be reported. (4) The clinical grading of acute GVHD, on the other hand, is based on the updated version of Glucksberg (14) and the International Bone Marrow Transplant Registry scoring system developed by Przepiorka et al. (15) and is based on the affected organ system, including the gastrointestinal tract, skin and liver, and the severity of the injury. (14,15) Criteria based on the severity of mucosal damage are often used for endoscopic classification of acute intestinal GVHD Freiburg. (16,17) Clinical symptoms and endoscopic findings are very useful in the workup of gastrointestinal GVHD when histopathological examination supports the diagnosis, but clinicopathological correlation is not always present. A limited number of studies examined the correlation between clinical and pathological grading systems and found little or no correlation between clinical and pathological grades of gastrointestinal GVHD. (5,12) The correlation between endoscopic and pathologic grade also remains controversial, with some studies showing concordance (16-18) while others show discordance. (19.20)

In this study, we aimed to investigate the patterns of gastrointestinal (GI) involvement in acute GVHD at different sites of the gastrointestinal tract and assess their value in terms of clinicopathological correlation.

MATERIALS AND METHODS

patients

Pathology reports of endoscopic GI biopsies from patients after hematopoietic stem cell transplantation (HSCT) between 2005 and 2019 were retrieved from the automated hospital database of the Department of Pathology, Ankara University Faculty of Medicine. A total of 479 GI biopsies from 164 patients with at least one GI biopsy diagnosed by two GI pathologists (AE, BS) as “consistent with acute GVHD” were included in the study. Patients with repeat biopsies were excluded. The majority of patients (n=140, 85.4%) had more than one biopsy taken from different gastrointestinal sites, while the remaining patients (n=24, 14.6%) had only one biopsy taken mainly from the left colon ( n=20) ). ). Clinical information such as age, sex, underlying disease, graft types, conditioning regimens, stem cell source, donor type, tissue match, presence of GVHD in other organ systems, endoscopic findings, GI symptoms, and time between transplant and biopsy were obtained from electronic medical records and manual records of endoscopy units. The study complies with ethical standards and was found adequate by the Institutional Review Board of Ankara University Faculty of Medicine (Ref. no.: 2019/1).

Histopathological assessment and classification

At the time of diagnosis, the GI biopsies were evaluated by experienced GI pathologists (AE, BS) who were critical of the differential diagnosis of GVHD, including radiochemotoxicity, drug side effects, and co-infections, all of which were ruled out by careful histopathologic analysis. The classification of acute GVHD was based on the globally recognized system of Lerner et al. (7) and the NIH consensus criteria. (21) The grading system classifies the severity of the disease into 4 groups (Figure 1), including grade 1: apoptosis of gland/crypt epithelium; grade 2: apoptosis of epithelial cells with isolated gland/crypt destruction; Grade 3: Apoptosis of epithelial cells with extensive gland/crypt destruction resulting in crypt loss; Grade 4: Total mucosal denudation/ulcer ratio n. Apoptosis was more pronounced in the regenerative area of ​​the gland/crypt: in the gastric body, apoptotic cells were mainly found in the cervical region of the glands and in the deeper glands in the antrum. In the duodenum and other areas of the small intestine, apoptotic bodies were more abundant in the neck and deep crypts than in the villous epithelium, while they were restricted to the deep crypts of the colon. CMV infection was suspected in 62 cases, which was determined by immunohistochemistry using anti-CMV antibodies (clone DDG9/CCH2, Cell Marque, Rocklin, CA) and the streptavidin-biotin-peroxidase technique on the Benchmark XT automated stain (Ventana , Tucson, AZ).

Clinical and endoscopic classification

Clinical grading of acute GVHD was performed using the modified Seattle-Glucksberg grading system (14), which is based on grading of the most affected organs, consisting of skin, gastrointestinal tract, and liver. While the extent of the maculopapular rash and serum bilirubin levels were considered to determine the clinical stage for skin and liver, respectively, stool volume or the presence of severe abdominal pain, ileus, and profuse bloody stools were assessed to determine the clinical GI stage. Based on the stages of these three organs, namely skin, liver and gastrointestinal tract, an overall clinical GVHD grade for the patient was determined (Table 1).

The endoscopic classification was based on the Freiburg criteria, a 4-level classification system: Grade 1: no clear criteria; grade 2: patchy erythema; grade 3: aphthous lesions; and grade 4: confluent defects, ulcers, mucosal detachment. (16.17)

Statistical analysis

Analysis of non-parametric variables was performed using either the chi-square test or the Mann-Whitney U test. Spearman's rank correlation coefficient test was used to determine the correlations between clinical, pathologic, and endoscopic grades. For statistical purposes, the comparisons were repeated by grouping the cases into low (classes 1 and 2) and high (classes 3 and 4) classes. Social Sciences Version 25.0 Statistical Package (IBM Corp., Armonk, NY, USA) was used for all analyses, and P<0.05 was considered statistically significant.

RESULTS

Clinical characteristics

Our series included 164 patients, including 133 (81.1%) adults and 31 (18.9%) children/adolescents (age <18 years) with a mean age of 38 years (range: 1–68 years) and male predominance from 7 to 5 (n=97, 59.1% men; n=67, 40.9% women). More than half of the patients received bone marrow transplantation for acute leukemia (n=90, 54.9%), followed by myelodysplastic syndrome (n=18, 11.0%), chronic leukemia (n=15, 9.1%) and multiple myeloma (n). =3, 1.8%). Other disorders included Hodgkin or non-Hodgkin lymphoma, idiopathic myelofibrosis, thalassemia major, aplastic anemia, autoimmune hemolytic anemia, sickle cell anemia, hemophagocytic lymphohistiocytosis, severe generalized immunodeficiency, and cirrhosis. The majority of all transplants were allogeneic (n = 153, 93.3%), while only 6 were autologous and 2 liver transplants. A significant proportion of patients (n=112, 68.3%) had myeloablative conditioning, and 72.6% (n=119) used peripheral blood as a stem cell source. Almost all patients (n=163, 99.4%) had known symptoms and in 44.8% (n=73) of patients diarrhea was the main symptom followed by upper and lower gastrointestinal symptoms (28.8%; n=47) . and 26.4% (n=43) upper gastrointestinal symptoms. In 128 of 164 patients, endoscopic findings were extracted from the endoscopy reports. According to the reports, 50 (39.0%) patients had normal mucosa (grade 1), 27 (21.0%) had erythema (grade 2), 32 (25.0%) had aphthous lesions (grade 3), and 19 (15 .0%) presented with ulceration (grade 4). The clinical characteristics of the patients are shown in Table 2.

Histopathological findings

A total of 479 biopsies were taken from 164 patients included in the study. The distribution of biopsies from different GI sites was as follows: 44 (9.2%) esophagus, 90 (18.8%) stomach, 91 (19.0%) duodenum, 20 (4.2%) of the terminal ileum, 32 (6.7%) right colon, 87 (18.2%) left colon and 115 (23.9%) rectum. As 78 cases had simultaneous left colon and rectal biopsies, they were pooled for statistical purposes, resulting in a total of 124 biopsies.

The majority of patients evaluated with multiple biopsies (n=1.40, 85.4%) had biopsy-proven GI-GVHD at more than one site (n=115/140, 82.1%), and half of these patients (n=72/140, 51.4%) had biopsy-proven GI-GVHD in all regions examined. In cases with simultaneous upper and lower gastrointestinal biopsies, GVHD involvement of the lower gastrointestinal tract, upper gastrointestinal tract, and both upper and lower gastrointestinal tracts was detected in 13.5%, 10.1%, and 76.2% of the cases.

Due to the high frequency of duodenal GVHD (n = 80/91, 87.9%), the total participation in LG I (n = 118/126, 93.6%) and UGI (n = 91/97, 93 .8%) similar ( P =). .3). The duodenum (n=80/91, 87.9%) was the most commonly (P<0.0001) affected upper GI site, followed by the stomach (n=52/90, 57.8%) and the esophagus (n = 16/44, 36.4). %), while in the lower gastrointestinal tract both the right (n = 29/32, 90.6%) and left colon including the rectum (n = 114/124, 91.9%) were equally frequently affected, followed by the terminal . ileum (n=15/20, 75%). The severity of GVHD in the duodenum and lower gastrointestinal tract was higher than GVHD in the stomach (P=0.021; P=0.003) and esophagus (P=0.005; P=0.001) (Figure 2).

A correlation analysis of pathologic GVHD grades in different GI domains revealed that only esophagus and stomach had a weak positive correlation (coefficient: 0.334, P=0.029) in the upper GI, while no such correlation was found for the lower GI - domains. In the lower gastrointestinal tract, terminal ileum showed a moderate positive correlation with right (coefficient: 0.658, P=0.003) and left colon (coefficient: 0.540, P=0.021), similar to the correlation between right and left colon (coefficient). : 0.649, P < 0.000).

Clinical-pathological findings

To assess the association between pathologic GVHD grades and clinical features, the highest GVHD grade was accepted as the pathologic grade for each patient. Accordingly, the distribution of pathologic GVHD grades was as follows: 46 (28.0%) grade 1, 42 (25.6%) grade 2, 44 (26.8%) grade 3, and 32 (19.5%) grade 4. Statistical analysis revealed this evidence. GVHD rates at more than one site increased in parallel with pathologic GVHD grades (P<0.001). The frequency of UGI and LGI involvement was similar in patients with UGI symptoms, LGI symptoms, and both upper and lower GI symptoms. Although not significant, the frequency of upper gastrointestinal symptoms alone (nausea and vomiting) decreased as pathologic GVHD grade increased, while the presence of both upper and lower gastrointestinal symptoms (nausea, vomiting, and diarrhea) increased (P =0.077). . . Other clinical characteristics such as age, gender, donor type, underlying disease, conditioning regimens, and tissue match were not significantly associated with pathologic GVHD grade. When the same comparisons were made by classifying the cases as low-grade (grades 1 and 2) and high-grade (grades 3 and 4) GVHD, similar results were obtained.

The clinical GI stages obtained from 113 patients were distributed as follows: 22 (19.5%) stage 0, 19 (16.8%) stage 1, 39 (34.5%) stage 2, 23 (20.4%) ) stage 3 and 10 (8.8%) stage 4. A small positive correlation was found between clinical GI stage and the highest pathologic GVHD grade (coefficient: 0.308, P=0.001). Correlation analysis for pathologic and endoscopic grades revealed a weak positive correlation (coefficient: 0.261, P = 0.003). However, no such correlation was found between endoscopic grades and clinical GI stages (coefficient: 0.100, P=0.3).

Some patients were suspected of having extra-gastrointestinal GVHD on clinical follow-up, and because of changing symptoms extra-gastrointestinal biopsies were taken.

Biopsy-proven extra-GI GVHD consisting of skin (n=65, 39.6%), liver (n=21, 12.8%), oral mucosa (n=14, 8.5%), eye (n=15 , 9.1%) and pulmonary GVHDs (n=2, 1.2%) were found in 71.3% (n=117) of patients. Immunohistochemical CMV staining was completely negative in 62 patients with pathological or clinical suspicion.

DISCUSSION

The results of this study showed that GVHD in the gastrointestinal tract predominantly and more severely affected the lower gastrointestinal tract and duodenum with increasing severity compared to other upper gastrointestinal regions. In addition, all sites in the lower gastrointestinal tract were invariably affected, while uneven distribution was observed in the upper gastrointestinal tract, where the duodenum was affected more frequently than the esophagus and stomach. Studies published to date have primarily highlighted involvement of the distal colon (rectum or sigmoid) as the most common pattern of GVHD in the gastrointestinal tract (8,9,11,12,22), whereas few have reported more frequent involvement of the upper gastrointestinal tract. intestinal tract (13 ) or equal to a lower GI. (23,24) The proximal and distal colons were also shown to be equally affected by GVHD, prompting some investigators to perform proctosigmoidoscopy as the initial investigation. (9,10,13,23-26) Based on the predominant and diffuse involvement of GVHD in the lower gastrointestinal tract, our results confirm previous studies and suggest that examination of the distal colon is an effective approach for diagnosing GVHD in in the majority of cases, the lower gastrointestinal tract and the associated risks will be reduced by a detailed endoscopic examination. This also applies to cases with upper GI symptoms alone, as similar GVHD involvement was observed in the UGI and LGI tract, regardless of symptoms. On the other hand, some investigators suspected that gastrointestinal GvHD might be associated with panintestinal or "patchy-diffuse" disease (3,13,23,24,27-30), which was also confirmed in this study, where the majority of cases were present of patients with simultaneous upper and lower gastrointestinal biopsies had GVHD in both cases. Therefore, a full GI endoscopy should be considered, especially in patients diagnosed as "incompatible with GVHD" with single biopsies, in order not to miss such cases with "flexed" disease. Given the high degree of involvement compared to the esophagus and stomach, the duodenum should also be carefully examined, especially in patients with a lower GI biopsy.

There are only a limited number of studies comparing the severity of GVHD at different gastrointestinal sites (see Table 3). According to Ma et al. (9) The stomach was the least affected, while the rectum was the most affected place in the gastrointestinal tract. Pathological grades were similar at different sites in the upper gastrointestinal tract, in contrast to the lower gastrointestinal tract, where a significant increase in grades was observed from the proximal to the distal colon. (9) In contrast, Nomura et al. (25), who compared only lower gastrointestinal sites, found that severity increased from left to right colon and terminal ileum. Our results are partially consistent with the previous study, as the stomach and esophagus were the least affected sites, but the severity was similar at sites in the lower gastrointestinal tract. While a poor positive correlation was found between esophageal and gastric GVHD grades, the duodenum exhibited similar GVHD grades to the colon. This confirms that the duodenum, as the most common and most affected site in the upper gastrointestinal tract, is the primary biopsy site that should be carefully investigated, especially when only upper gastrointestinal endoscopy is performed. However, in the lower gastrointestinal tract, all regions showed a moderately positive correlation with each other in terms of severity.

Conflicting results were found in studies comparing pathologic and clinical grades, with some investigators finding little or no concordance (5,12) and suggesting that pathologic grading of GVDH should not be considered mandatory practice, while others found a high degree of correlation between pathological and clinical scores observed. (31-33) Conflicting results have also been shown in studies comparing pathological and endoscopic grades. (16-20) According to our results, pathological grades were in poor agreement with clinical and endoscopic grades. There are several reasons behind this poor agreement, including a possible fragmented distribution of the disease in the gastrointestinal tract, the lack of standardized biopsy protocols, or the early endoscopic biopsy protocol in the clinical work-up of GVHD in our institute, before the clinical picture is complete. On the other hand, these results emphasize the importance of pathological examination for GVHD diagnosis in the GIT.

The retrospective nature and the lack of a standardized endoscopic biopsy protocol represent the major limitations of our study. Had the biopsies included every gastrointestinal site in every patient, the comparisons of these sites would have been optimal in terms of incidence and severity of pathology. On the other hand, the thorough clinical and histopathological evaluation of such a large cohort from a single institution makes the study results remarkable.

In summary, the results of this study suggest that examination of the distal colon with rectosigmoidoscopy together with upper gastrointestinal endoscopy, especially duodenoscopy, would be the first and probably sufficient approach in patients with suspected gastrointestinal GVHD, as the duodenum and lower gastrointestinal tracts are similarly affected by GVHD in terms of frequency and severity. However, because of the possible "flexed" nature of the disease, negative biopsies of the left colon and duodenum before performing a complete colonoscopy and esophageal stroduodenoscopy with multisite biopsies cannot rule out the diagnosis of GVHD. Finally, since GI-GVHD has nonspecific characteristics, clinical, endoscopic, and pathologic evaluation should be performed by an experienced medical team of hemato-oncologists, gastroenterologists, and GI pathologists for accurate diagnosis.

main points

* Although graft-versus-host disease tends to affect the lower gastrointestinal tract, duodenal graft-versus-host disease should not be ignored as it has a similar involvement in graft-versus-host disease as the lower gastrointestinal tract.

* The similar frequency of graft-versus-host disease involvement throughout the lower gastrointestinal tract makes proctosigmoidoscopy the most practical first approach to the diagnosis of graft-versus-host disease.

* Nonspecific microscopic findings of graft-versus-host disease require an integrated approach consisting of clinical, endoscopic, and pathologic evaluation.

Ethics Committee Approval: Approval was obtained from the Institutional Review Board, Faculty of Medicine, Ankara University (Ref. No.: 2019/1).

Informed consent: Written informed consent was obtained from patients participating in this study.

Peer Review: Externally assessed.

Author Contributions: Concept - S.K.O., A.K., C.C.E., E.U.i., B.S., A.K., H.Q., AE.; Design - S.K.O., A.K., C.C.E., E.U.i., B.S., A.K., H.Q., AE.; Supervision – A.E.; Resources - S.K.O.; Materials - S.K.O., H.Q.; Data collection and/or processing – S.K.O., N.A., M.K.Y., M.E., H.Q.; Analysis and/or interpretation - S.K.O., A.K., N.A., C.C.E., B.S., AE.; Literature search - S.K.O., A.K., N.A., C.C.E., AE.; Writing manuscript – S.K.O., A.K., C.C.E., M.K.Y., B.S., M.E., A.E.; Critical review – B.S., A.K., A.E.

Acknowledgments: The authors would like to thank the Department of Biostatistics, Ankara University for their help with the statistics used in this manuscript.

Declaration of interest: The authors have no conflict of interest to declare.

Funding: This study received no funding.

REFERENCES

(1.) Society G, Blazar BR. Acute graft-versus-host disease: from the laboratory bench to the bedside. Blood. 2009;114(20):4327-4336.

(2.) Jagasia MH, Greinix HT, Arora M, et al. National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-versus-Host Disease: I. The Report of the Diagnosis and Staging Task Force 2014. Biol Blood Marrow Transplant. 2015;21(3):389-401.e1.

(3.) Washington K, Jagasia M. Pathology of graft-versus-host disease of the gastrointestinal tract. Hum Pathol. 2009;40(7):909-917.

(4.) Shulman HM, Cardona DM, Greenson JK, et al. NIH consensus development project on criteria for clinical trials in chronic graft-versus-host disease: II. The report of the Pathology Working Group 2014. Biol Blood Marrow Transplantation. 2015;21(4):589-603.

(5.) Abraham J, Janin A, Gornet JM, et al. Clinical severities in gastrointestinal graft-versus-host disease. Transplantation. 2014; 97(9):965-971.

(6.) Couriel D, Caldera H, Champlin R, Komanduri K. Acute graft-versus-host disease: pathophysiology, clinical manifestations, and management. Cancer. 2004;101(9):1936-1946.

(7.) Lerner KG, Kao GF, Storb R, Buckner CD, Clift RA, Thomas ED. Histopathology of graft-versus-host disease (GvHR) in human recipients of bone marrow from HL-A-matched sibling donors. transplant procedure 1974;6(4):367-371.

(8.) Ross WA, Ghosh S, Dekovich AA, et al. Endoscopic biopsy diagnosis of acute gastrointestinal graft-versus-host disease: rectosigmoid biopsies are more sensitive than upper gastrointestinal tract biopsies. Bin J Gastroenterol. 2008;103(4):982-989.

(9.) Ma C, Maluf HM, Liu TC. Acute graft-versus-host disease occurs more frequently and more severely in the lower than in the upper gastrointestinal tract. Hum Pathol. 2015;46(10):1480-1487.

(10.) Nydegger A, Catto-Smith AG, Tiedemann K, Hardikar W. Diagnosis of gastrointestinal graft-versus-host disease - is rectal biopsy sufficient? Pediatric blood cancer. 2007;48(5):561-566.

(11.) Crowell KR, Patel RA, Fluchel M, Lowichik A, Bryson S, Pohl JF. Endoscopy for the diagnosis of intestinal graft-versus-host disease: is inferior endoscopy with biopsy as effective for diagnosis as superior endoscopy combined with inferior endoscopy? Pediatric blood cancer. 2013;60(11):1798-1800.

(12.) Thompson B, Salzman D, Steinhauer J, Lazenby AJ, Wilcox CM. Prospective endoscopic evaluation of gastrointestinal graft-versus-host disease: determining the best diagnostic approach. bone marrow transplant. 2006;38(5):371-376.

(13.) Ip S, Marquez V, Schaeffer DF, Donnellan F. Sensitivities of biopsy sites in the endoscopic assessment of graft-versus-host disease: Retrospective review from a tertiary center. You Dis Sci. 2016;61(8):2351-2356.

(14.) Rowlings PA, Przepiorka D, Klein JP, et al. IBMTR grade index for grading acute graft-versus-host disease: retrospective comparison with Glucksberg grade. Br J Hematol. 1997;97(4):855-864.

(15.) Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on the Classification of Acute GVHD. bone marrow transplant. 1994;15:825-828.

(16.) Kreisel W, Herbst EW, Schwind B, et al. Diagnosis and classification of acute graft-versus-host disease after allogeneic bone marrow transplantation using sigmoidoscopy. Eur J Gastroenterol Hepatol. 1994;6(8):723-730.

(17.) Kreisel W, Dahlberg M, Bertz H, et al. Endoscopic diagnosis of acute intestinal GVHD after allogeneic hematopoietic SCT: a retrospective analysis in 175 patients. bone marrow transplant. 2012;47(3):430-438.

(18.) Cheung DY, Kim JI, Kim SS, et al. Endoscopic examination in gastrointestinal graft-versus-host disease: comparisons with histologic findings. You Dis Sci. 2008;53(11):2947-2954.

(19') Borges LR, Vilela EE, Ferrari MLA, Cunha AS, Vasconcelos AC, Torres HOG. Diagnosis of acute gastrointestinal graft-versus-host disease in patients undergoing allogeneic hematopoietic stem cell transplantation. A descriptive and critical study of diagnostic tests. Hematol Transfusion Cell Ther. 2020;42(3):245-251.

(20.) Thompson B, Salzman D, Steinhauer J, Lazenby AJ, Wilcox CM. Prospective endoscopic evaluation of gastrointestinal graft-versus-host disease: determining the best diagnostic approach. bone marrow transplant. 2006;38:371-376.

(21.) Shulman HM, Kleiner D, Lee SJ, et al. Histopathologic diagnosis of chronic graft-versus-host disease: National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: report of the II Pathology Working Group. Biol Blood Marrow Transplantation. 2006;12(1):31-47.

(22.) Lee KJ, Choi SJ, Yang HR, et al. Stepwise endoscopy based on sigmoidoscopy to assess graft-versus-host disease in children. Pediatric Gastroenterol Hepatol Nutr. 2016;19(1):29-37.

(23.) Aslanian H, Chander B, Robert M, et al. Prospective evaluation of acute graft-versus-host disease. You Dis Sci. 2012;57(3):720-725.

(24.) Roy J, Snover D, Weisdorf S, Mulvahill A, Filipovich A, Weisdorf D. Simultaneous upper and lower endoscopic biopsy in the diagnosis of intestinal graft-versus-host disease. Transplantation. 1991;51(3):642-646.

(25.) Nomura K, Lizuka T, Kaji D, et al. Usefulness of endoscopic examination in the diagnosis of acute lower gastrointestinal graft-versus-host disease. Gastroenterol Res Pract. 2017;2017:2145986.

(26.) Minamino H, Machida H, Tominaga K et al. Rectal biopsy is more appropriate than ileal biopsy to confirm the diagnosis of early gastrointestinal graft-versus-host disease. Scand J Gastroenterol. 2015; 50(11):1428-1434.

(27.) Naymagon S, Naymagon L, Wong SY, et al. Acute intestinal graft-versus-host disease: considerations for the gastroenterologist. Nat Rev Gastroenterol Hepatol. 2017;14(12):711-726.

(28.) Ferrara JLM, Levine JE, Reddy P, Holler E. Graft-versus-host disease. Lancet. 2009;373(9674):1550-1561.

(29.) Choi SW, Levine JE, Ferrara JL. Pathogenesis and management of graft versus host disease. Immunol Allergy Clinic Nord Am. 2010; 30(1):75-101.

(30.) Vogelsang GB, Lee L, Bensen-Kennedy DM Pathogenesis and management of graft-versus-host disease after bone marrow transplantation. Annu Rev Med. 2003;54:29-52.

(31.) Velasco-Guardado A, Lopez-Corral L, Alvarez-Delgado A, et al. Endoscopic examination and histological findings in graft versus host disease. Rev Esp Enferm Dig. 2012;104(6):310-314.

(32.) Melson J, Jakate S, Fung H, Arai S, Keshavarzian A. Crypt loss is a marker of clinical severity in acute gastrointestinal graft-versus-host disease. Bin J Hematol. 2007;82(10):881-886.

(33.) Sauvestre F, Belleannee G, Breal C, et al. Histologic analysis has prognostic value in colorectal biopsies evaluated for suspected graft-versus-host disease. Virchow's arch. 2018;472(2):213-220.

Sonay Kus Ozturk [1] [ID], Ayca Kirmizi [1] [ID], Nermin Aras [1] [ID], Cevriye Cansiz Ersoz [1] [ID], Meltem Kurt Yuksel [2] [ID], Elif Unal Ince [3] [ID], Berna Savas [1] [ID], Mehmet Ertem [3] [ID], Aydan Kansu [4] [ID], Hulya Cetinkaya [5] [ID], Arzu Ansari [1] [ AUSWEIS]

[1] Department of Pathology, Ankara University Faculty of Medicine, Sihhiye, Ankara, Turkey

[2] Department of Hematology, Ankara University Faculty of Medicine, Sihhiye, Ankara, Turkey

[3] Department of Pediatric Hematology, Ankara University Faculty of Medicine, Sihhiye, Ankara, Turkey

[4] Department of Pediatric Gastroenterology, Ankara University Faculty of Medicine, Sihhiye, Ankara, Turkey

[5] Department of Gastroenterology, Ankara University Faculty of Medicine, Sihhiye, Ankara, Turkey

Corresponding author: Sonay Kuf Ozturk, E-mail: sonayku@gmail.com

Received: January 12, 2022 Accepted: August 3, 2022 Published date: February 20, 2023

DOI: 10.5152/tjg.2023.22012

Caption: Figure 1. Histological grading of GVHD. (A) Grade 1, apoptosis in crypt epithelium (H&E, x200); (B) Grade 2, isolated crypt destruction (H&E, x200); (C) Grade 3, contiguous crypt destruction (H&E, x200); (D) Grade 4, diffuse mucosal defect/ulceration (H&E, x40). GVHD, graft versus host disease; H&E, hematoxylin and eosin.

Table 1. Clinical stage of acute GVHD (14) Target organ stage Skin (active liver (serum total erythema only) bilirubin)0 None active <2 mg/dL (<34.21 (erythematous) rash [micro]mol/L)1 Maculopapular Rash, 2-3 mg/dL (34.2151.31 <25% BSA [micro]mol/L)2 Maculopapular rash, 3.1-6 mg/dL (53.02-25%-50% BSA 102, 62 μmol/ L)3 Maculopapular rash, 6.1-15 mg/dL (104.33->50% BSA 256.56 [micro]mol/L)4 Generalized >15 mg/dL (>256, 56 Erythroderma ( >50% [micro]mol/L) ) BSA), plus bullous formation and desquamation (>5% BSA) Target organ stage skin (active upper GI erythema only)0 None active None or intermittent nausea, rash (erythematous), vomiting or anorexia1 Rash maculopapular, persistent nausea, <25% BSA Vomiting or anorexia2 Rash maculopapular, -- 25%-50% BSA3 Rash maculopapular, -- >50% BSA4 Generalized -- Erythroderma (>50% BSA), plus bullous formation and peeling (>5% BSA) Target organ stage Skin (lower gastrointestinal active erythema). Only) (faeces)0 Not active Adult: <500 mL per day Rash (erythematous) Child: <10 mL/kg per day 1 Maculopapular rash, Adult: 500-999 mL per day <25% BSA Child: 10-19.9 ml/kg per day2 maculopapular rash, adult: 1,000-1,500 ml/day, 25-50% BSA, child: 20-30 ml/kg/day3 maculopapular rash, adult: >1,500 ml per day, >50% BSA, child: > 30 ml/kg per day4 Generalized Severe abdominal pain with erythroderma (>50% or no ileus or heavy BSA), plus bullous bloody stools (regardless of formation and volume), desquamation (>5% BSA) BSA, body surface area.

Table 2. Clinical characteristics of patients Patient characteristics n (%)Age (years), median (range) 38 (1-68) Gender Female 67 (40.9) Male 97 (59.1) Underlying disease Acute myeloid leukemia 56 (34, 1 ) Acute lymphoid leukemia 34 (20.7) Chronic leukemia 15 (9.1) Myelodysplastic syndrome 18 (11.0) Myeloma 3 (1.8) Other 35 (21.3) Unknown 3 (1.8) Graft types Allogeneic 153 (93.3) Autologous 7 (93.3) Autologous 7 (6.3) organ. . ) Unknown 3 (1.8) Conditioning regimens Myeloablative 112 (68.3) Non-myeloablative 15 (9.1) Unknown 37 (22.6) Stem cell source 119 (72.6) Peripheral blood Bone marrow 9 (5.5) 1 Umbilical cord blood 8 (Peripheral cord blood) + Bone Bone marrow 3 (1.8) Unknown 30 (18.3) Donor type related 106 (64.6) Unrelated 38 (23.2) Unknown 20 (12.2) Tissue match Complete 99 ( 60 .4) Partially 21 (12.8) Unknown 44 (26). by biopsy GVHD 117 (71.3) Skin 65 (39.6) Liver 21 (12.8) Oral mucosa 14 (8.5 ) Eye 15 (9.1) Lung 2 (1.2) Gastrointestinal symptoms Nausea-vomiting 18 (10.0) Diarrhea 50 (10.0) Diarr. + diarrhea 40 (24.4) Unknown 56 ( 34.1) Endoscopic findings Normal mucosa 23 (33.8%) Erythema 17 (25%) Thawed lesion 20 (29.4%) Ulceration 8 (11.8%) Time between transplant 59 (10-1500) and biopsy (days). ), median (range)

Table 3. Summary of previous studies on GI GVHDAuthor Year Case n Frequency of GI GVHDMa et al (9) 2015 110 LGT > UGT Duo > Stom Similar LGTRoss et al (8) 2008 112 LGT > UGTCrowell et al (11) 20013 LGT > UGTThompson et al (12) 2006 24 LGT > UGTNydegger et al (10) 2007 11 Stom-Duo > Oes LGT > UGTAslanian et al (23) 2012 27 LGT= UGT Similar to LGTRoy et al (24) 77 LGT 71 UGTIp et al (13) 2016 46 Duo > Oes -Stom Similar to LGTNomura et al (25) 2017 186 Similar to LGTMinamino et al (26) 2015 16 Rec > T.ileAbraham et al (5) 2014 210 --Velas et al (31) 2012 197 --Melson et al (32) 2007 23 --Sauvestre et al (33) 2018 112 --Current study 2023 164 LGT> UGT Duo > Oes-Stom Similar to LGTAuhor Severity of GI GVHD ( 9) LGT> UGT Similar in UGT LC > RCross et al (8) --Crowell et al (11) --Thompson et al (12) --Nydegger et al (10) --Aslanian et al (23) - - Roy et al (24) -- Ip et al (13) --Nomura et al (25) T.ile > RC > LCMinamino et al (26) --Abraham et al (5) --Velasco-Guardado et al ( 31) --Melson et al ( 32) --Sauvestre et al (33) --Current study LGT > UGT Duo > Oes-Stom Duo = LGT Similar under LGTA Author clinical-pathological correlation Ma et al (9) --Ross et al (8) --Crowell et al (11) --Thompson et al (12) NoNydegger et al (10) --Aslanian et al (23) --Roy et al (24) --Ip et al (13) ) - - Nomura et al (25 ) --Minamino et al (26) --Abraham et al (5) NoVelasco-Guardado et al (31) JaMelson et al (32) JaSauvestre et al (33) Low survival rate with higher pathological grade. Recent study Low positive correlationLGT, lower gastrointestinal tract; UGT, upper gastrointestinal tract; LC, left colon; RC, right colon; duo, duodenum; power, stomach; Esophagus; rec, rectum; T.ile, terminal ileum.

Figure 2. Pathological GVHD grades along the gastrointestinal tract. GVHD, graft versus host disease; GIT, gastrointestinal tract. Esophagus Stomach Duodenum Terminal Right Left Ileum Colon Colon Grade 1 14 33 37 8 10 28 Grade 2 1 11 17 1 6 32 Grade 3 0 6 14 2 7 33 Grade 4 1 6 214