Histopathologic, phenotypic, and molecular criteria to discriminate low-grade intestinal T-cell lymphoma in cats from lymphoplasmacytic enteritis

Abstract

Background: Differentiation of low-grade intestinal T-cell lymphoma (LGITL) from lymphoplasmacytic enteritis (LPE) in cats is a diagnostic challenge for pathologists.

Objective: Characterize histologic, immunohistochemical, and molecular features of LGITL and LPE.

Animals: Forty-four client-owned cats, 22 diagnosed with LGITL and 22 with LPE.

Methods: Prospective, cohort study. Clinical suspicion of LGITL or LPE was based on persistent gastrointestinal signs, unresponsive to empirical treatments. All cats underwent a standardized diagnostic evaluation, including biopsy (preferentially full-thickness), and were diagnosed with LGITL or LPE after review of clinical, laboratory, sonographic, histologic, immunohistochemical, and clonality results.

Results: A monomorphic lymphocytic population (22/22, 100%) and in-depth mucosal infiltration (15/22, 68%) were hallmarks of LGITL. Epithelial patterns (nests and plaques) were significantly more frequent in LGITL (11/22, 50%) than in LPE (1/22, 5%) cases (P = .001). A CD3+ lymphocytic apical-to-basal gradient was observed in 9/22 (41%) of LGITL vs 1/22 (5%) of LPE cases (P = .004). Most LPE cases (17/18, 94%) featured marked fibrosis in the superficial part of the lamina propria. The Ki-67 20%- and 30%-thresholds discriminated between LGITL and LPE within both the epithelium (specificity >95%) and lamina propria (specificity >95%), respectively. All LGITL cases were CD3+ pSTAT3- and pSTAT5+. T-cell receptor gamma chain gene rearrangements indicated monoclonality in 86% of LGITL cases. Surprisingly, 70% of LPE cases featured monoclonality (40%) or monoclonality on a polyclonal background (30%).

Conclusions and clinical importance: We identified new histologic, immunohistochemical, and clonality criteria to distinguish LGITL from LPE.

Keywords: CD20; CD3; JAK-STAT; Ki-67; PARR; alimentary lymphoma; clonality; epithelium; fibrosis; full-thickness intestinal biopsies; gradient; histology; immunohistochemistry; inflammatory bowel disease; lamina propria; monoclonal; nest; plaque; polyclonal.

FIGURE 1

Pattern A shows apical‐to‐basal gradient and massive infiltration by low‐grade intestinal T‐cell lymphoma (LGITL) (HES ×100) with small and monomorphous CD3+ T‐cells, displaying an intraepithelial lymphocytosis (HES ×400) or lymphocytic cryptitis. In rare cases neutrophilic cryptitis are observed. Both cases present variable villous blunting. Masson Trichrome staining show a thin fibrosis of the lamina propria in the apical‐to‐basal gradient onset and no fibrosis of the lamina propria in the massive infiltrative onset. Pattern B represents 2 cases of lymphoplasmacytic enteropathies. HES staining shows a superficial fibrosis of the lamina propria in both cases, confirmed by Masson's Trichrome staining. High power field magnification pictures show a polymorphous infiltrate including neutrophilic cryptitis or abscesses. CD3 staining confirms the absence of IEL increase and show a heterogeneous CD3+ inflammatory infiltrate in the lamina propria

FIGURE 2

Pattern A shows low‐grade intestinal T‐cell lymphoma (LGITL) case with CD20 aberrant expression and differential expression of negative pSTAT3 and positive pSTAT5. At higher magnification (×400), intraepithelial and lamina propria lymphocytes display the same phenotype. Pattern B shows LGITL case with an apical‐to‐basal infiltration (CD3+ partial infiltration of the villi and superficial lamina propria) and a lymphoplasmacytic enteritis (LPE) case. Differential expression of pSTAT3 and pSTAT5 between LGITL and LPE may help to discriminate both diagnoses: tumoral T‐cells are pSTAT3− and pSTAT5+ whereas reactive inflammatory cells are pSTAT3+ and pSTAT5 mostly negative

FIGURE 3

Receiver operator characteristic (ROC) curve for differentiating cats with low‐grade intestinal T‐cell lymphoma (LGITL) from cats with lymphoplasmacytic enteritis (LPE) based on Ki67 expression in the epithelium and lamina propria. AUC, area under the curve

FIGURE 4

Survival data for 20 cats with low‐grade intestinal T‐cell lymphoma (LGITL) and 21 cats with lymphoplasmacytic enteritis (LPE). Eight LGITL cats were censored, as they were all lost to follow‐up at the time of analysis. Sixteen LPE cats were censored, including 14 cats that were lost to follow‐up at the completion of the study period and 2 cats that were still alive at the time of analysis. Median survival time for LGITL cats was 719 days (range, 4‐1272 days). Median survival times for LPE cats was not defined because more than 50% of the LPE cohort was censored