Enteropathy-Associated T-Cell Lymphoma and Other Primary Intestinal T-Cell Lymphomas


Enteropathy-Associated T-Cell Lymphoma (Type I, Classical)

An association between malabsorption and intestinal lymphoma was first reported in 1937, at which time lymphoma was considered responsible for the malabsorption. However, in 1962, Gough and coworkers demonstrated that intestinal lymphoma was a complication of celiac disease or gluten-sensitive enteropathy. In 1978, Isaacson and Wright characterized celiac disease–associated lymphoma as a single entity, originally considered a form of malignant histiocytosis. Later, Isaacson and coworkers showed that the neoplastic cells were of T-cell lineage. This lymphoma subtype was categorized in the 2008 World Health Organization (WHO) classification as enteropathy-associated T-cell lymphoma (EATL) type I, or classical EATL. In the revised 2016 WHO classification, it is designated simply as EATL.

Definition

EATL is a neoplasm derived from intraepithelial lymphocytes (IELs) showing variable degrees of cellular pleomorphism, often within a polymorphic background.

Epidemiology

EATL is the most common subtype of primary intestinal T-cell lymphoma (66% to 80%). However, it is a rare lymphoma, accounting for less than 1% of non-Hodgkin's lymphomas, less than 5% of primary gastrointestinal lymphomas, and approximately 5% of peripheral T-cell lymphomas (PTCLs) in Western countries. EATL characteristically occurs in the sixth and seventh decades of life, although it has been reported in younger adults. A mild male predominance (1.04-2.8 : 1) has been observed in most series. Most, if not all, patients with EATL have the celiac disease–associated HLADQA1*0501, DQB1*0201 (HLA-DQ2) genotype. EATL is more common in regions with a high seroprevalence of celiac disease, such as Europe (EATL incidence of 0.05 to 0.14/100,000) and the United States (EATL incidence of 0.016/100,000). The reasons for a higher incidence of EATL in individuals of northern European descent (and in certain European countries) despite a similar incidence of celiac disease in most Western countries are unknown. EATL is virtually non-existent in regions such as the Far East, where most lack the celiac disease susceptibility alleles. There is an increased incidence of EATL in older individuals, 2.92/100,000 in the 60- to 69-year age group and 0.05/100,000 in people older than 60 years in Europe and the United States, respectively. The incidence of EATL in the celiac population is 0.22 to 1.9/100,000. A wide range in the relative risk of non-Hodgkin's lymphoma, including EATL, has been reported for celiac disease (and dermatitis herpetiformis) patients (3 to 100), with population-based studies providing more reliable (and lower) estimates. On the basis of recent studies, the prevalence of EATL in at-risk individuals is likely to be much lower (≤1%) than estimated previously.

Etiology

EATL is a recognized complication of celiac disease, a common autoimmune disorder (incidence, 1% in most regions of the world) with myriad intestinal and extraintestinal manifestations that occur in genetically susceptible individuals intolerant to gluten-containing grains (e.g., wheat, barley, and rye). In retrospective studies of patients with EATL, celiac disease has been diagnosed in 38% to 100% of affected patients; this variation is likely a result of incomplete data or inclusion of non-EATL cases. Evidence for an association between celiac disease and EATL comes from identical human leukocyte antigen (HLA) types in celiac disease patients and those with EATL, the demonstration of gluten sensitivity in EATL patients, and the protective effect of a gluten-free diet against the development of lymphoma ; the risk of lymphoma decreases, albeit gradually, after commencing a gluten-free diet. Homozygosity for HLA-DQ2 alleles, observed in 53% of cases (compared with 21% in uncomplicated celiac disease), is considered a risk factor for development of EATL.

A diagnosis of celiac disease is established in 20% to 73% of cases before the diagnosis of EATL. A short history of “adult-onset” celiac disease usually predates EATL, but it may occur in individuals with long-standing disease, the mean time between diagnosis of celiac disease and EATL ranging from 46.8 months to 10 years. In 10% to 58% of cases, celiac disease and EATL are diagnosed simultaneously, with up to a third being discovered at surgery for intestinal obstruction or perforation, and the diagnosis of EATL is occasionally made at autopsy. At times, EATL might occur in patients only manifesting celiac-associated extraintestinal conditions, such as dermatitis herpetiformis. Gastrointestinal symptoms in celiac disease do not correlate with the degree of small intestinal mucosal damage, and more than 50% of patients may be asymptomatic. Hence, certain individuals with EATL might have had lifelong “silent” or “cryptic” gluten sensitivity because jejunal villous atrophy and crypt hyperplasia are found in uninvolved small intestinal mucosa when the tumor is resected. The only manifestation of celiac disease in some cases is an increase in IELs, and in a minority, the jejunum appears nearly normal. Studies showing that the jejunal mucosa can appear normal in “latent” celiac disease might provide an explanation for this finding, which was previously thought to argue against a strict association between EATL and celiac disease.

Clinical Presentation

EATL most commonly presents with abdominal pain (65% to 100%) and recurrence of gluten-insensitive malabsorption or diarrhea (40%-70%) in previously well individuals or those with a history of adult-onset (or childhood-onset) celiac disease and a prior response to a gluten-free diet. * Other presentations include weight loss (50% to 80%), acute abdominal emergency due to intestinal perforation (25% to 50%), hemorrhage, anorexia, fatigue, and early satiety or nausea or vomiting due to intestinal obstruction. * Ichthyotic rash and finger clubbing may be observed. B symptoms, besides weight loss, are present in less than a third of patients. The duration of symptoms can range from 1 week to 5 years, and a symptom duration of less than 3 months before EATL diagnosis was observed in 59% of cases in one study.

* References .

Small intestinal involvement is detected in 90% to 96% of “de novo” EATLs, most commonly in the jejunum. The frequency of small intestinal involvement by EATL with prior refractory celiac disease type II (RCD II, discussed later; Table 38-1 ) is lower (65%). Multifocal lesions involving different segments of small intestine are seen in 32% to 54%, ulcers or strictures in 51%, and mass lesions in 32% of cases. The next most common gastrointestinal sites are the large intestine and stomach. EATL might occasionally present at extraintestinal sites (e.g., skin, lymph nodes, spleen, or central nervous system), usually in cases evolving from RCD II, as the aberrant IELs in RCD II frequently disseminate to extraintestinal sites. Common sites of EATL dissemination include intra-abdominal lymph nodes (35%), bone marrow (3% to 18%), lung and mediastinal lymph nodes (5% to 16%), liver (2% to 8%), and skin (5%). High-stage disease at diagnosis has been reported in 43% to 90% of cases, although the staging systems used in different studies have varied. An Eastern Cooperative Oncology Group (ECOG) score higher than 1 is noted in 88%, and many patients have a poor performance status. Elevated lactate dehydrogenase levels are observed in 25% to 62%, low serum albumin concentration in 76% to 88%, and low hemoglobin level in 54% to 91% of patients. Abnormalities of the white blood cell count, renal function tests, erythrocyte sedimentation rate, C-reactive protein level, and alkaline phosphatase level are detected in more than a third of patients. A hemophagocytic syndrome may occur in 16% to 40% of cases.

Table 38-1
Pathologic, Molecular, and Biological Features of Refractory Celiac Disease (Types I and II) and Enteropathy-Associated T-Cell Lymphoma
Refractory Celiac Disease Type I Refractory Celiac Disease Type II EATL
Mucosa Villous atrophy and crypt hyperplasia Villous atrophy and crypt hyperplasia Villous atrophy and crypt hyperplasia
IEL number Increased Increased Increased
IEL morphology Normal Normal * Atypical/pleomorphic
IEL phenotype sCD3 + , cytCD3 + , CD8 + , sTCRαβ + , CD5(variable) + , CD30 sCD3 , cytCD3 + , CD8 , sTCRαβ , CD5 , CD30 sCD3 −/+ , cytCD3 + , CD8 +/− , sTCRαβ , CD5 , CD30 +/−
TCRγδ IELs Increased Decreased Decreased
TCR gene rearrangement Polyclonal Clonal Clonal
Lamina propria infiltration by neoplastic cells No Yes (up to 50% of cases) Yes
Peripheral blood involvement No Yes (44%-60%) Yes
Transformation to EATL (4-6 years) 3%-14% 30%-52%
EATL, enteropathy-associated T-cell lymphoma; IEL, intraepithelial lymphocyte; s, surface; cyt, cytoplasmic; TCR, T-cell receptor.

* IELs might show atypia in occasional cases.

Non-neoplastic (reactive) TCRγδ IELs.

Pathology

Macroscopic Appearance

The tumor may form ulcerating nodules, plaques, strictures, or, less commonly, large masses ( Fig. 38-1 ). The uninvolved mucosa might appear thin and show loss of mucosal folds. The mesentery is often infiltrated, and mesenteric lymph nodes are commonly involved. There is sometimes remarkably little macroscopic evidence of disease in the intestine as opposed to the mesenteric lymph nodes.

Figure 38-1, Resected jejunum from a patient with enteropathy-associated T-cell lymphoma shows multiple ulcerating tumors.

Histopathology

The histologic features of EATL show great variation both between cases and within a single case ( Fig. 38-2 ). The infiltrate is generally polymorphic, comprising numerous inflammatory cells, particularly eosinophils and plasma cells, which may be so great as to almost obscure the neoplastic lymphocytes ( Fig. 38-2 ). Large cell or anaplastic features are seen in 40% of cases. In some cases, the tumor cells can be monomorphic and display prominent central nucleoli (immunoblastic appearance). Angiocentricity and angioinvasion leading to destruction of blood vessels are not uncommon, and foci of necrosis are seen in a high proportion of cases. Increased numbers of histiocytes are often seen admixed with the neoplastic cells, and granulomas may be present, causing confusion with Crohn's disease. Increased mucosal and submucosal vascularity can occasionally be a prominent feature. Intraepithelial spread of tumor cells may be striking; but in some cases, only single scattered atypical lymphocytes are observed within the epithelium ( Fig. 38-3 ).

Figure 38-2, Histologic appearance of enteropathy-associated T-cell lymphoma.

Figure 38-3, Intraepithelial spread of enteropathy-associated T-cell lymphoma in the small intestinal mucosa.

The histology of the small intestinal mucosa remote from the tumor is an important consideration in the diagnosis of EATL. In most cases, the changes are identical to those of celiac disease, consisting of villous atrophy and crypt hyperplasia, lymphoplasmacytic infiltrates in the lamina propria, and increased IELs ( Fig. 38-4 ). As in uncomplicated celiac disease, the mucosal changes are maximal proximally and improve distally so that the lower jejunum and ileum may be normal. In some cases, the mucosal changes are less severe.

Figure 38-4, Small intestinal mucosa remote from enteropathy-associated T-cell lymphoma shows villous atrophy, crypt hyperplasia, and increased intraepithelial lymphocytes.

Numerous shallow ulcers extending into the submucosa are often present remote from the lymphoma. The ulcer bases contain an infiltrate of small lymphocytes and plasma cells, with an overlying acute inflammatory exudate ( Fig. 38-5 ). Episodes of ulceration followed by healing lead to scarring with stricture formation and distortion of the mucosal architecture, accentuated by destruction of the muscularis mucosa and the emergence of glands lined by cells of the ulceration-associated cell lineage, previously called pseudopyloric metaplasia ( Fig. 38-6 ).

Figure 38-5, A, Shallow mucosal ulcer in a case of enteropathy-associated T-cell lymphoma. Similar changes can also be seen in cases of ulcerative jejunitis. B, High magnification of the ulcer base shows granulation tissue and no evidence of lymphoma.

Figure 38-6, Healed ulcer in the non-lymphomatous small intestinal mucosa in a case of enteropathy-associated T-cell lymphoma. There is destruction of the muscularis mucosa, fibrosis of the mucosa and submucosa, and ulcer-associated lineage metaplasia of the intestinal crypts.

Lymph Node Involvement

The pattern of mesenteric (or other) lymph node involvement may be predominantly intrasinusoidal or paracortical, or both compartments may be involved ( Fig. 38-7 ). Lymph nodes remote from the EATL can show varying degrees of necrosis in the absence of a morphologically recognizable neoplastic cellular infiltrate ( Fig. 38-8 ). Abdominal (and extra-abdominal) lymph nodes may show a spectrum of changes ranging from lymphocyte depletion and fibrosis to dissolution of the node and replacement with lymph fluid ( Fig. 38-8 ), referred to as lymph node cavitation. At times, these lymph nodes can undergo calcification. Similar lymph node alterations can also be observed in individuals with long-standing untreated or refractory celiac disease, often accompanied by splenic atrophy. Lymphatic and blood vessel damage by the neoplastic (or non-neoplastic) cytotoxic intraepithelial T-cells trafficking to the lymph nodes and bystander killing of lymph node constituents possibly contribute to lymph node destruction.

Figure 38-7, Dissemination of enteropathy-associated T-cell lymphoma to a mesenteric lymph node.

Figure 38-8, Lymph node changes without overt evidence of enteropathy-associated T-cell lymphoma.

Immunohistochemistry

In most cases of EATL, the neoplastic lymphocytes express CD103, CD3 (cytoplasmic), CD7, T-cell intracellular antigen-1 (TIA-1), perforin, and granzyme B. They are usually negative for CD5, CD4 and CD8, and CD56, and similar to RCD II, many EATLs lack surface and cytoplasmic T-cell receptor (TCR) expression ( Fig. 38-9 ). However, these immunophenotypic features are not consistent. Cytoplasmic TCRβ chain (βF1) expression may be detected in approximately 25% of cases. Expression of TCRγ chain has also been reported in some cases ( Fig. 38-10 ). This might imply derivation from TCRγδ T-cells or lineage infidelity of a proportion of EATLs. CD8 expression has been described in 19% to 30% of EATLs overall, with a higher frequency (50%) reported in cases not associated with RCD II. In occasional cases, the neoplastic cells fail to express CD3, perforin, or CD103. CD30 expression varies in the different cytomorphologic variants, but almost all EATLs manifesting large cell morphology are CD30 + (see Fig. 38-9 ). Although, these tumors may resemble anaplastic large cell lymphoma, EATLs are anaplastic lymphoma kinase negative except in very rare instances. EATLs are Epstein-Barr virus (EBV) negative lymphomas, but admixed EBV-positive lymphocytes may be present. Virtually all cases display elevated Ki67 proliferation indices (>50%). Immunohistochemistry is useful in detecting single scattered neoplastic cells in cases lacking a mass lesion and also for discerning increased IELs, as evidence of celiac disease, when villous architecture of small intestinal mucosa–adjacent EATLs appears normal ( Fig. 38-11 ). The immunophenotype of the IELs in the uninvolved small intestinal mucosa may be normal in “de novo” EATLs, but in most of the cases preceded by RCD II, the IELs exhibit an aberrant phenotype (see later; see also Table 38-1 ).

Figure 38-9, A, Enteropathy-associated T-cell lymphoma showing angiocentricity and angiodestruction. B, The neoplastic cells express CD3. C, They lack CD8 expression. D, Virtually all cells show intense CD30 expression.

Figure 38-10, Enteropathy-associated T-cell lymphoma involving stomach.

Figure 38-11, A, Enteropathy-associated T-cell lymphoma with overlying mucosa composed of well-formed villi. B, Higher magnification of villi immunostained with CD3 shows increased numbers of intraepithelial lymphocytes.

Clinical Course

The clinical course of EATL is unfavorable, with a median survival of 7 months and 1- and 5-year overall survival of 31% to 39% and 0% to 59%, respectively. * Prognostic factors are not well established for this entity. Malnutrition, which is common in EATL patients with prior RCD II, is considered responsible for their markedly lower 5-year survival (0% to 8%) compared with those lacking prior RCD II (59%). Surgical resection is impractical in most cases because of involvement of multiple intestinal segments and dissemination of disease beyond the mesenteric lymph nodes. Most chemotherapeutic regimens are usually ineffective, resulting in only temporary remission. Better outcomes have been reported for patients treated with intensive chemotherapy followed by autologous stem cell transplantation (5-year overall and progression-free survival of 60% and 52%, respectively). Death is usually attributed to EATL, malnutrition, or infectious complications.

* References .

Pathogenesis

Postulated Cell of Origin

EATL is derived from intraepithelial T-cells on the basis of shared immunophenotypic features, including integrin α E β 7 (HML-1, CD103) expression. IELs comprise a phenotypically heterogeneous population of T-cells and other innate lymphoid cells. Many EATLs are thought to arise from the neoplastic transformation of thymus-derived intraepithelial T-cells that have rearranged TCRβ chain genes, express the CD8αβ heterodimer (CD8αβ) and TCRαβ, and exhibit major histocompatibility complex class I restriction. These lymphocytes, which are referred to as conventional, type a, or induced IELs, account for 80% of all human small intestinal IELs. They have latent cytotoxic potential (TIA-1 + , perforin , granzyme B ) and migrate to the intestine in response to antigen stimulation. T-cells that have rearranged gamma delta but not beta chain genes (TCRγδ) account for up to 15% of IELs, and the majority display a “double negative” (CD4 CD8 ) phenotype. Studies suggest that a subset of EATLs might also be derived from TCRγδ T-cells; however, the frequency of such cases is presently unknown.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here