Nodal and extranodal lymphoproliferative disorders in sjogren's syndrome: A clinical and immunopathologic study

Nodal and extranodal lymphoproliferative disorders in sjogren's syndrome: A clinical and immunopathologic study

Nodal and Extranodal Lymphoproliferative Disorders in Sjogren’s Syndrome: A Clinical and lmmunopathologic Study THOMAS 1. McCURLEY, MD, R. DEAVER COLL...

6MB Sizes 0 Downloads 23 Views

Nodal and Extranodal Lymphoproliferative Disorders in Sjogren’s Syndrome: A Clinical and lmmunopathologic Study THOMAS 1. McCURLEY, MD, R. DEAVER COLLINS, MD, EUGENE BALL, MD, AND ROBERT D. COLLINS, MD Sjogren’s syndrome (SS) is frequently associated with both reactive and neoplastic lymphoproliferative disease. Over a 12-year period beginning in 1970, 21 of 138 patients with SS followed at two tertiary university medical centers had biopsies taken of enlarged lymph nodes (18) or extranodal lymphoid infiltrates (8). Many had immunologic studies performed on fresh tissue and all had paraffin-embedded tissue available for histochemical and immunoperoxidase studies. Eight of our patients had malignant lymphomas which were chiefly B cell neoplasms including two lymphoplasmacytic lymphomas and two follicular center cell lymphomas. The remaining 13 patients had either reactive adenitis (usually with follicular hyperplasia) or atypical lymphoid hy perplasia which failed to meet both histopathologic and immunopathologic criteria for malignancy. None of the nine patients with reactive hyperplasia has yet progressed to lymphoma, while one of four patients with atypical lymphoid hyperplasia progressed to overt lymphoma. Clinical features such as age, duration of disease, extent of lymphadenopathy, splenomegaly, or parotid swelling failed to identify those subsets of patients with lymphadenopathy at increased risk for lymphoma. Recognition of lymphoma in two patients was greatly facilitated by tissue immunologic studies demonstrating focal areas of monotypic B cell proliferation. In one patient in whom the histopathologic diagnosis was immunoblastic sarcoma of B cells, tumor cells were L26-negative and strongly UCHLl-positive suggesting T cell differentiation. In three patients with relatively homogeneous extranodal lymphoid infiltrates, B cell polyclonality on tissue immunoperoxidase studies, and the absence of cytologic atypia, precluded a diagnosis of malignant lymphoma; none of these three patients has progressed to overt lymphoma. Our results indicate that (1) patients with SS develop a variety of B cell lymphomas and other lymphoproliferative disorders, and (2) the nature of the lymphoproliferative disorder is best determined by multiparameter analysis including immunologic phenotyping. HUM PATHOL 21:482-492. 0 1990 by W.B. Saunders Company. The diagnosis of Sjogren’s syndrome (SS) is based on the presence of at least two components of a clinical triad that includes keratoconjunctivitis sicca, xerostomia, and a connective tissue disease, usually rheumatoid arthritis (RA).’ Lymphoproliferative dis-

From the Department of Pathology, Divisions of Immunopathology and Hematopathology, Vanderbilt University School of Medicine, Nashville, TN, and the Department of Medicine, Division of Rheumatology and Immunology, University of AlabamaBirmingham, Birmingham, AL. Accepted for publication September 1. 1989. Key words; Sjogren’s syndrome, malignant lymphoma, pseudolymphoma, reactive adenitis. Address correspondence and reprint requests to Thomas L. McCurley, MD, Department of Pathology. Vanderbilt University School of Medicine, 21st and Garland Ave, Nashville, TN 37232. 0 1990 by W.B. Saunders Company. 0046-H 17719012 105-0004$5.00/0

orders are the most frequent of the serious complications of SS and may involve lymph nodes or extranodal sites, particularly lung and salivary gland.2-s Based on histopathologic features, lymphoid proliferations in SS may be characterized as (1) reactive changes, (2) malignant lymphomas (ML), and (3) atypical lymphoid hyperplasia or pseudolymphomas.” Clinical features associated with the development of lymphoma include increasing adenopathy, parotid enlargement, splenomegaly, and a fall in rheumatoid factor and IgM levels.2,7 Pathologic criteria for diagnosis of ML have recently been modified by the availability of immunologic studies for demonstration of monotypism, facilitating recognition of lymphomas which appear to be heterogenous lymphoid proliferations by light microscopy.“,” While there are numerous case reports of lymphoproliferative disorders complicating Sjogren’s syndrome,“J-‘2 the most comprehensive clinical pathologic analysis of lymphoproliferation in SS was based on a group of 136 patients followed at the National Institutes of Health (NIH) from 1954 to 1975. Ten of these patients developed ML including three patients with Waldenstrom’s macroglobulinemia.’ Four of these lymphomas were included in an immunopathologic study which showed that most lymphomas in Sjogren’s syndrome are monoclonal B cell lymphoproliferaneoplasms. ” While nonneoplastic tive disease was described in early reports of some of the NIH patients, 2~.7the last NIH study did not include an analysis of nonneoplastic lymphoid proliferation. The purpose of our study was to comprehensively examine both reactive and neoplastic lymphoproliferation in patients with Sjogren’s syndrome using recently available immunopathologic techniques and monoclonal antibodies. We reviewed the clinical and pathologic records of 138 patients with Sjogren’s syndrome seen at two tertiary care university medical centers from 1970 to 1982. Twenty-one of these patients had biopsies of nodal or extranodal lymphoid infiltrates. Eight of our patients had ML, nine had reactive lymphoid hyperplasia, and four had atypical lymphoid hyperplasia. Our group of patients constitutes the largest single clinical pathologic analysis of neoplastic and nonneoplastic lymphoproliferative disease complicating Sjogren’s syndrome. Diagnosis of both lymphomas and nonlymphomatous reactions was facilitated by immunophenotypic studies.

LYMPHOPROLIFERATIVE DISORDERIN SJOGREN’S(McCurleyet al)

MA1ERiAl.S AND METHODS

Classification

Patient Population

Patients were classified into three groups according to their i&al biopsy: Rsactiw infiltmtes. Sections from these patients demonstrated architectural preservation usually with follicular hyperplasia. lmmunopathologic studies failed to demonstrate monotypism by restriction of immunoglobulin heav! chains and light chains. Malignant lymphoma. Sections showed effacement of architecture by lymphoid infiltrates, frequently accompanied by additional features of malignancy such as necrosis, cytologic atypia, and invasion of perinodal tissue. Immunologic criteria for neoplasia included the demonstration of monotypism by immunoperoxidase evaluation of immunoglobulin heavy and light chain distribution. Malignant lymphomas were categorized by the Lukes-Collins classification.” Atypical lymphoid hyperplasia. Sections showed partial or total effacement of normal architecture by infiltrates that were relatively homogeneous by light microscopy. However, cytologic atypia and necrosis were not present. Immunopathologic analysis failed to demonstrate monotypic cell populations in all cases studied.

The study population consisted of 138 patients meeting criteria for primary SS; these patients had been followed at Vanderbilt Hospital, Nashville VA Hospital, and the University of Alabama at Birmingham from 1970 to 1982. Patients with primary Sjogren’s syndrome had keratocon.jullcti\~itis sicca and xerostomia while those with secondary Sjogren’s had an associated connective tissue disease. Thle study group included 2 1 patients who had biopsies confined either to lymph nodes alone (13 patients), extranodal lymphoid infiltrates (3 patients), or both extranodal lymphoid infiltrates and lymph nodes (5 patients). Patients with pre-existing lymphotna, sarcoidosis, or other diseases known to mimic SS were excluded from the study. One of these patients has previously been described in the literature. “I Evaluation of each patient included review of the medical records and all histopathologic material from biopsies and autopsy material as well as immunologic studies from biopsy sprdimens. Clinical criteria for vasculitis included palpable purpura, mononeuritis multiplex, and multiple or punched out cutaneous ulcers. Tissue immunologic studies were performed on paraffin-embedded blocks in selected cases.

RESULTS Demographic and Clinical Features

Histopathology Five micron paraffin-embedded stained with hematoxylin and eosin. and methll grren pyronin stains.

Major clinical and demographic features are summarized in Table 1. Females predominated in all Thirteen patients had prihistopathologic groups. mary SS. Eight patients had secondary SS with RA. In these patients sicca symptoms followed the onset of RA by a mean of 8 years (range 3- I.3 years). The patients with ML were slightly older with a somewhat greater duration of disease prior to biopsy than the groups with reactive infiltrates and atypical lymphoid hyperplasia. Lymphadenopathy invol\.ing three 01 more anatomic sites was present in five patients with reactive infiltrates, five patients with XIL, and one patient with atypical lymphoid hyperplasia. Splenomegaly was more common in patients with ML and atypical lymphoid hyperplasia. The patients with re-

tissue sections were periodic acid-Schiff,

lmmunopathology Paraffin-embedded tissue was stained using a peroxiclase-antiperoxidase technique by methods previously described. I5 PI-imarv antibodies included monospecific goat antihuman antibodies directed against kappa and lambda light chains. IgG. IgM. and IgA heavy chains, lysozyme and the monoclonal antibodies LNl (follicular B cells), LN2 (pan B).‘” LCA (pan leukocyte), If UCHLI (T cell restricted patients CDG), Ix and L26 (pan B).‘” In selected Vanderbilt immunologic typing studies were performed on cell sus1)ensions of. frech tissue hy methods previously described.“” TABLE 1.

Summary of Clinical and Demographic Features in Patients With Lymphoproliferation Complicating Sioaren’s Svndrome Atypical Reactive Infiltrates (n = 9)

Group

hlalignant Lymphoma (n = 8)

__.__ Sex (M/F) Primarv Sicqqwl‘s syndrome Sledian age in v~2rs at hiops\

7.1

9.x

:3 .i x (,5OMi) i.0

(I-20)

(5-2))

(I-Ifi)

:i 60

5 59

(24.73) hfean duration in yars of rheumatologic disease at hiopsv (range) Lymphadenopathv of 3 01 mire sites Splenomegal\ Parotid swelling \‘asculitis ~rher clinical features in individual patients

014

915

2iT (range)

I.vmphoid Hvperplasia 111 = 1) ---~

5 2 1 3 Felty’s (n = I) pneumonitis (n = 1) nwropathy (n = I) rhvroiditis (n = I) hyperviscosit~ svndrome (n = 2)

_-___.-.

483

(46.69)

I :5 ?I

.: 4 4 ?I

pneumonitis

I (n = 1)

leukopenia (n = 1) neuropathy (n = 1) hvperviscosity (n = 1) __-

Volume 21, No. 5 (May 1990)

HUMAN PATHOLOGY

trafollicular plasma cells were prominent in four patients. The interfollicular areas were expanded in four patients with focal dermatopathic changes. Sinuses were open, containing varying numbers of macrophages in all cases and granulocytes in four cases. Hyperplasia of parafollicular IS cells was not identified. Interfollicular plasma cells were prominent in eight biopsies from six patients. In the two patients with sequential lymph node biopsies, the second biopsy in each showed architectural preservation with follicular hyperplasia and interfollicular plasmacytosis (cases 7 and 9). In patient 7, open lung biopsy at the time of the second lymph node biopsy demonstrated an interstitial lymphoplasmacytic infiltrate compatible with lymphoid interstitial pneumonia. Two patients (cases 2 and 4) had concurrent biopsies of the superficial parotid gland which showed myoepithelial sialadenitis. In patient 2 there was proliferation of small cells with abundant cytoplasm around myoepithelial islands (Fig 1). Lip biopsies in patients 3, 5, 6, 7, and 9 showed focal lymphocytic infiltrates supporting a diagnosis of SS.

active infiltrates and ML were similar in the frequency of vasculitis and parotid swelling. Rheumatoid factors (RF) were present as follows: reactive infiltrates, 6 of 9; ML, 4 of 8; atypical lymphoid hyperplasia, 2 of 4. Antinuclear antibody-positivity was detected in: reactive infiltrates, 4 of 9; ML, 4 of 8; atypical lymphoid hyperplasia, 3 of 4. Two patients with ML and three patients with reactive infiltrates had significant decreases (twofold or greater) in RF titer under observation, but these were not temporally related to changes in adenopathy or development of ML. REACTIVE INFILTRATES Histopathology All eleven lymph node biopsies from nine patients showed preservation of architecture (Table 2). Capsules were normal in width in nine biopsies with two showing capsular thickening. Nine biopsies from seven patients demonstrated follicular hyperplasia with large follicles present in both the cortex and medulla. A biopsy from one patient demonstrated an admixture of small atrophic follicles. Distinct mantle zones were present in lymph nodes from all patients with follicular hyperplasia. Prominent follicular zoning was noted in five biopsies from four patients. InTABLE 2.

nodal

Immunologic studies on cell tissue in three cases showed

suspensions polyclonal

of lym-

Clinical and lmmunopathologic Features in Patients With Reactive Infiltrates Disease Durationi (yr) at Biops!

Ages/Sex

Clinical Diagnosis

1

51/F

ss-1

2

2

73IF

ss-1

1

36/F

SS-1

6

24/M

ss-1

.5

s-2

6

Patient

lmmunopathology

9-2

ss-1

90 P

52lF

SS-2

I?

ti’L/F

ss-2

IO

Pathologic Diagnosis LN: Follicular hyperplasia LN: Focal dermatopathia parotid: myoepithelial sialadenitis LN: Folliculal hyperplasia LN: Follicular hyperplasia parotid: myoepithelial sialadenitis LN: Follicular hyperplasia LN: Follicular hyperplasia LN(2): Follicular hyperplasia Lung: Lymphoid interstitial pneumonitis LN: Sinus histiocytosis LN(2): Follicular hyperplasia

Abbreviations: SS-1. primal-v Sjogren’s syndrome; SS-2, secondary paraffin immunoperoxidase. * Age at initial biopsy,. + Time from onset of rheumatologic disease to biopsy. $ A. alive; D. deceased; follow-up in number of years.

484

Sjogren’s

Immunologic Studies

Status$ (Cause of Death)

Poloclonal

(CS. PIP)

Polyclonal

((2% PIP)

A. $1

Pol~clonal

((3)

A. 7

Polyclonal

(PIP)

D, 6 (Bronchiectasis) A, 9

I). I (Lung

Polyclonal

(PIP)

Polyclonal

(PIP)

l’ol,clonal

(PIP)

(Sepsis,

cancer)

D. 5 Lrukopenia) A, 5

A, 7 Polyclonal

syndrome:

(PIP)

LN, lymph

D. 2 (Intracranial hemorrhage) node:

(:S, cell

suspension; PIP,

LYMPHOPROLlFERATlVE DISORDER IN SJOGREN’S [McCurley et al)

FIGURE 1. Myoepithelial sialodenitis with proliferation areas. (Left) Parotid gland from case 2 showing confluent areas of cellular proliferation around myoepithelial island (Hematoxylin and eosin stain; magnification x 74,) [Right] The cells have central nuclei with abundant clear cytoplasm and resemble monocytoid or parafollicular B cells. (Hematoxylin and eosin stain; magnification x 740.)

6), Iymphopfasmacytic (cases 1 and X), large cleaved cell (case 3). large B cell (cases 4 and 5). one intermediate grade lymphoma of undefined Immunologic phenotype (case 7). and one large-cell immunoblastic lymphoma (case 2. Table 3). ‘I-he diagnosis of’ seven of’ the eight lvmphomas was made on lvmph node biopsy while c&e 4 had a large noncleavbd follicular center cell (FCC) lymphoma at thyroidectomv. Four of’ the fymphomas showed partial lymph node involvement on initial biopsy; in the large cell immunoblastic lvmphoma (case 2) there was focal

f>hoid pojpiilations. Stains fi)r intracytoplasmic immiinoglobulin in eight biopsies f‘rom six fjatients re\,ealed pofvt,pic marking in plasma cells.

MALIGNANT LYMPHOMA Histopathology ~I‘he l~mf~hontas histologically ranged from indolent to high-grade and were small B cell in tvpe (case

-~~~

I’atirnt -_______

TABLE

AgrlSes

3.

Clinical

(Xnical Diagnosis ss-I

and lmmunopathologic Disease Dut-arion in Years at Inittal Uiopsk ti

SS-?

ss-L’

ss- I

ss-I ss- I ss-2

ss-1

Ii

Features

in Patients With Malignant Lymphoma

Pathologic Diagnosis LN: ML. L~mphoplasmaotic type 1.X: XII.. large-cell immunohlastic 1.X: ML. probable per-ipher-al T cell (postmortem. 3 1110 aftu initial hiopsv) LX: ML. FCC:. I.arge cleaved cell Thyroid: XIL. FCX:. Lar-ge nrmcleaved rell. Hashimoto’r thr-roiditi5 LN: ML. large transformrd cell LN: Small H cell lvmphoma 1.x: ML. tvpc unc crtaill LN: ML. Lvmpboplasmas ytic- tape. parotit!: mwepithelial sialadenitis (2 \r3 prior to LN hiopsv)

485

Immunologic~ S1udies Monotypic IgM-K (PIP) Polvclonal (PI 1’. (25) CrCHL I equivocal, I.%-negative L~(:HLl positive (PIP), LTti-negkati\e

sr
I). 8 mphoma) D. I

~I\rrlphoma)

,A 7 A. 7

I,gWK (PIP) LN 1-negativr, LNZ-positive Ighl-K (PIP) I.(:.\-positi\

e (1’11’)

Polykmal (CS) IgM-K (PIP)

1). :i il~tnphoma)

Volume 21, No. 5 (May 1990)

HUMAN PATHOLOGY

lymph node involvement with tumor cells principally located in follicular and parafollicular areas (Fig 2, top left). Neoplastic cells in this case resembled immunoblasts with large nuclei and prominent nucleoli associated with abundant amphophilic cytoplasm (Fig 2, top right). The apparent homing of tumor cells to follicular centers and their cytologic features suggested histopathologically that this was an immunoblastic sarcoma of B cells, but that diagnosis was not supported by immunologic studies. Postmortem material on this patient demonstrated lymphoma involving all lymph node groups as well as the lung, vertebral column, peritoneum, pleura, and pericardium. In case 7 there was focal involvement of the lymph node cortex by an infiltrate of irregular small and large cells associated with abundant cytoplasm. In cases 1 and 8 there was interfollicular expansion with a heterogeneous infiltrate of plasma cells, lymphocytes, plasmacytoid lymphocytes, and macrophages (Fig 3). A follow-up biopsy in patient 8 showed progressive interfollicular expansion with retention of follicular centers. Biopsy of the superficial parotid in this patient 2 years prior to the initial lymph node biopsy showed typical myoepithelial sialadenitis without areas of cellular proliferation.

lmmunopathology Immunologic studies confirmed the B cell nature of six of the eight lymphomas. Four stained monotypically for IgM heavy chain and K light chain. Interestingly, in the two lymphoplasmacytic lymphomas, (cases 1 and 8), the interfollicular infiltrate was monotypic (Fig 3 bottom left and right), while the residual follicular centers marked polytypically. and cell suspension studies in case 8 were polytypic as well. The large cleaved FCC lymphoma showed distinct follicular nodulation and marking of tumor cells with LNl and LN2. In case 2 cell suspension studies revealed polyclonal B cell marking of lymph node cells. Tumor cells were L26-negative by paraffin immunoperoxidase and showed equivocal positivity for UCHLl. Immunoperoxidase studies of postmortem tissue showed the tumor cells were L26-negative and strongly UCHL 1-positive, suggesting that this neoplasm was a peripheral T cell lymphoma. (Fig 2, bottom left and right). In case 4 tumor cells marked with LCA and were negative for K and A light chains. LN 1, LN2 and L26 positivity indicated that this process was a B cell lymphoma. Case 7 did not mark as a B cell lymphoma,

FIGURE 2. Large-cell lymphoma. probable peripheral T-cell type. (Top left) This lymph node from case 2 demonstrates focal involvement by malignant lymphoma (arrows) surrounding hyperplastic follicular centers. (Hematoxylin and eosin stain; magnification x 73.) (Top right) Tumor cells exhibit large nuclei with prominent nucleoli associated with abundant amphophilic cytoplasm. (Hematoxyin and eosin stain; magnification X 280.) (Bottom left] Tumor cells from postmortem lymph node stained with antibody to UCHLI. (Bottom right) Tumor cells did not stain with L26. (Magnification X 444.)

486

LYMPHOPROLIFERATIVE DISORDER IN SJOGREN’S (McCurley et al)

FIGURE 3. Malignant lymphoma, lymphoplasmacytic type. (Top left] Residual follicular centers (arrows) are present in case 8 with intervening expansion of interfollicular areas. (Hematoxylin and eosin stain; magnification x 40.) (lop right) There is a heterogeneous infiltrate of lymphocytes, macrophages. and plasma cells. (Hematoxylin and eosin stain; magnification *I 740.1 (Bottom left] Plasma cells were shown by immunoperoxidase procedures to contain monotypic kappa light chains. Compare with bottom right. (Magnification x 1000.) (Bottom right) Very few cells marked in the anti-lambda preparation. Monotypic marking in the interfollicular areas shown was associated with polytypic marking in adjacent reactive follicular centers [not shown). [Magnification A 1000).

487

Volume 21, No. 5 (May 1990)

HUMAN PATHOLOGY

and immunoperoxidase to antibodies against LN2.

studies showed immunoglobulins,

showed cytologically phangitic extension

no reactivity LN 1, and

atypical cells or pulmonary of the lymphoid infiltrate.

lym-

lmmunopathology ATYPICAL LYMPHOID HYPERPIASIA

In case 1, initial diagnostic material was unavailable for typing studies. Subsequent biopsies with immunologic typing studies on cell suspensions and paraffin-embedded tissue documented a monotypic IgM-K bearing lymphoid population. In case 2 the plasma cells marked predominantly for Ig:G-K, but the admixture of lambdaand IgA-bearing plasma cells indicated polytypism. In cases 3 and 4 tissue immunoperoxidase studies demonstrated polytypic marking of plasma cells.

Histopathology Four patients failed to meet all histopathologic and immunopathologic criteria for neoplasia on their initial biopsy (Table 4). Case 1 had partial effacement of lymph node architecture with residual follicular centers and focal collections of interfollicular and sinusoidal cells characterized by small nuclei and clear cytoplasm resembling parafollicular B cells (Fig 4, top left and right). The extent of this parafollicular proliferation was not sufficient for a diagnosis of malignant lymphoma. Subsequent lymph node biopsies were diagnostic of ML as they demonstrated architectural effacement by a monomorphic infiltrate of large cells with round nuclei, small nucleoli, and abundant cytoplasm. While this lymphoma had areas suggestive of a sinusoidal distribution, the cell cytology, vascular proliferation, and absence of residual follicles precluded a diagnosis of a parafollicular or monocytoid B cell lymphoma (Fig 4, bottom left). In case 2 an extranodal lymphoplasmacytic infiltrate was demonstrated in the submucosal tissue of the cheek. Lip biopsy in the same patient showed a focal lymphocytic infiltrate compatible with SS. Case 3 showed extensive alteration of lymph node architecture by a dense infiltrate of small lymphocytes mixed with plasma cells and macrophages; lung biopsy in this case showed multiple noncaseating granulomas containing polarizable material. The granulomas were associated with a lymphoid infiltrate similar to that in the node. In case 4 the normal lung architecture was obliterated by dense infiltrates of small lymphocytes with scattered plasma cells and macrophages. No identifiable follicular centers were present in the lung in cases 3 or 4. Neither case TABLE 4.

Patient 1

Age/Sex

Clinical Diagnosis

Disease Duration in Years at Initial Biopsy

50/F

ss-2

Iti

62/F

ss-

3

55/F

ss-1

65/F

Abbreviations: CS, cell suspension;

Four of nine patients in the reactive group have died, with their rheumatologic syndrome a major contributing factor to death in three patients. One patient had bronchiectasis with recurrent pulmonary infection, hemorrhage, and ultimately developed right heart failure, while the second patient had Felty’s syndrome and died with sepsis. The third patient developed hyperviscosity and died with intracranial hemorrhage. Five of eight patients with malignant lymphoma are dead. Four died of progressive lymphoma while one recently developed an intrathoracic poorly differentiated carcinoma. The remaining three patients are alive, although case 8 has required chemotherapy for a type I cryoglobulin associated with the lymphoma. In the abnormal immune reaction group, case 1 developed lymphoma and died of progressive disease. The remaining three patients are stable at a follow-up interval of up to 5 years. Case 3 had multiple monoclonal paraproteins with antibody activity directed against intermediate filaments. This patient was treated with cyclophosphamide and prednisone and is stable clinically on follow-up.‘4

Clinical and lmmunopathologic Features in Patients With Atypical Lymphoid Hyperplasia

2

4

OUTCOME

ss-1

SS- 1, primary PIP, paraffin

I

5 6

1

Sjogren’s syndrome; immunoperoxidase.

Pathologic Diagnosis

Immunologic Studies

LN: Atypical lymphoid hyperplasia LN: ML, large transformed cell predominant ( 1 Year after inittal biopsy) Cheek: Plasmacytic infiltrate LN: Atypical lymphoid hyperplasia Lung: Multiple granuiomas, Atypical lymphoid hyperplasia Lung: Atypical lymphoid hyperplasia SS-‘1 -, secondary

488

Sjogren’s

syndrome:

(Cause

-

Status of Death)

D, 2 (Iymphoma)

IgM-K (CS), LNI-positive (PIP)

Predominantly IgG-K (PIP) Polytypic (PIP) Polytypic

(PIP)

Polytypic

(PIP)

LN, lymph

node;

ML. malignant

A, 3 A, 5

A .9

lymphoma;

LYMPHOPROLIFERATiVE DISORDER IN SJOGREN’S (McCurley et al)

FIGURE 4. Atypical lymphoid hyperplasia with progresslon to malignant lymphoma. nap left] In the first lymph node biopsy from case 1 with atypical lymphoid hyperplasia, there is partial effacement of architecture in which sinuses are distorted by adjacent focal proliferations of parafollicular B cells (Hematoxylin and eosin stain; magnification x 275.) (lop right] Nodule of parafollicular B cells and adjacent sinus. (Hematoxylin and eosin stain; magnification x 689.) [Bottom left) Second node biopsy from case 1 in which there was diffuse architectural obliteration by mamalignant lymphoma composed of monomorphic cells with large nuclei, small nucleoli, and clear cytoplasm. (Periodic acidschiff stain; magnification x 727). This lymphoma marked as a B cell process and was judged to be intermediate in grade.

DISCUSSION

three tertiary care university hospitals. We used monoclonal antibodies and immunopathologic techniques not available at the time the NIH and Mayo Clinic series were reported. As in previous clinical studies of lyrnphoproliferative disorders in SS, a majority (13 of 21 patients) had primary SS, while eight had secondary SS with RA. Patients with lymphoma, reactive adenitis, and atypical lymphoid hyperplasia were similar in age, sex, and duration of disease prior to biopsy. Earlier clinical studies have suggested that clinical features such as splenomegaly, parotid enlargement, or falling rheumatoid factors presage the development of lymphoma,“,’ findings we did not confirm in our patients. Distinct clinical subsets of patients with lymphadenopathy who were more likely to have lymphoma were not identified. Most of the nine patients with reactive infiltrates had striking follicular hyperplasia. The triad of follicular hyperplasia, interfollicular plasmacytosis, and sinusoidal neutrophils previously described in RA was present in three patients, two with primary SS and one with secondary Necrosis (often obSS. y“z.l served in systemic lupus erythematosus) was not present.‘” Hyperplasia of parafollicular B cells (monocytoid B cells) has been noted in pseudolymphomas in SS and toxoplasmosis but was not present in lymph nodes in this group. In one case a concurrent parotid biopsy showed proliferation of cells re-

Lymphoproliferative disease is a major cause of morbidity and mortality in SS, as established by the reports of ‘I‘alal, Anderson, Kassan and coworkers.“.“,’ ‘I‘hey followed 136 women with sicca svndrome at the NIH over a 22-year period and identified 10 patients who developed lymphoma. Four of the NIH patients had tissue immunologic studies reported in a concurrent study that showed lymphomas in SS were predominantly B cell neoplasms.‘:’ In a 1976 Mayo Clinic study of the pulmonary manifestations of SS, eight of 343 patients had pulmonary lymphoproliferative disease including three with lymphoma, three with lymphoid interstitial pneumonitis, and two with pseudolymphoma (one of these had a subsequent extranodal lymphoma of bowel).” The lymphomas were not subclassified nor were tissue immunologic studies performed. Since these two series were reported, the diagnosis and classification of lymphoproliferative disease has been facilitated by the availability of’ immunologic methods for demonstration of nlor~otypism. As th’ese previous large series were limited in their evaluation of reactive lymphoid proliferations and in the application of tissue immunologic studies, we undertook a comprehensive clinical and immunopathologic analysis of both reactive and neoplastic lymphoproliferation in patients with SS followed at 489

Volume 21. No. 5 [May 1990)

HUMAN PATHOLOGY

sembling monocytoid or parafollicular B cells around myoepithelial islands. r2z5 No patient exhibited the exuberant follicular hyperplasia of human immunodeficiency virus infection.*” Finally the follicular hyperplasia was not so extensive as to produce architectural distortion and thereby be confused with follicular lymphoma. As expected from the histopathologic findings, immunopathologic studies documented polytypic typing for immunoglobulin heavy and light chains in both cell suspensions (three patients) and paraffin immunoperoxidase (seven patients). None of these patients progressed to overt malignant lymphoma (mean follow-up 7.4 years for living patients). However, three have died from complications directly related to their underlying rheumatologic disease. Eight patients had ML on their initial biopsy material. These lymphomas were notably heterogeneous in type with four indolent lymphomas, and four intermediate- to high-grade lymphomas. As reported previously in SS, most were B cell neoplasms;4.‘” four lymphomas marked monotypically for IgM and K. Two other cases demonstrated staining with LN 1, LNZ, or L26-three putative B cell markers.‘“,‘” The two lymphoplasmacytic lymphomas were initially thought to be abnormal immunologic reactions as sections showed partial architectural effacement by a very heterogeneous infiltrate of lymphocytes, plasma cells, and macrophages. In one of these, cell suspension studies showed a polytypic lymphoid population. However, immunoperoxidase studies demonstrated monotypic marking (IgM-tc) of the interfollicular infiltrate, while B lymphocytes in follicular centers in both cases marked polytypically. Therefore, tissue immunoperoxidase studies facilitated diagnosis of these two cases as indolent lymphomas rather than as atypical lymphoid hyperplasias. While case 2 was initially thought to have an immunoblastic sarcoma of B cells, paraffin immunoperoxidase studies on the postmortem material showed distinct UCHL 1 positivity in tumor cells; these findings indicate this patient had a peripheral T cell lymphoma, which has only rarely been described in Sjogren’s syndrome.” Patients with ML generally had a shorter length of survival than those with reactive infiltrates or abnormal immune reactions. Five patients have died and ML was a major contributing cause in four of these deaths. Another patient (case 8) with a lymphoplasmacytic lymphoma remained stable for 3 years until treatment was required for a monoclonal cryoglobulin (cryocrit 15%) with vasculitis. Four patients were diagnosed as having atypical lymphoid hyperplasias. One of these patients had partial alteration of nodal architecture with focal collections of parafollicular B lymphocytes similar to those previously described in pseudolymphoma in SS.” Unfortunately tissue was unavailable for immunologic studies. Subsequently this patient developed an aggressive B cell lymphoma and died of disseminated disease. In another case a soft tissue plasmacytic infiltrate showed a predominance of cells mark490

ing with IgG-K, a finding suggestive of an evolving lymphoid neoplasm. However, follow-up has not revealed evidence of overt lymphoma. The remaining two patients had extensive alteration of pulmonary architecture by dense lymphoid infiltrates that did not contain follicular centers. Cytologically atypical lymphoid cells were not present, and neither showed the lymphangitic spread of lymphoid cells described studin pulmonary lymphomas .8 Immunoperoxidase ies showed polytypic marking. The subsequent course of these two patients has not revealed clinical progression, although one was treated with chemotherapy. l-1 The reason for the increased incidence of malignant lymphoma in SS is unknown. Longitudinal studies in many patients with SS have documented a stepwise evolution to B cell lymphoma from antecedent reactive lymphoid proliferations.2~1~~z7~z* New Zealand Black mice have a lymphocytic sialadenitis similar to that seen in SS, and have a similar susceptibility to B cell lymphomas following reactive lymphoid hyperplasias. 29-31 This mouse model shows other features of B cell activation including polyclonal hypergammaglobulinemia with the production of multrple autoantibodies.3z The reason for polyclonal B cell activation in SS is unknown. There is little evidence suggesting a primary B cell defect. Unlike patients with systemic lupus erythematosus, patients with SS apparently do not have increased numbers of B cells spontaneously secreting immunoglobulin in either bone marrow or peripheral blood.J” Patients with SS do have abnormalities in the number and function of immunoregulatory T cell subsets including a significant reduction in circulating CD8 (suppressoricytotoxic) ‘I’ cells3’ The immunohistologic composition of lymph node and salivary glands was studied in three patients with pseudolymphoma in SS, with one of these patients developing an overt lymphoma. Many of the cells in these reactive infiltrates had a T helper (CD3, CD2, CD4) phenotype and bore the lymphocyte activation markers HLA-DR and T10.“5 On the basis of these studies, Fox, et al proposed that the primary defect in SS predisposing to lymphoma was excessive T helper activity with chronic B cell stimulation and eventual escape of neoplastic B cell clones.- 75 Recently, Fishleder, Tubbs, et al have shown that histopathologically benign lymphoid infiltrates in salivary tissue from patients with SS exhibit clonal rearrangements of immunoglobulin heavy chain and light chain genes.3” This may represent an early step in progression to overt lymphoma. Focal monotypic areas of B cell proliferation in salivary glands with myoepithelial sialadenitis may identify patients at increased risk for development of lymphoma. 27 Monomorphic B cell proliferations have been described in salivary glands with myoepithelial sialadenitis composed of “centrocyte’‘-like cells with light chain restriction on tissue immunologic studies.“’ These areas of “early lymphoma” are morphologically and phenotypically similar to lymphomas of mucosal-associated tissues8 and to parafollicular or monocytoid B cell lyn~phomas.3”,“0 The latter

LYMPHOPROLIFERATNE DISORDER IN SJOGREN’S (McCurley et al)

I~mphomas have recently been well-described in patlents with SS.-“,“9 In our own patients we found proliferation areas:17.:‘5 m . a parotid biopsy of one patient who has nor been treated and has not developed any clinical evidence of lymphoma after a Y-year follo\vup. A second patient with a parafollicular B cell proliferation in lymph node developed an aggressive B cell lympboma within 1 year of the initial biopsy. This may represent transformation of the process we called al.ypical lymphoid hyperplasia in the first node. While this case and the studies cited4”‘,+’ suggest that lymphomas in SS may evolve from monocytoid B cell proliferations, the morphologic and immunophenotypic diversity of lymphomas in our series suggests other pathways in the development of lymphoid neoplasms in SS. Analysis of’ patient tissue in SS requires both histopathologic and immunopathologic techniques for diagnosis. OUT studies show that recognition of lymphomas that are heterogeneous or only focally involve no’des is facilitated by tissue immunopathologic studies to demonstrate their monotypism. On the other hand, dense, relatively homogeneous proliferations of‘small lymphocytes, as demonstrated in three of our patients, may be immunologically heterogeneous-a finding that may preclude diagnosis of malignant Iymphoma. An aggressive approach with early biopsy of nodal and extranodal lymphoid infiltrates in patients with SS is warranted by the frequency of’ lymphoma: ‘3.5 Nearly half of our patients undergoing biopsy had ML, but analysis of clinical features did not predict this complication. Patients with follicular hyperplasia on initial biopsy appear to be at low risk tor subsequent development of lymphoma. Patients with previously diagnosed atypical lymphoid hyperplasia or low-grade Iymphoma should be given a repeat biopsy in the setting of inL.reasing or new adenopathy, because the former group may develop overt lymphoma, and the latter high-grade Ivn~phoma.‘~12~9i .~cknozc'le~l~rrlerLf‘I‘he authors ca1assistalice of‘ Harriet Davis.

acknowledge

the techni-

REFERENCES I. B1oc.h KJ, Buchanan W\v, Wohl MJ, et al: Sjogren‘s syrldrome. A clinical. pathological. and serological stud! of 62 cases. Sledicine 44: 187-23 I. 1965 2. ,4nderson L. Talal N: rhe spectrum of benign to malignant l~mphoproliferation in Sjogren‘s svndrome. Clin Exp Immunol CJ.lYY-221. lY71 3. Talal N. Sokoloft L. Barth W: Extrasalivary lymphoid abnormalities rn Sjogren’s syndrome (Reticulum cell sarcoma, Macroglobulinemia). Am J Med 13:50-65, “l’seudolvm~phonla.” 1067 4. van (;umbrrr (1. Seitrrt G: Immunoglobulir~-containing plasma ceils in chronic parotitis and malignant lymphomas of'the parotid gland. \‘irchows Arch [A] 389:79-92. 1980 5. Stt-imlan 0. Rosenow EC:, Divertie MB. et al: Pulmonary manifestaticllls 01 Sjogren’s syndrome. Chest 70:354-36 I, 1976

6. Talal IN, Schnitzer B: Lymphadenopath) dt-ome. Clin Rhrum Dis 3:421-432. 1977 7. Kassan SS. Thomas TL, Moutsopoulos

and Sjogren‘s

syn-

WM. et al: Increased

risk oflvmphoma in sicca svndrome. .\nn Intern Med X9:888-XY2. I Y78 8. Colby TV. Barrington C:B: Pulmonary Ivmphomas: Current concepts. Hcs% PA~‘HOI. 14:88+887. lYX3 Y. Mann RB. Jaffe ES, Berard CW: Malignant lymphomas: a conceptual understanding of morphologit diversity. .\m J Pathol %:I05192. 1970 10. Faguet GB. Webb HH. Agee JF, et al Immunologically diagnosed malignancv in Sjogren’s pseLltl(,l~Inphonla. Am J .&led 65:424-429. lY7X 1 I. Wilke WS, Tubbs RR, Bukowski KM, et al: ‘I‘ cell Ivmphoma occurring in Sjogren’s syndrome. Arthrili:, Rhrum 27:45l95.5, 198‘4 12. Scullv RE, Mark EJ, McNeelv BL’ (edsl: Presentation of Case 29-1981: Case Records of the Massachusetth lienera Hospital YO.5:15:i-l6O. 1981 13. Zulman J, Jaffe R, Talal h’: Evidence that the malignant lymphoma of Sjogren’s svndrome is a monoclonal B-cell neoplasm. New EnEl 1 Med 299:1215-1220. 1978 I-I. Landay A, Kubagawa H. Coopet- MD: Idiotypic analvsis of a B cell clone with anti-intermediate filament speciticit! in a patient with Sjogren’s syndrome: Involvement of tive subpopulations producink different immunoglobulin isotvpes. J Immunol 137: 12% 136, 1986 15. (:asev TT. Cousar JB. (:ollins RI): .2, simplified plastic embedding and immunohistologic technique for immunophenotvpic analysis of human hemapoietic and Ivmphoid tissues. Am J Pathol 131:18:4-189, I988 16. Epstein AL, Marcher R.1. Winter ]A, et al. Two new monoclonal antibodies (LN-1, LX- 1) reactive‘in B5 formalin fixed. paraffin embedded tissues with follitular center and mantle zone B lymphocytes and derived tumors. J Immunol 133: 1028- 1036. 1984 17. Kurtin PJ. Pinkus GS: Leukoclte common antigen--a diagnostic discriminant between hematopoietic and nonhematopoietic neoplasms in paraffin sections using mono, lonal antibodies: C;ot-r-elation with immunologic studies and ultrasrructut-al localization. HC.M PATH~L 16:353-365. lY85 IX. Smith SH. Brown MN. Rowe I). et al: Euncttonal subset5 01 human helper-inducer cells detined by a new monoc lonal antibodv. L’CHLI. Immunology 58:63-70. I YXfi 1Y. Norton AJ, Isaacson PC;: Mono&ma] antibodv L%: An antibody that is reactive with normal and neoplastic B lymphocyte\ in routinely fixed and paraffin wax embedded tissues. ,J (:lin Pathol 40:1405-1412, I987 20. Leech JH, Glick AD, Waldron JA, et al Malignant Iymphomas of follicular center cell origin ‘in man. I. Immunologic~ studies. ,JNCI 54:l l-21. 107.5 ?I. Stein KS, Cousar .J, Flexner JM. et ~1: (.orrelations between immunologic markers and histopathologlc classifications: (;linical implications. Semin Oncol 7:244-251. 19x0 22. Motolsky .4G. Weinberg S. Saphir 0. et al: Lymph node< in rheumatoid arthritis. Arch Intern Med YO:660-676. I952 23. Nosanchuk ,JS, Schnitzer B: Folliculal hyperplasia in Iymph nodes from patients with rheumatoid at-thritis. Cancer 21:343-3%. lY6Y 24. Fox RA. Rosahn PD: The lymph node 111disseminated lupus er!thematosus. Am J Pathol lY:7385. lY4L’ 25. Dorfman RF. Remington JS: Value of lymph-node biopsy in the diagnosis of acute acquired toxoplasmosis. Uew Engl J Med 28Y:878-X8 I. I Y73 26. Ioachim HL. Lerner CW. Tapper hlL. I he lymphoid lesions associated with the acquired immunodeficientr syndrom. Am J Surg Pathol 7:543-553, 1983 27. Schmid LT. Helbron D, Lennert K: Development of malignant lvmphoma in myoepithelial sialadenitis (Siogren’s syndrome). Virchows Arch [A] 395: 1143. 1982 28. Diaz-Jouanen E, Ruiz-Arguelles (;J, Vega-Ortiz JM. et al: From benignn oolvclonal to malignant monoclonal Ivmphoproliter, ation in a pitient with primary Sjogrrtl’a svndrome. Arthritis Rheum 24:850-X53. IYXI 2Y. Kessler HS: A laboratory model for Slogrcn’r svndrome. Am J Pathol .52:671-680. 1968 30. Kessler HS. Cubberly M, Manski W: Lye changes in autoimmune NZB and h’ZB x NZW mice. Arch Ophthalmol 85:21 l219. 1971 Y

491

3

L

1

HUMAN PATHOLOGY

Volume 21, No. 5 (May 1990)

31. Taylor CR: Immuno-histological observations upon the development of reticulum cell sarcoma in the mouse. J Path01 118:201-219, 19’76 32. Howie JB. Heyer BJ. The immunology and pathology of NZB mice. Ret Adv Immunol 9:215-245. 1968 33. Fauci AS, Montsopoulos HM: Polyclonally triggered B cells in the peripheral blood and bone marrow of normal individuals and in patients with systemic lupus erythematosus and primary Sjogren’s syndrome. Arthritis Rheum 24:577-584, 1981 34. Fox RI, Carstens SA, Fong S, et al: Use of monoclonal antibodies to analyze peripheral blood and salivary gland lymphocyte subsets in Sjogren’s syndrome. Arthritis Rheum 25:419-426, 1982 35. Fox RI, Adamson TC, Fong S, et al: Lymphocyte phenotype and function in pseudolymphoma associated with Sjogren’s syndrome. J Clin Invest 72:52-62, 1983 36. Fishleder A, Tubbs R, Hesse B, Uniform detection of immunoglobulin-gene rearrangement in benign lymphoepithelial lesions. New Engl J Med 316:1118-1121. 1987

492

37. Hyjek E, Smith WJ, Isaacson PC;: Primary B-cell lymphoma of salivary glands and its relationship to myoepithelial sialadenitis. HUM PATHOL 19:766-776, 1988 38. Isaacson P, Wright DH: Extranodal malignant lymphoma arising from mucosa-associated lymphoid tissue. Cancer 53:25152524, 1984 39. Cousar JB, M&inn DL, Glick AD, et al: Report of an unusual lymphoma arising from parafollicular B-lymphocytes (PBLs) or so-called “monocytoid” lymphocytes. Am J Clin Path01 87:121-128, 1987 40. Sheibani K, Sohn CC, Burke JS, et al: Monocytoid B-cell lymphoma. A novel B-cell neoplasm. Am J Pathol 124:310-318. 1986 41. Sheibani K. Burke JS, Swartz WG. et al: Monocytoid B-cell lymphoma. Clinicopathologic study of 21 cases of a unique type of low-grade lymphoma. Cancer 62:1531-1538. 1988 42. Ngan B-Y, Takagi D, Dombrowski A, et al: Monocytoid B-cell (parafollicular B-cell) lymphoma: a study of 20 cases. Mod Pathol 8:201. 1989 (abstr)