James J. Butler

We describe the histologic and clinical features of non-Hodgkin's lymphoma diagnosed between January 1980 and December 1983 in 90 homosexual men from San Francisco, Los Angeles, Houston, and New York. The median age was 37 years, with an age distribution identical to that for cases of AIDS reported to the Centers for Disease Control. Sixty-two per cent of the patients had high-grade (aggressive) subtypes of lymphoma, 29 per cent had subtypes of intermediate grade, and 7 per cent had low-grade subtypes. Histologic subtypes and malignant cell phenotypes were consistent with a B-cell origin. All but two men had extranodal lymphoma: central-nervous-system, bone-marrow, bowel, and mucocutaneous sites were most commonly involved. Thirty-five of 66 evaluable men (53 per cent) had complete responses to combination chemotherapy or radiotherapy or both, and thus far, 19 (54 per cent) of them have had a relapse. Mortality and morbidity were closely related to prodromal manifestations; death or illness have occurred in 19 (91 per cent) of the 21 men who presented with AIDS, in 26 (79 per cent) of the 33 who presented with generalized lymphadenopathy, and in 5 (42 per cent) of the 12 who had no prodromal manifestations. Mortality rates analyzed according to histologic grade were higher than currently reported rates in other patient populations. Kaposi's sarcoma or severe opportunistic infections characteristic of AIDS developed in 14 of 33 men (42 per cent) who presented with generalized lymphadenopathy and in 3 of 12 (33 per cent) without prodromal manifestations. We conclude that non-Hodgkin's lymphoma in members of an AIDS risk group is a serious manifestation of AIDS and the AIDS-related complex.

This paper evaluates the significance of the clinical stages and histologic features of Hodgkin's disease in 377 U. S. Army cases from World War II with a 15- to 18-year follow-up. From this study 6 histologic types have emerged: (1) lymphocytic and/or histiocytic (L & H), nodular; (2) lymphocytic and/or histiocytic (L & H), diffuse; (3) nodular sclerosis; (4) mixed; (5) diffuse fibrosis and (6) reticular. There is a definite relationship between histologic types, clinical stages and survival. The L & H types are expressions of lymphocytic proliferation and diffuse fibrosis and reticular types represent lymphocytic depletion while the mixed is intermediate between these extremes. Nodular sclerosis appears to be a regional expression of Hodgkin's disease in the mediastinum and is of major prognostic significance in stage I. The histologic types are regarded as expressions of an attempted host response and possibly evidence of the dramatic interplay between the host and the factors responsible for the development of Reed-Sternberg cells. The authors suggest that the Hodgkin's disease process represents the attempted induction of malignant neoplasia and that the evolution of the histologic process is a manifestation of the natural history of the disease.

Sixteen patients presenting with granulocytic sarcoma without evidence of acute leukemia were seen and diagnosed at The University of Texas M.D. Anderson Hospital and Tumor Institute at Houston from 1962 to 1985. Seven of them (44%) did not develop acute leukemia. Of these seven, four are alive with no evidence of disease 3.5 to 16 years after initial presentation; the remaining three patients died of their disease within 2 to 8 months of presentation. Two of 16 patients were diagnosed within the last 15 months and do not have adequate follow-up. The seven remaining patients developed acute leukemia within 1 week to 13 months of the diagnosis of granulocytic sarcoma. Six of them died 5 weeks to 16 months after diagnosis; one patient has been in complete remission for 8 years. Twelve of these 16 cases (75%) were initially misdiagnosed, most frequently as large cell lymphoma. The remaining four cases were correctly diagnosed as granulocytic sarcoma. The naphthol-ASD-chloroacetate esterase stain was required to make the correct diagnosis in all cases. Contrary to findings in other series, granulocytic sarcoma arising in nonleukemic patients does not necessarily progress to acute leukemia. At least four of 16 (25%) patients in this series did not develop acute leukemia during the 3.5 to 16 years they have been followed. No prognostic factors were identified in this series to predict which patients would develop acute leukemia and which ones would not.

The clinical and pathologic findings for 35 patients with malignant lymphoma presenting in the thyroid are reviewed. The lymphomas tended to occur in females with a median age of 65 years and clinically were manifested by a mass in the neck. The majority of patients were euthyroid and thyroid scans demonstrated cold nodules. In none of the patients was there clinical suspicion of lymphoma prior to surgery. Thirty-four of the cases were histiocytic lymphomas; the one exception; a patient with nodular poorly differentiated lymphocytic lymphoma, had histiocytic lymphoma in a subsequent biopsy of the soft tissues of the neck. Although classified as histiocytic, the lymphomas had the histologic and ultrastructural features of transformed lymphocytes or immunoblasts. Lending possible additional credence to the immunoblastic nature of these lymphomas was the histologic documentation of chronic lymphocytic thyroiditis in all 27 cases where residual thyroid parenchyma remained. This relationship suggests possible evolution of thyroid lymphomas from chronic lymphocytic thyroiditis and probably is analogous to the malignant lymphomas developing in other altered immune states, including Sjogren's syndrome. In the current study the overall 5-year survival was 54%. Patients under age 65, without local soft tissue extension or regional lymph node involvement, and with stage I disease survived the longest; a nodular histologic pattern also appeared to favorably influence the prognosis. Improved staging procedures and newer modes of therapy appear essential, particularly for those patients with clinical stage II disease and with local extension to soft tissues.

Fifty (5%) of 1039 patients with non-Hodgkin's lymphomas registered on two Southwest Oncology Group clinical trials between 1972 and 1977 developed evidence of central nervous system (CNS) lymphoma. Thirty-nine patients (3.7%) had leptomeningeal involvement, 10 patients (1%) had focal cerebral involvement and 2 patients (0.1%) had spinal cord compression. Cytologic examination of the cerebrospinal fluid (CSF) was the most reliable diagnostic technique. Patients with diffuse histologies and those with nodular histiocytic lymphoma had the highest incidence of CNS disease. Diffuse histiocytic lymphoma accounted for 22 patients with CNS involvement. Lymphoblastic lymphoma, convoluted type, was found in 6 patients, although in none was the diagnosis established prospectively. Patients who developed CNS lymphoma usually had extranodal disease and systemic symptoms at initial staging (90% stage IV, 54% “B” symptoms). The most commonly involved extranodal sites included the bone marrow (56%), gastrointestinal tract (26%), skin (18%), and lung (14%). CNS lymphoma was observed at all times during the clinical course but was most common in patients with active, poorly controlled disease. In 13 patients (26%), however, CNS lymphoma was the first sign of relapsing lymphoma. Median survival after recognition of CNS lymphoma was 2 months. The incidence of CNS disease was similar for 5 remission induction regimens employed, but maintenance chemotherapy with vincristine, prednisone and parenteral cytosine arabinoside appeared to substantially reduce the incidence of late CNS relapse (1 of 90 patients) compared to maintenance treatment with oral cyclophosphamide in place of cytosine arabinoside (9 of 90 patients, p = 0.02). This study indicates that there is a subgroup of patients with readily identifiable clinical and pathologic features who might benefit from prophylactic therapy to prevent CNS lymphoma.