Peripheral T-cell lymphoma

Peripheral T-cell Lymphoma JAMES 0.ARMITAGE, MD,' JOHN P. GREER, MD,t ALEXANDRA M. LEVINE, MD,* DENNIS D. WEISENBURGER, MD,§ SlLVlA C. FORMENTI, MD,* MARTIN BAST, BS,* SUE CONLEY, BA," JENE PIERSON, BS," JAMES LINDER, MD,§ JOHN B. COUSAR, MD,II AND BHARAT N. NATHWANI, MDT

Peripheral T-cell lymphoma is the most common type of T-cell lymphoma seen in adults in the United States. Clinical data were reviewed from 134 cases of peripheral T-cell lymphoma diagnosed in three centers. The median age of the patients was 57 years (range, 4-97 years), 59%were male, and 36 patients (27%)had a history of a preceding disorder of the immune system. The tumors were grouped histologically into large cell (43%), mixed large and small cell (40%), and small cell (17%).The stage at diagnosis was I (7%),I1 (21%),I11 (22%),and IV (50%).B symptoms were present in 57%.The most frequent sites of extranodal involvement were bone marrow (35%), skin (13%), and lung (11%). Eighty patients were treated with a multiagent chemotherapy regimen with proven curative potential in aggressive non-Hodgkin's lymphomas and the remainder of the patients received less intensive chemotherapy (36 patients), radiotherapy (nine patients), or no treatment (nine patients). Fifty percent of the intensively treated patients achieved complete remission and the actuarial 4-year survival was 45%. However, the 4-year, diseasefree survival in patients with Stage IV disease was only 10%. Although peripheral T-cell lymphomas appeared similar in many ways to their B-cell counterparts, disease-free survival by stage was low and patients with Stage IV disease had an especially poor outlook. Cancer 63:158-163, 1989.

R

could be subdivided based on surface membrane characteristics led to new classification systems for lymphoid malignanc i e ~ . ' -In ~ the United States, distinct from some other parts of the world, it quickly became apparent that malignant lymphomas bearing T-cell surface markers were much less frequent than B-cell lymphomas. With the advent of monoclonal antibodies against T-cell determinants that appear early or late in normal T-cell development," it was possible to identify some tumors as developing from immature (Le., thymic or central) T-cells and others from more mature (i.e., peripheral) T-cells. Lymphomas presenting as tumorous masses which bear an immature Tcell phenotype generally have been classified as lymphoblastic lymphomas and those bearing a mature T-cell phenotype (ie.,excluding mycosis fungoides or cutaneous T-cell lymphoma) have been grouped together as peripheral T-cell lymphoma (PTCL).' The latter comprise the largest group of T-cell lymphomas in the United state^.^'^ Peripheral T-cell lymphomas exhibit a variety of his-

tologic appearances,generally being classified in the diffuse mixed or immunoblastic catagories of the Working Formulation.' In Japan and the Caribbean PTCL are frequently associated with infection by human T-lymphotropic virus4 (HTLV-I).9,'oPrevious series of patients with PTCL reported from the United States have found a high frequency of cutaneous involvement, advanced age, widely disseminated disease, and systemic symp'-I4 , ' Many investigators have suggested that patom~.~ tients with PTCL have a particularly poor prognos ~ s , ' * ~ whereas ' ~ - ' ~ others have found no difference in response rates when compared to B-cell lyrnphorna~.'~~'~ We have combined the clinical data from all patients with PTCL seen at our institutions to better characterize this malignancy.

ECOGNITION THAT LYMPHOCYTES

Patients and Methods

The 134 patients included in this report represent all the cases of malignant lymphoma with a proven T-cell immunophenotype and meeting our criteria for PTCL seen at our three institutions between 1973 and July 1986. Peripheral T-cell lymphoma was diagnosed when a patient with non-Hodgkin's lymphoma had a T-cell immunophenotype proven with the exclusion of cases diagnosed as lymphoblastic lymphoma or mycosis fungoides/cutaneous T-cell lymphoma. A T-cell immunophenotype was proven by rosette assay for sheep erythrocyte receptors, or by the use of immunofluorescence or immunoperoxidase techniques as previously described.'8-20

From the Departments of *Medicine and §Pathology and Microbiology, University of Nebraska Medical Center, Omaha, Nebraska; Departments of ?Medicine and 11 Pathology, Vanderbilt University School of Medicine, Nashville, Tennessee; and Departments of $Medicine and TPathology, University of Southern California School of Medicine, Los Angeles, California. Address for reprints: James 0.Armitage, MD, Department of Internal Medicine, University of Nebraska Medical Center, 42nd and Dewey Avenue, Omaha, NE 68 105. Accepted for publication July 1 I , 1988.

158

No. 1

PERIPHERAL T-CELLLYMPHOMA

Patients determined to have PTCL using the E-rosette assay were included only when 30%or greater of the cells formed E-rosettes at 4”C, the rosette-forming cells appeared neoplastic in stained cytocentrifuge preparations, and a low percentage of cells with polyclonal surface immuinoglobulin were present. Cases identified by direct im munofluorescence or immunoperoxidase techniques were accepted only when the vast majority of the neoplastic appearing cells from nodal or extranodal tumors stained with one or more T-cell-specific antibodies (i.e., OKT 4, 8, and 11 from Coulter [Hialeah, FL], and Leu 1 tlo 4 from Beckton-Dickinson [Sunnyvale, CAI) in the absence of a monotypic population of B-cells. Thirty-five percent of the cases were identified using frozen section im munoperoxidase studies, 29% with E-rosetting, 17% wii h both direct immunofluorescence and E-rosetting, 16‘%with direct immunofluorescence, 2% with all three studies, and 1% with frozen section immunoperoxidase and E-rosetting. ,4 detailed histologic analysis of the cases included in this study is in progress and will be presented separately. For the purpose of this presentation, the cases were grouped into those with predominantly large cells (i.e., diffuse large cell and immunoblastic lymphomas in the Working Formulation), mixed cell tumors (i.e., diffuse mixed cell lymphoma in the Working Formulation), and those with predominantly small cells (i.e.,difficult to classifiq in the Working Formulation). The latter generally do not fit the usual criteria for small lymphocytic or diffuse small cleaved cell lymphoma (i.e., both typically B-cell neoplasms) from the Working Formulation. The clinical data in each case was collected by the clinical investigator at each center. Several of these patients have been included in previous reports.’8-21The Ann Arbor staging system was used.22The staging studies employed at the three centers were not uniform. However, all patients had histories taken and physical examinations, co mplete blood counts, screening chemistry studies, chest x-rays, and investigations for intraabdominal disease utiliz ing imaging studies (i. e., computerized axial tomography [CAT] scans and/or lymphangiograms) and/or laparotomy. Bone marrow biopsies were performed on 118 patients. Serum lactate dehydrogenase (LDH) levels were recorded as normal or elevated depending on the normal values for each individual laboratory. The treatment regimens utilized varied considerably among the three centers. Patients were grouped into those receiving intensive combination chemotherapy regimens with curative potential in diffuse large cell lymphoma, less intensive chemotherapy regimens, radiotherapy only, or no treatment. The intensive chemotherapy regimens used in 80 patients included cyclophosphamide, doxoruibicin, vincristine, and prednisone (CHOP)23with or without bleomycinZ4in 30 patients; cyclophosphamide, doxorubicin, procarbazine, bleomycin, vincristine, pred-

-

Armituge et ul. TABLEI .

159

Patient Characteristics No. of patients

Characteristic Total patients

134

Median age (range) Ma1e:female

57 yr (4-97) 7955

Preceding disorder of the immune system Other lymphoproliferative disorders Angioimmunoblastic lymphadenopathy Atypical dermatitis Mononucleosis Lymphomatoid papulosis Lymphomatoid granulomatosis Different lymphoma “Autoimmune” arthritis Miscellaneous*

36 (27%) 15 (1 1%) 6 4 2 2 1 1 1 (8%) 6 (4%) 4 (3%)

Preceding nonhematologic malignancy

4 (3%)

Stage I I1 111 IV

10 (7%) 28 (21%) 29 (22%) 67 (50%)

Symptom status A B

58 (43%) 76 (57%)

Histologic group Large cell Mixed Small cell

58 (43%) 53 (40%) 23 (17%)

Hypercalcemia (Caf+ measured in 122 patients)

3 (2%)

HTLV-1 positive (measured in 24 patients)

1 (4%)

Elevated LDH (measured in 1 13 patients)

98 (87%)

HTLV-I: human T-lymphotropic virus-I; LDH lactate dehydrogenase. * Grave’s disease, systemic vasculitis, sarcoidosis, ataxia telangiectasia.

nisone (CAP-BOP)” in 22 patients; cyclophosphamide, vincristine, methotrexate, and cytosine arabinoside (COMLA)26-27 in seven patients; BACOP28in five patients; TABLE2. Sites of Proven Extranodal Involvement by Peripheral T-cell Lymphoma in 134 Patients Organ involved

No. of patients

Bone marrow Skin Lung Liver Gastrointestinal tract Bone Nasopharynx Adrenal Salivary gland Brain Other *

41 18 15 10 8 5 3 2 2 2 7

~

* Other extranodal sites involved in one patient: testicle, muscle, kidney, meninges, orbit, pancreas, tongue.

CANCERJanuary 1 1989

160

TABLE3. Relation of Site of Involvement and Histologic Group Site

TABLE4. Treatment Outcome in Patients Receiving Intensive Combination Chemotherapy Regimens

No. of patients (%)

Histologic group

Group Bone marrow*

Skin

Lung

Large (5 1 patients) Mixed (48 patients) Small (19 patients)

12 (24%) 16 (33%) 13 (68%)

Large (58 patients) Mixed (53 patients) Small (23 patients)

5 (9%) 8 (15%) 5 (22%)

Large (58 patients) Mixed (53 patients) Small (23 oatients)

4 (7%) 1 1 (21%) 0 10%)

* Only patients who had bone marrow biopsies are included. methotrexate with leucovorin rescue, doxorubicin, cyclophosphamide, vincristine, prednisone, and bleomycin (MACOP-B)29in three patients; NHL-330in three patients; M-BACOD3'in two patients; COPP/MOPP32in two patients; and other, miscellaneous regimens in six patients. A complete remission was defined as the disappearance of all disease present at the initiation of therapy documented by repeating previously abnormal studies. Because less than complete response has little clinical usefulness, no distinction was made between partial responders and nonresponders. Survival was calculated from the time of the initiation of therapy. Deaths from causes other than lymphoma were not censored. Complete remission duration was calculated from the time of documentation of a complete response. Either relapse or death from any cause were considered to terminate remission in the calculations. Survival curves were calculated using the life table method and comparisons between curves were performed using the log-rank test. Comparisons between groups were done using the chi-square test with the Yates correction.

t-

z W

0

a W

n

2ol 10

00

10

20

30

40

0 50

MONTHS FIG.1. Overall survival for all 134 patients with PTCL. The numbers in parenthesis indicate patients surviving to the indicated interval.

Vol. 63

All patients Stage I & I1 111 IV

Histologic type Large Mixed Small

No. of patients

Complete responders (%)

80

40 (50%)

24 20 36

18 (75%) 12 (60%) 10 (28%)

42 24 14

23 (55%) 13 (54%) 4 (29%)

Results The clinical characteristics of the 134 patients with PTCL are presented in Table 1. The patients ranged in age from 4 to 97 years (i.e., median 57 years) and 59% were male. Preceding disorders of the immune system had been diagnosed in 27% of the patients. These included various autoimmune processes in 7%, a different (i.e., usually B-cell) non-Hodgkin's lymphoma in 8%, and a variety of other lymphoproliferativedisorders in 1 1%. The latter group included such conditions as angioimmunoblastic lymphadenopathy and lymphomatoid granulomatosis which might well have represented early manifestations of PTCL.33In one patient with lymphomatoid papulosis, the condition has persisted after apparent cure of a large cell PTCL. The majority of patients had widely disseminated disease and 50% were Stage IV. Fevers, night sweats, or significant weight loss were present in 57% of the patients at diagnosis. Histologic classifications were large cell in 43%, mixed large and small cell in 40%, and small cell in 17%.Hypercalcemia was present at diagnosis in only 2% of patients in whom a serum calcium determination was performed and only one of 24 patients tested (i.e., a patient with diffuse mixed lymphoma)had evidence for infection by HTLV-I. In contrast, 87% of the patients in whom the serum LDH determination was performed the level was elevated. Table 2 presents the sites of extralymphatic involvement by PTCL. The extranodal site most frequently proven to be involved was the bone marrow, which was found in 41 of 118 (35%) patients. The next most frequently involved sites were the skin in 13%, the lung in 11%, the liver in 7%,and the gastrointestinal tract in 6%. Of course, since all patients did not have biopsies performed on all organs, the frequency of involvement of sites such as liver might be under estimated. As illustrated in Table 3, the frequency of bone marrow involvement by PTCL was related to histologic type, which was significantly more frequent in patients with small cell tumors ( P = 0.002). The overall survival for all 134 patients (i.e., median


No. 1

161

Armituge et al.

30-

2ol

10

10

0

10

20

30

40

50

- 1

0

MONTHS FIG. 2. The curve indicates the durability of complete remission in patients who received intensive combination chemotherapy regimens for PTCL. Numbers in parenthesis indicate the patients remaining in remission to the indicated interval.

follow-up for surviving patients of 20 months) is presented in Figure 1. The median survival was 17 months and the proportion surviving at 4 years was 28%. The treatment regimens received by these patients varied by institution and over time. Sixty percent of the patients received a mudtiagent chemotherapy regimen with curative potential in diffuse aggressive non-Hodgkin’s lymphoma, whereas 27% of the patients received less intensive chemotherapy, 7% of the patients were treated with radiotherapy alone, and 7% of the patients received no treatment. The treatment outcome for patients who received intensive combination chemotherapy regimens is presented in Table 4. Fifty percent of these patients achieved complete remissio n. Attainment of complete remission was significantly less likely in patients with Stage IV disease (P = 0.001). Tbe duration of complete remission in these patients is presented in Figure 2. Four-year durability of remission was 41%. The overall survival of the intensively treated patients (i.e.,45% at 4 years) is presented in Figure 3 and contrasted to patients receiving other therapies. Diseasefree survival by stage is presented in Figure 4. Patients with Stage I and I1 were significantly more likely to have durable remissions than those with Stage IV disease (P = 0.01), although the Stage I and I1 patients had a poorer disease-free survival than has been reported recently for loc,alizedB-cell lymphomas. Overall survival by histologic group is presented in Figure 5. Despite a lower complete remission rate, patients with small cell histologic type did no1 have a shorter survival. Discussion

The term “peripheral T-cell lymphoma” encompasses lymphomas with a variety of histologic appearances and clirdcal patterns. In this regard, PTCL might be compared to the B-cell lymphomas which are included under the

~

lb

2b

310

410

~~

50

MONTHS

3. The curves represent the survival of patients receiving intensive combination chemotherapy regimens and those patients receiving less aggressive therapy or no treatment. The difference in survival between the two groups was significantly different (P = 0.01). FIG.

term “follicular center cell lymphoma.” Almost certainly, the wide variety of histologic appearances which are encompassed by PTCL represent several clinical entities. Thus, our findings in this large number of patients suggest that all patients with PTCL are not the same, and should not be considered as such. This would be fully analogous with the B-cell follicular center cell lymphomas, which, although all of B-cell origin, represent several distinct clinicopathologic entities. As is the case with B-cell lymphomas, histologic subtyping of PTCL appears to be of prognostic importance. Approximately 7% of the patients in our series had prior histories of various autoimmune disorders, whereas 19% had prior lymphoproliferative disorders. The finding of prior immune disease was documented by Levine et al.

O01

90i I-

z W

0

a W

n

-6

1‘0

I

20

I

30

410

50

MONTHS FIG.4.The three curves represent the disease-free survival of patients with F’TCL who received intensive combination chemotherapy regimens. Patients with Stage I I1 had a superior survival to patients with Stage IV (P = 0.01).

+

CANCERJ a mrary 1 1989

162

70 -

I2

60-

0

50-

W

a

40-

30 2010-

- 0

10

do

30

I

40

50

MONTHS FIG.5. The curves represent the overall suMval of patients with PTCL who received intensive combination chemotherapy regimens grouped by histologic subtype. There was no significant difference among the groups.

in 16%of 19 cases of T-cell immunoblastic sarcoma, including patients with Hashimoto’s thyroiditis and immune thrombocytopenic purpura.” Greer et al. cited a 24% incidence in their series of 42 cases, but included patients with angioimmunoblastic adenopathy and lymphomatoid granulomatosis, which may represent early manifestations of PTCL6333 and were classified as lymphoproliferative disease in the current series.18 The demonstration of prior autoimmune disease in these patients with T-cell immunoblastic sarcoma (T-IBS) was not distinct from patients with B-cell immunoblastic sarcoma, occurring in 3 1% of Levine’s series. It is thus apparent that some increase in prior immune disease may be seen in either B-cell or T-cell lymphoma. Although it is difficult to estimate the frequency of autoimmune disorders in the general population, it is probably no greater than 2% to 3%. Aside from prior history of autoimmune disorders, 19% of our cases had histories of prior lymphoproliferative disorders. Again, such a finding has been reported previously, with 2 1% of Levine’s cases of T-IBS presenting with histories of such prior malignancy, including T-cell chronic lymphocytic leukemia (T-CLL) and lymphoepithelioid cell lymphoma19and with 24% of Greer’s cases, including 12%B-cell lymphoma and 12%either angioimmunoblastic adenopathy or lymphomatoid granulomatosis.’8 In the current series, the majority of prior lymphoproliferative malignancies were of B-lymphoid origin. Other lymphoproliferative disorders included conditions such as angioimmunoblastic lymphadenopathy and lymphomatoid granulomatosis, which might well have represented early manifestations of PTCL, as has been documented by ~ t h e r s .Winberg ~ , ~ ~ et al., for example, have described such a progression of histologic type in patients with PTCL.33

Vol. 63

The frequency and sites of extranodal involvement by PTCL has varied in previous series. The biggest variation has been in the frequency of skin involvement. Cutaneous involvement has been documented in six of nine patients reported by Grogen et a1.,’* and in 49% of a series of 41 patients by Homing et a l l 3We found cutaneous involvement in 13%of the 134 patients in this series, a frequency that did not vary significantlyamong the three institutions. The higher frequency of cutaneous involvement in previous reports probably does reflect patient selection, as ’ ~ patients with cutasuggested by Horning et ~ 1 . Thus, neous involvement can undergo repeat biopsy quite easily, and with additional biopsy material for performance of immunophenotype, these cases may be more readily recognized as T-cell lymphomas. The frequency of bone marrow involvement which we observed was similar to that found in most previous reports, which have ranged Interestingly, the frequency of bone from 1 1% to 44%.13,16 marrow involvement was significantlyrelated to histologic subtype in our series. As in B-cell lymphomas, patients with small cell histologic type had a significantly higher frequency of marrow infiltration than patients with mixed or large cell histologic type. In southwest Japan and the Caribbean, various T-cell lymphoproliferativedisorders, including PTCL, have been associated with infection by the retrovirus designated HTLV-I.9*’oThese patients often have widely disseminated disease, hypercalcemia, bone lesions, and leukemic transformati~n.~,’~ The frequency of the association between HTLV-I and PTCL in the United States has been uncertain but some have suggested such a relationship in 10% of cases.34Only one of our 24 cases tested (a case from the southeastern United States) demonstrated presence of antibodies to HTLV-I. Thus, at least from the regions on the United States represented by the institutions contributing to this study, it does not appear that HTLVI plays a major etiologic role in PTCL. Several previous reports have suggested that PTCL is associated with an especially dire prognosis. 12314-16However, other reports have found no significant difference in outcome between patients with PTCL and those with Bcell lymphoma, or between patients with large cell lymphoma with either T-cell or B-cell immunophenoIn contrast to Weis et a1.,6we did not find a significant survival difference between patients with large and small cell PTCL when only those patients receiving aggressive chemotherapy were considered. The results from our study suggest that PTCL does have a poor outlook, especially with high-stage disease. With the treatment regimens utilized in these patients, only 10%of those patients with Stage IV disease had prolonged disease-free survival. These results are poorer than generally reported for patients with aggressive non-Hodglun’s lymphomas Also,more than irrespective of immunologic s~btYpe.~~-~’

No. 1


50% of the complete responders have relapsed. Again,

this is higher than generally found in previous reports of series regardless of i m m ~ n o p h e n o t y p e . ~ ~ - ~ ~ In conclusion, patients with PTCL are similar in many ways to patients who present with various B-cell lymphomas. Prior histories of autoimmune disease may be seen, and many of these individuals have histories of prior lymphalproliferative disorders. The majority of patients have systemic “B” symptoms, with advanced stage of disease at diagnosis. With use of intensive, multiagent chemotherapy, complete remission and long-term survival may be achieved, although the disease-free survival we observed was lower than might have been expected and patients with Stage IV disease responded very poorly. The results of prospective therapeutic trials are awaited to clarify this initial observation. REFERENCES 1. Lukes RJ, Collins RD. Immunologic characterization of human malignant lymphomas. Cancer 1974; 34: 1488- 1503. 2. Lukes RJ, Collins RD. New approaches to the classification of the lymphomata. Br J Cancer 1975: (Suppl)2:1-28. 3. Lennert K. Histopathology of Non-Hodgkin’s Lymphomas. New Yorlc: Springer-Verlag, 198I ; 4 1-44;92- 1 I 1. 4. Foon KA, Todd RF. Immunologic classification of leukemia and lymphoma. Blood 1986; 68: 1-3 1. 5. Waldron JA, Leech JH, Click AD, Flexner JM, Collins RD. Malignant lymphoma of peripheral T-lymphocyte origin. Cancer 1977: 40: 1604-1 6 17. 6. Weis JW, Winter MW, Phyliky RL, Banks PM. Peripheral T-cell lymphomas: Histologic, immunohistologic, and clinical characterization. Mayo Clin Proc 1986: 61:411-426. 7. Rudders RA, DeLellis RA, Ah1 ET Jr, Bernstein S, B e g CB. Adult non-Hodglun’s lymphoma: Correlation of cell surface marker phenotype with prognosis, the New Working Formulation, and the Rappaport and Lukes-Collins histomorphologic schemes. Cancer 1983; 52:2289-2299. 8. The Non-Hodgkin’s Lymphoma Pathologic Classification Project. National Cancer Institute sponsored study of classificationof non-Hodgkin’s lymphomas: Summary and description of a working formulation for clinical usage. Cancer 1982; 49:2112-2135. 9. Uchiyama T, Yodoi J, Sagawa K, Takatsuki K, Uchino H. Adult T-cell leukemia: Clinical and hematologic features of 16 cases. Blood 1977; 50:481-492. 10. Broder S, Bunn PA, Jaffe ES et a/. T-cell lymphoproliferative syndrome associated with human T-cell leukemia/lymphoma virus. Ann Intern Med 1984: 100543-557. 1 1. Jaffe ES. Pathologic and clinical spectrum of post-thymic T-cell malignancies. Cancer Invest 1984a; 2:413-426. 12. Grogan TM, Fielder K, Rangel C et al. Peripheral T-cell lymphoma: Aggressive disease with heterogeneous immunotypes. Am J Clin Pathol 1985; 83:279-288. 13’. Homing SJ, Weiss LM, Crabtree GS ef al. Clinical and phenotypic diveisity of T-cell lymphomas. Blood 1986: 67: 1578-1 582. 14.. Brisbane JU, Berman LD, Neiman RS. Peripheral T-cell lymphoma: A clinicopathologic study of nine cases. Am J Clin Patho/ 1983: 79~285-293. I S . Helbron D, Brittinger G, Lennert K. T-zonen-lymphom: Klinisches bild, therapie und prognose. Hamatol Bluttransfis 1979: 39: 1 17131. 161. Brouet J-C, Rabian C, Gisselbrech C, Flandrin G. Clinical and immunological study of non-Hodgkin’s T-cell lymphomas (cutaneous

-

Armitage et ul.

163

and lymphoblastic lymphomas excluded). Br JHaernafoll984 57:3 15327. 17. Cossman J, Jaffe ES, Fisher RI. Immunologic phenotypes of diffuse, aggressive non-Hodgkin’s lymphomas: Correlation with clinical features. Cancer 1984: 54:1310-1317. 18. Greer JP, York JC, Cousar JB et al. Peripheral T-cell lymphoma: A clinicopathologic study of 42 cases. J C h Oncol 1984; 2:788-798. 19. Levine AM, Taylor CR, Schneider DR et al. Immunoblastic sarcoma of T-cell versus B-cell origin: I. Clinical features. Blood 198 I: 58: 52-6 1. 20. Weisenburger DD, Astorino RN, Glassy FJ et a/. Peripheral Tcell lymphoma: A clinicopathologic study of a morphologically diverse entity. Cancer 1985; 56:2061-2068. 2 I . Weisenburger DD, Linder J, Armitage JO. Peripheral T-cell lymphoma: A clinicopathologic study of 42 cases. Hematologic Oncoll987: 5: 175-187. 22. Carbone PP, Kaplan HS, Musshoff K ef al. Report of the committee on Hodgkin’s disease staging classification. Cancer Res 1971: 31: 1860-186 1. 23. McKelvey E, Gottlieb J, Wilson H et al. Hydroxyldaunomycin (Adriamycin) combination chemotherapy in malignant lymphoma. Cancer 1976: 38:1484-1493. 24. Rodriguez V, Cabanillas F, Burgess MA et a/. Combination chemotherapy (“CHOP-Bleo”) in advanced (non-Hodgkin’s) malignant lymphoma. Blood 1977; 49:325-333. 25. Armitage JO, Weisenburger DD, Hutchins M et al. Chemotherapy for diffuse large-cell lymphoma: Rapidly responding patients have more durable remissions. J Clin Oncol 1986: 4~160-164. 26. Sweet DL, Golomb HM, Ultmann JE et al. Cyclophosphamide, vincristine, methotrexate with leucovorin rescue, and cytarabine (COMLA) combination sequential chemotherapy for advanced diffuse histiocytic lymphoma. Ann Intern Med 1980; 92:785-790. 27. Halnsworth JD, Wolff SN, Stein RS, Greer JP, Cousar JB, Greco FA. Effects of Mega-COMLA (cyclophosphamide,cytarabine, vincristine, and methotrexate followed by leucovorin and prednisone) plus CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) in the treatment of lymphoid neoplasms with very poor prognosis. Cancer Treat Rep 1986; 70:953-958. 28. Schein PS, DeVita VT Jr, Hubbard SP et al. Bleomycin, Adriamycin, cyclophosphamide, vincristine and prednisone (BACOP) combination chemotherapy in the treatment of advanced diffuse histiocytic lymphomas. Ann Intern Med 1976: 85:4 17-422. 29. Klimo P, Connors JM. MACOP-B chemotherapy for the treatment of diffuse large-cell lymphoma. Ann Intern Med 1985; 102:596-602. 30. Koziner B, Sklaroff R, Little C ef al. NHL-3 protocol: Six-drug combination chemotherapy for non-Hodglan’s lymphoma. Cancer 198% 53:2592-2600. 3 I . Skarin AT, Canellos GP, Rosenthal DS et al. Improved prognosis of diffuse histiocytic and undifferentiated lymphoma by use of high dose methotrexate alternating with standard agents (M-BACOD).J Clin Oncol 1983: 1:91-98. 32. DeVita VT, Canellos G, Chabner B et a/. Advanced diffuse histiocytic lymphoma, a potentially curable disease: Results with combination chemotherapy. Lancet 1975: 1:248-250. 33. Winberg CD, Krance R, Sheibani K, Rappaport H. Peripheral Tcell lymphoma: A case report demonstrating morphologic heterogeneity and progression, atypical immune reactions. Cancer 1986: 57:2329-2342. 34. Blayney DW, Jaffe ES, Blattner WA et al. The human T-cell leukemia/lymphoma virus associated with American adult T-cell leukemia/lymphoma. Blood 1983: 62:401-405. 35. Fisher RI, Hubbard SM, DeVita VT et al.Factors predicting longterm survival in diffuse mixed, histiocytic, or undifferentiated lymphoma. Blood 1981: 58:45-51. 36. Jagannath S, Velasquez WS, Tucker SL et a/. Stage IV diffuse large-cell lymphoma: A long-term analysis. J Clin Oncol 1985: 3:39-47. 37. Laurence J, Coleman M, Allen SL et al. Combination chemotherapy of advanced diffuse histiocytic lymphoma with the sixdrug COPBLAM regimen. Ann Intern Med 1982: 97:190-195.