Leukemia (1998) 12, 1457–1460  1998 Stockton Press All rights reserved 0887-6924/98 $12.00 http://www.stockton-press.co.uk/leu Biweekly THP-COPBLM (pirarubicin, cyclophosphamide, vincristine, prednisone, bleomycin and procarbazine) regimen combined with granulocyte colony-stimulating factor (G-CSF) for intermediate- and high-grade non-Hodgkin’s lymphoma N Niitsu and M Umeda

First Department of Internal Medicine, Toho University School of Medicine, 6-77-7 Omori-Nishi, Ota-ku, Tokyo 143-8541, Japan

Biweekly THP-COPBLM including pirarubicin (THP), which is relapse and refractory NHL is treated by high-dose chemo- thought to be less toxic than doxorubicin, was used to treat therapy with peripheral blood stem cell transplantation non-Hodgkin’s lymphoma (NHL) and the remission rate and adverse events were studied in 42 patients younger than 69 (PBSCT), decreased cardiac function also not infrequently years. Complete remission (CR) was achieved in 37 patients poses a problem. (88.1%) and partial remission in five (11.9%). Classified by inter- In the present study, we used a biweekly THP-COPBLM national prognostic index, CR was achieved in 16 out of 17 low– regimen including a new anthracycline anticancer agent, pira- intermediate-risk patients, 14 out of 16 high–intermediate-risk rubicin (THP), which is considered to be less cardiotoxic than patients and seven out of nine high-risk patients. The 3-year DXR, to increase the dose intensity, and we assessed the effi- survival rate was 72.1% and the 3-year event-free survival rate was 58%. Grade 3 or higher adverse events included granulocy- cacy and occurrence of adverse events. topenia in 39 patients (92.9%) and thrombocytopenia in seven (16.7%). The biweekly THP-COPBLM regimen appears useful for the treatment of aggressive intermediate- and high-grade NHL, and G-CSF made it possible to shorten the interval Patients and methods between courses of chemotherapy. Further studies regarding adverse events on organs, other than on bone marrow are required to improve the long-term results of combination ther- Patients apy on NHL. Keywords: non-Hodgkin’s lymphoma; pirarubicin; granulocyte col- Forty-two Japanese patients with previously untreated NHL ony-stimulating factor; international prognostic index who received combination therapy between April 1994 and August 1997 are included in this report. The median age was 56 years (range 18–69). There were 24 males and 18 females. Introduction According to the Ann Arbor classification,8 staging was determined by bone marrow aspiration, bone marrow biopsy, In recent years, non-Hodgkin’s lymphoma (NHL) has been computerized tomography (CT) of the chest, abdominal son- treated successfully in an increasing number of patients. ography, abdominal CT, gastrointestinal endoscopy and gal- Third-generation chemotherapy has achieved complete lium scintigraphy as well as routine examinations. Their per- remission in 80–85% of the patients treated, and long-term formance status (PS) was between 0 and 3 according to the survival in 60–70%.1,2 According to Fisher et al,3 however, Eastern Cooperative Oncology Group (ECOG).9 Of the 42 intensified chemotherapy with third-generation multiple patients, 12 had stage II, 17 had stage III and 13 had stage IV. anticancer agents is not significantly different in efficacy from Histopathology of the NHLs was classified according to the the CHOP regimen, but it causes more adverse effects. When Working formulation.10 They were of the diffuse large cell patients treated mainly with the CHOP regimen were classi- type in 22 patients, the diffuse mixed cell type in 15, the dif- fied into prognostic groups according to the international fuse small non-cleaved cell type in three, and the follicular prognostic index,4 the 5-year survival rate was 73% for the large cell type in two. All patients tested negative for human low-risk group, while the rates for the low–intermediate (L-I), T cell lymphotropic virus type-I (HTLV-I) antibody. The score high–intermediate (H-I), and high (H)-risk groups were 51, 43 system of the International Non-Hodgkin’s Lymphoma Prog- and 26%, respectively. This indicates that the CHOP regimen nostic Factors (INHLPF) Project4 was used to evaluate thera- is satisfactorily effective for low-risk group non-Hodgkin’s peutic effects by age-adjusted risk group. The three prognostic lymphoma, but is inadequate when the prognosis is poorer. In factors employed were: Ann Arbor stages III and IV, a PS score order to improve the effectiveness of chemotherapy for these of 2 or higher as determined according to the ECOG criteria, poorer prognostic groups, a biweekly dose-intensified CHOP and an increased lactate dehydrogenase (LDH) level. They regimen has been used, in which CHOP is administered with were of the L-I risk group in 17, the H-I group risk in 16, and G-CSF to increase the doses of the key agents the H risk group in nine. (cyclophosphamide (CPA) and doxorubicin (DXR)).5,6 The total dose of DXR is limited by cardiotoxicity. When given alone, the dose limit is 450–550 mg/m2. Our previous study Treatment showed that cardiotoxicity appeared when the total dose of DXR in the COP-BLAM regimen was 350–400 mg/m.2,7 If it is All patients were treated with THP-COPBLM chemotherapy remembered that 40–50% of patients with NHL can usually plus recombinant human granulocyte colony-stimulating fac- achieve long-term survival, cardiotoxicity associated with pro- tor (G-CSF). THP-COPBLM regimen consisted of CPA longed administration of DXR should be considered. When 750 mg/m2, THP 50 mg/m2 and vincristine 1.4 mg/m2 (maximum dose, 2 mg) on day 1, and prednisone 40 mg/m2, procarbazine 100 mg/m2 on days 1–7 and bleomycin 10 mg Correspondence: N Niitsu; Fax: +81 03 3763 8298 on day 10 for six cycles every 2 weeks if absolute neutrophil Received 23 March 1998; accepted 27 May 1998 count (ANC) was у2 × 109/l and platelets у10 × 109/l. Dose Biweekly THP-COPBLM combined with G-CSF for NHL N Niitsu and M Umeda 1458 modification was planned for CPA and THP as follows: Table 1 Therapeutic effects decrease of 25% if ANC count was Ͻ2 × 109/l or platelet count Ͻ100 × 109/l on the planned day of treatment; decrease CR (%) PR (%) P value of 50% if the leukocyte count was Ͻ1.5 × 109/l or platelet count Ͻ70 × 109/l; and chemotherapy delay until hematologic Overall 37/42 (88.1) 5/42 (11.9) Ͻ × 9 Stage recovery if the leukocyte count was 1 10 /l or platelet  Ͻ × 9 II 12/12 (100) — NS count 50 10 /l. Doses were not adjusted for patient’s age. III, IV 25/30 (83.3) 5/30 (16.7)  G-CSF (lenograstim) was administered as a daily subcutaneous Histology dose of 2 ␮g/kg from day 5 up to day 13 without day 10 or Diffuse large 21/22 (95.5) 1/22 (4.5)   ANC Ͼ5 × 109/l. Diffuse mixed 12/15 (80) 3/15 (20)  Diffuse small non-cleaved 2/3 (66.7) 1/3 (33.3) NS Follicular large 2/2 (100) —  Phenotype Analysis and statistics T cell 14/16 (87.5) 2/16 (12.5)  B cell 23/26 (88.5) 3/26 (11.5) NS Performance status Response to treatment was documented on completion of 0, 1 28/31 (90.3) 3/31 (9.7)  chemotherapy. Complete remission (CR) was defined as 2, 3 9/11 (81.8) 2/11 (18.2) NS absence of detectable disease on clinical, radiologic and his- LDH; Ͼ1 × normal 33/37 (89.2) 4/37 (12.1) NS у Ͻ tologic criteria. Partial remission (PR) required a Ͼ50% Soluble IL-2R; 1000 U/ml 9/14 (64.3) 5/14 (35.7) 0.01 Bone marrow involvement 2/5 (40) 3/5 (60) Ͻ0.01 reduction of tumor volume. Duration of CR was calculated Bulky mass (diameter у10 cm) 2/4 (50) 2/4 (50) Ͻ0.01 from completion of chemotherapy to relapse or last follow- AAII up. Overall survival (OS) was the interval from initiation of L-I 16/17 (94.1) 1/17 (5.9)  therapy to the time of death or last follow-up. Survival analysis H-I 14/16 (87.1) 2/16 (12.5) NS was performed according to Kaplan–Meier.11 The generalized H 7/9 (77.8) 2/9 (22.2) Wilcoxon test (rank sum test) was used to compare mean dif- ference between the two groups. A level of P Ͻ 0.05 was accepted as statistically significant. WHO criteria12 were used to define adverse events. Dose intensity (DI) was evaluated according to Hryniuk and Bush.13 The actual DI of CPA, THP were calculated by dividing the actual total amount of each drug by the time needed for delivery. The expressions of the actual DI as fractions of the stated protocol dosages were defined as relative DI. The calculation was made for the whole treatment course and separately for the initial cycles needed to achieve maximal response (either CR or maximal degree of partial remission) or to determine the progression of disease.

Results

Therapeutic effects Figure 1 Survival time and event-free survival of non-Hodgkin’s All patients were evaluated for response and survival. Of the lymphoma patients treated with biweekly THP-COPBLAM therapy. 42 patients treated, 37 (88.1%) achieved a CR and five (11.9%) achieved a PR, for a response rate of 100%. The response rate was 100% for stage II disease and 83.3% for stage III and IV disease. With respect to histopathology, the response rate was 95.5% for diffuse large cell type, 80% for diffuse mixed cell type, 66.7% for diffuse small non-cleaved cell type and 100% for follicular large cell type, with no sig- nificant difference observed among histologic subsets. The CR rate for patients with bone marrow involvement was 40%, that for patients with bulky masses of 10 cm or greater in diameter was 50%, and soluble interleukin-2 receptor level Ͼ1000 U/ml was 64.3%. These rates were somewhat lower than the overall rate. According to the INHLPF (age-adjusted (AAII)) index, L-I risk was 94.1%, H-I risk 87.1% and H risk 77.8% (Table 1). The 3-year overall survival and event-free survival were 72.1% and 58%, respectively (Figure 1). According to the INHLPF index, the 42 patients were divided into the three risk groups. There was no significant difference in the survival and event-free survival among the three groups Figure 2 Survival time and event-free survival by the International (log-rank, Wilcoxon test) (Figure 2). Non-Hodgkin’s Lymphoma Prognostic Factors. Biweekly THP-COPBLM combined with G-CSF for NHL N Niitsu and M Umeda 1459 Dose intensity mens,3 3-year survival rates were almost the same among the four groups of patients receiving CHOP (54%), m-BACOD The median of the average relative dose intensity calculated (52%), ProMACE-Cyta-BOM (50%) and MACOP-B (50%). for the initial cycles of THP-COPBLM given to all patients was Consequently, it was concluded that CHOP, the least lethal 0.94, ie 50% of the patients received more than 94% of the chemotherapy regimen, should be the standard chemotherapy projected dose rate, whereas 50% of the patients were given for advanced aggressive lymphoma. However, CHOP has not less intensive treatment. The median of the relative DI of CPA succeeded in improving the CR and long-term survival rates and THP were 0.95 and 0.93, respectively. All patients of NHL adequately, particularly in the L-I to H risk groups of received six cycles of chemotherapy which was the full dose IPI. New chemotherapy regimens for NHL need to be for 39 patients and less than the full dose for three patients. developed to attain CR more effectively. DeVita et al16 The median duration of G-CSF administration was 7 days. described a close association between dose intensity (DI) and G-CSF administration was well tolerated. survival time. Kwak et al17 reported a significant correlation between the relative DI of CPA and DXR and the survival per- iod in patients with diffuse large-cell lymphoma. Adverse events (Table 2) The use of G-CSF with the CHOP regimen has been increas- ing the DI in recent years. Usually, doses of key drugs (CPA Grade 3 or higher adverse events included granulocytopenia and DXR) are increased with a reduction of the interval in 39 patients (92.9%), and thrombocytopenia in seven between courses from 3 to 2 weeks. Shipp et al4 treated (16.7%). Median time of granulocytopenia was 3 days (range patients with advanced NHL and a poor prognosis using 0–10 days). Median time of thrombocytopenia was 5 days CHOP combined with G-CSF, which was administered at 3- (range 0–7 days). week intervals by a dose escalation method or by increasing In three patients with a reduced %DLCO, the dose of BLM the doses of CPA and DXR step to the maximum tolerated was reduced in the fifth and sixth cycles. An arrhythmia doses of 4000 mg/m2 and 70 mg/m2, respectively, and achi- (уgrade 3) was not found by EKG in any of the patients. The eved a high CR rate of 86%. left ventricular ejection fraction assessed by echocardiography In the present study, a CR rate of 88.1% was achieved with showed no change related to chemotherapy in any patients. a 3-year survival rate of 72.1% and 3-year event-free survival Fever (у38.0°C) was reported in three patients (7.1%), pneu- rate of 58% by the biweekly THP-COPBLM regimen given monia in two (4.7%) and acute bronchitis in one (2.4%). Anti- with G-CSF. None of the patients died of adverse events, indi- biotics and antifungal agents controlled infection in all cating that this chemotherapy was safe. According to the IPI patients. classification, the CR rates for the L-I, H-I and H risk groups were 94.1, 87.1 and 77.8%, respectively. The rates were rela- tively high even for the H-I and H risk groups, and were higher Discussion than those which we had achieved previously using biweekly COP-BLAM with G-CSF.1 The increased DI achieved by The results of trials of third-generation chemotherapy for non- increasing the dose of CPM from 500 to 750 mg/m2 and that Hodgkin’s lymphomas (NHL) were reported between the late of VCR from 1.2 to 1.4 mg/m2 may account for the high effi- 1970s and the 1980s. The phase II studies performed in single cacy. This suggests that improvement of efficacy depends on institutions using MACOP-B,2 ProMACE-Cyta-BOM14 and how far G-CSF used with cancer chemotherapy can prevent COP-BLAM III15 achieved CR rates of 70–84% and long-term neutropenia and increase the DI. Long-term studies are con- survival rates of 51–76%. In a study performed by the South- sidered necessary to assess the possible adverse events such west Oncology Group (SWOG) and the Eastern Cooperative as cardiomyopathy, liver dysfunction, as well as neutropenia. Oncology Group (ECOG) to compare four chemotherapy regi- In the case of cardiac toxicity, the cumulative dose is critical. Cardiomyopathy should be detected as early as possible because it is an irreversible process. Our previous study on the Table 2 Adverse events of biweekly THP-COPBLM therapy cardiotoxicity of COP-BLAM showed that cardiac adrenergic abnormalities and impaired mitochondrial function were asso- Grade ciated with a cumulative dose of DXR in excess of 400 mg/m.2,7 Because DXR was administered at 3-week inter- 01234 vals in the study, however, cardiac toxicity should be assessed with COP-BLAM given at 2-week intervals. In the present Hematological toxicity study, we used THP which has been reported to be less toxic Granulocytopenia 0 0 3 22 17 18 Thrombocytopenia 20 9 6 6 1 than DXR. Patients who achieved CR had received three Anemia 30 6 6 0 0 additional courses or a total of 300ෂ450 mg/m2 of THP with- Non-hematological toxicity out apparent cardiac adrenergic dysfunction or mitochondrial Nausea, vomiting 36 2 4 0 0 dysfunction. The left ventricular ejection fraction, as assessed Stomatitis 37 2 3 0 0 by echocardiography, was also not decreased. When THP was Diarrhea 40 2 0 0 0 used for multi-drug chemotherapy in place of DXR to treat Constipation 40 1 1 0 0 elderly patients with NHL, the modified treatment was less Liver dysfunction 32 6 4 0 0 cardiotoxic than the original regimen.19,20 Essentially, no stud- Renal dysfunction 40 2 0 0 0 Proteinuria 41 1 0 0 0 ies have been performed, however, to assess the cardiotoxicity Hematuria 39 3 0 0 0 of THP in younger patients. Pretreatment cardiac function is Arrythmia 40 2 0 0 0 not infrequently decreased in elderly patients with NHL. Some Peripheral neuropathy 35 4 3 0 0 elderly patients have a history of angina pectoris or myocar- Pulmonary toxicity 39 3 0 0 0 dial infarction. Drugs which will not cause acute or subacute adverse cardiac effects seem to be required. Because about Biweekly THP-COPBLM combined with G-CSF for NHL N Niitsu and M Umeda 1460 50% of patients with NHL currently achieve long-term sur- the Eastern Cooperative Oncology Group. Am J Clin Oncol 1982; vival as a result of progress in treatment, drugs are particularly 5: 649–655. required to treat younger patients. A long-term follow-up 10 The Non-Hodgkin’s Lymphoma Pathologic Classification Project. National Cancer Institute sponsored study of classifications of non- study is also necessary to assess the cardiotoxicity of the Hodgkin’s lymphomas; summary and description of working for- biweekly THP-COPBLM regimen used in the present study. mulation for clinical usage. Cancer 1982; 49: 2112–2135. 11 Kaplan EL, Meier P. Nonparametric estimation from incomplete observation. J Am Stat Assoc 1958; 53: 457–481. 12 World Health Organization. WHO handbook for reporting results References of cancer treatment. Geneva, Switzerland: WHO, 1979. 13 Hryniuk WM, Bush H. The importance of dose intensity in chemo- therapy of metastatic breast cancer. J Clin Oncol 1984; 2: 1 Niitsu N, Umeda M. Biweekly COP-BLAM regimen combined 1281–1288. with granulocyte colony-stimulating factor (G-CSF) for intermedi- 14 Goldie JH, Coldman AJ. A mathematics model for relating the ate-grade non-Hodgkin’s lymphoma. Eur J Haematol 1996; 56: directivity of tumors to their spontaneous mutation rate. Cancer 163–167. Treat Res 1977; 63: 1727–1733. 2 Klimo P, Connors JM. MACOP-B chemotherapy for the treatment 15 Boyd DB, Coleman M, Papish SW, Topilow A, Kopel SK, of diffuse large cell lymphoma. Ann Intern Med 1985; 102: Bernhardt B, Files JC, Schwartz S, Gaynor M, McDermott D. COP- 596–602. BLAM III: infusional combination chemotherapy for diffuse large 3 Fisher RI, Gaynor ER, Dahlberg S, Oken MM, Grogan TM, Mize cell lymphoma. J CLin Oncol 1988; 6: 425–433. EM, Glick JH, Coltman CA Jr, Miller TP. Comparison of a standard 16 DeVita VT, SM Hubbard Jr, Longo DL. The chemotherapy of lym- regimen (CHOP) with three intensive chemotherapy regimens for phoma – looking back moving forward: The Richard and Hinda advanced non-Hodgkin’s lymphoma. New Engl J Med 1993; 328: Rosenthal Foundation Award Lecture. Cancer Res 1987; 47: 1002–1006. 5810–5824. 4 Shipp MA. International non-Hodgkin’s lymphoma prognostic fac- 17 Kwak LW, Halpern J, Olshen RA, Horning SJ. Prognostic signifi- tors project. A predictive model for aggressive non-Hodgkin’s lym- cance of actual dose intensity in diffuse large-cell lymphoma – phoma. New Engl J Med 1993; 329: 987–994. results of a tree-structured survival analysis. J Clin Oncol 1990; 8: 5 Tanosaki R, Okamoto S, Akatsuka N, ishida A, Michikawa N, 963–977. Masuda Y, Uchida H, Murata M, Kizaki M, Ikeda Y. Dose escal- 18 Temma K, Akera T, Chugun A, Kondo H, Hagane K, Hirano S. ation of biweekly cyclophosphamide, doxorubicin, vincristine, Comparison of cardiac actions of doxorubicin, pirarubicin and and prednisolone using recombinant human granulocyte colony aclarubicin in isolated guinea-pig heart. Eur J Pharmacol 1993; stimulating factor in non-Hodgkin’s lymphoma. Cancer 1994; 74: 234: 173–181. 1939–1944. 19 Bastion Y, Blay JY, Divine M, Brise P, Bordessoule D, Sebban C, 6 Ariyoshi Y, Ogawa M. Dose intensity in cancer chemotherapy. Jpn Michel B, Tilly H, Lederlin P, Deconinck E, Salles B, Dumontet C, J Cancer Chemother 1994; 21: 2699–2707. Briere J, Coiffier B. Elderly patients with aggressive non-Hodgkin’s 7 Niitsu N, Yamazaki J, Igarashi M, Umeda M, Morishita T. Clinical lymphoma: disease presentation, response to treatment, and sur- usefulness of 123I-metaiodobenzylguanidine (MIBG) myocardial vival. A group d’etude des lymphomes de l’adulte study on 453 SPECT for cardiomyopathy due to adriamycin. Nucl Med Com- patients older than 69 years. J Clin Oncol 1997; 15: 2945–2953. mun 1994; 31: 1051–1057. 20 Niitsu N, Umeda M. THP-COPBLM (pirarubicin, cyclophospham- 8 Carbone PP, Kaplan HS, Musshoff K, Smithers DW, Tubiana M. ide, vincristine, prednisone, bleomycin and procarbazine) regi- Report of the committee on Hodgkin’s disease staging classi- men combined with granulocyte colony-stimulating factor (G-CSF) fication. Cancer Res 1971; 31: 1860–1861. for non-Hodgkin’s lymphoma in elderly patients: a prospective 9 Oken M, Creech R, Tormey D. Toxicity and response criteria of study. Leukemia 1997; 11: 1817–1820.