1Lymphoma Service, and Departments of 3 Medicine, 4 Pathology, 2 Epidemiology and Biostatistics and 5 Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, USA
* Correspondence to: Dr C. S. Portlock, MD, Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10021, USA. Tel: +1-212-639-8109; Fax: +1-646-422-2285; Email: portlocc{at}mskcc.org
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods: The phase II NHL-15 protocol comprised: (i) induction [doxorubicin 60 mg/m2 i.v. on weeks 1, 3, 5 and 7 plus vincristine 1.4 mg/m2 i.v. (no cap) on weeks 1, 2, 3, 5 and 7]; and (ii) consolidation (cyclophosphamide 3000 mg/m2 i.v. on weeks 9, 11 and 13 plus granulocyte colony-stimulating factor 5 µg/kg subcutaneous on days 310 following each cyclophosphamide dose). Patients with aggressive non-Hodgkin's lymphomas (working formulation: intermediate grade or immunoblastic), bulky stage I and stages IIIV, were eligible.
Results: There are 165 eligible patients with a 6.9-year median follow-up (range 0.5141 months) and a median age of 48 years. For the entire group, 72.1% achieved complete remission, and at 5 years disease-free survival was 57.8% and overall survival (OS) was 62.2%. Ideal dose delivery was >90%. Acute and late toxicities of treatment were manageable and acceptable. Toxic death on treatment was 2.4%. When the diffuse large cell lymphoma histologies were grouped according to the International Prognostic Index (IPI), complete remission and OS in the low-intermediate (LI), and high-intermediate (HI) risk groups were improved by 5%15% compared with historical CHOP (cyclophosphamide, doxorubicin, vincristine and prednisone). This improvement was also noted for LI and HI risk groups in the age-adjusted (aa)IPI analysis for patients 60 years of age.
Conclusions: The NHL-15 program can be administered safely and effectively to achieve high rates of durable remission when used for the treatment of advanced stage, aggressive, non-Hodgkin's lymphomas. The 5%15% improvement in 5-year OS compared with historical CHOP, according to the IPI/aaIPI model (in LI and HI risk groups), is encouraging. Further evaluation and prospective testing of the NHL-15 protocol appears to be warranted.
Key words: aggressive lymphomas, dose-dense, dose-intense, NHL-15, non-Hodgkin's lymphomas
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Based on the tumor growth kinetic models of Norton and Simon [3], three concepts emerged: (i) dose intensification, in which the dose per unit time is increased; (ii) dose densification, in which chemotherapy is delivered at reduced intervals; and (iii) the utilization of sequential, non-cross-resistant agents/regimens to address tumor cell heterogeneity. Each of these concepts has been tested in aggressive lymphomas and, to date, the results have been mixed. Stumbling blocks to evaluating results have been excess toxicity and the ability to salvage some of those who fail initial therapy. In addition, the recent introduction of rituximab (anti-CD20 antibody) in combination with CHOP (R-CHOP) has led to a significant advance in outcome without major additional toxicity in CD20-positive aggressive lymphomas [4
].
Nevertheless, it is essential that further attempts to improve standard combination chemotherapy continue, as emphasized in a recent editorial by Coiffier [5]. The NHL-15 is a novel, dose-intense, dose-dense, sequential regimen designed to improve outcome, which was studied before the availability of rituximab. The results of this mature clinical trial are reported herein.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Dose modifications
Induction
For granulocyte counts >1000 at the time of treatment, doxorubicin was given at full dose. If granulocyte count was >200 but <1000, 65% of the doxorubicin dose was given. For granulocyte counts <200, treatment was held for 1 week. Vincristine dose was not modified for low granulocyte count unless <200, in which case vincristine was also delayed 1 week and administered along with doxorubicin. For grade 3 or 4 constipation, vincristine was held until resolution. The next vincristine dose was reduced by 50% and subsequent doses were escalated by 25% back to baseline as tolerated. Vincristine was discontinued if objective motor weakness was noted or the development of hoarseness of the voice.
Intensification
Cyclophosphamide was held until the granulocyte count was >1000 and platelet count was >50 000, at which point it was given in full dose. Platelet transfusions were administered for platelet counts <20 000/µl.
Patient eligibility
Patients could not have received previous chemotherapy. Enrollment was limited to patients with the following histological diagnoses: diffuse large cell, diffuse large cell immunoblastic type, diffuse mixed small and large cell, follicular large cell, follicular large cell with diffuse areas, and diffuse small cleaved cell lymphomas as defined in the Working Formulation [6]. Patients with Ann Arbor stage IIIV disease, or stage I with a >10 cm tumor mass, were eligible. Primary central nervous system lymphomas were not eligible, and neither were patients presenting with systemic disease who were known to have parenchymal brain or leptomeningeal involvement. All patients were required to have negative serologies for HIV and hepatitis B, and to have serum bilirubin <2 mg/dl and normal baseline cardiac function as defined by multigated equilibrium radionuclide angiography (MUGA) scan or echocardiogram.
Staging evaluation
Diagnostic tissue was obtained for all patients and was reviewed by one of two hematopathologists in the Department of Pathology, Memorial Hospital. A posterior iliac crest bone marrow biopsy and computerized axial tomographic scans of the chest, abdomen and pelvis were obtained for all patients. Lumbar puncture and cytological examination of the cerebrospinal fluid was performed on all patients who met the criteria for intrathecal prophylaxis listed above. Pre- and post-treatment gallium scanning was not performed routinely on the first 35 patients, but was done for subsequent patients. Additional diagnostic studies were performed as clinically indicated.
Response definitions
Response evaluation
Interim response evaluations and final determinations of response were conducted between weeks 7 and 9, and between weeks 16 and 17, respectively. Diagnostic studies that defined the initial extent of disease were repeated at these times. Patients demonstrating no response or those with evidence of progression following induction were considered treatment failures and did not go on to intensification. Patients found at the time of final response evaluation to have evidence of residual or progressive lymphoma were considered treatment failures. Residual clinical or radiographic abnormalities remaining at the end of treatment were subjected to surgical or computed tomography (CT)-guided needle biopsy in every case where the risk of the procedure could be justified. Patients whose post-treatment diagnostic evaluation failed to detect evidence of disease were considered to be in remission and received no further treatment. All instances of recurrent disease were confirmed pathologically.
Statistics
Relapse-free survival (RFS) for patients who achieved CR was defined as the time between the end of treatment and the time of relapse or death or time of last follow-up for patients who had not relapsed. Progression-free survival (PFS) was defined as the interval between the start of treatment and the time of treatment failure or relapse, or the time of last follow-up for patients who had neither failed nor relapsed. Overall survival (OS) was defined as the interval between the beginning of treatment and either death or time of last follow-up (Figure 2).
|
|
|
Drug delivery and dose intensity determinations
The ideal dose intensities for NHL-15 were calculated by dividing the total dose of drug planned by the planned duration of time in weeks over which the drug was to be administered, and was expressed in units of mg/m2/week. The actual dose intensity was calculated by dividing the actual amount of drug given by the actual number of weeks elapsed between administration of the first dose and recovery from the last dose. This was done for each patient treated and the data presented are the averages of the dose intensities for individual patients.
Toxicity
Toxicity assessment was made according to National Cancer Institute (NCI) Common Toxicity Criteria.
Informed consent
The Institutional Review Board of Memorial Hospital approved this protocol. All patients enrolled provided written informed consent.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Diffuse large cell lymphoma (including immunoblastic type) and diffuse mixed small and large cell lymphoma represented 73.3% and 10.3% of the evaluable patients, respectively, for a total of 138 (83.6%) patients. This group of patients was analyzed according to the IPI and the aaIPI for those 60 years of age [7
], since these were the histological subtypes for which this model was developed. All IPI analyses were performed retrospectively and data were reviewed independently by three investigators (C.S.P., A.D.Z. and C.H.M.). One patient had inadequate information and was not included in this portion of the analysis. Among the 137 patients in the IPI groups were 14 patients with T-cell lymphomas (three of whom were CD30-positive while 11 were CD30-negative), categorized as either diffuse large cell, or diffuse mixed small and large cell lymphomas. These patients were included in the IPI and aaIPI analyses and were distributed equally among prognostic groups. Follicular large cell lymphoma was present in 12.7% of patients and mantle cell lymphoma was present in 4%. Patient characteristics are shown in Table 1 and compared with the IPI model [8
].
|
|
|
Late toxicity assessment was made in 95 patients and these are tabulated in Table 4. Although a thorough review was made of these records, there may be under-reporting in this retrospective cohort. Late cardiovascular events were present in seven patients (7.4%), with other malignancies in 12 (12.6%). There has been no case of myelodysplasia; one patient developed acute myelogenous leukemia following autologous transplantation.
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
When initiated in 1991, it was the intention of this study to treat a representative population of patients with advanced-stage aggressive non-Hodgkin's lymphomas. For this purpose, advanced stage was considered any disease presentation other than non-bulky stage I, and aggressive lymphomas were considered to be those categorized as either intermediate grade or diffuse large cell immunoblastic (categories DH) according to the Working Formulation [6]. The median age of patients in this study was 48 years, similar to our previous experience [1
] but less than that of the IPI model (median 56 years).
A principal goal of the NHL-15 was to increase significantly the dose intensity of doxorubicin, vincristine and cyclophosphamide, and our data demonstrate clearly that this was achieved (see Table 2). The structure of NHL-15 differs from conventional chemotherapy programs by administering drugs sequentially by class (i.e. natural products followed by alkylating agent), thus separating the delivery of doxorubicin and cyclophosphamide, and eliminating the overlap of hematological toxicities that normally limit drug delivery. Moreover, no prednisone was utilized, eliminating some of the potential morbidity seen with CHOP and similar regimens. While facilitating increased dose intensity, the sequential delivery of chemotherapy was not simply a manoeuver to achieve a mathematical end point, but rather an approach emanating from the principles of the NortonSimon hypothesis [3]. Whether the results of the NHL-15 were adversely affected by the elimination of prednisone and the loss of potential synergy by separating cyclophosphamide and doxorubicin must also be considered.
The toxicity of the NHL-15 regimen (Table 3) generally required only routine medical management. Neutropenia did not limit the high percentage of targeted dose intensity that was achieved. However, almost 50% of patients were hospitalized at least once for transient nadir fever, 65% of these admissions during the high-dose cyclophosphamide phase. With rare exceptions, these episodes were brief and uncomplicated. Almost half of all patients received at least one red blood cell transfusion; however, erythropoietin was not utilized. Thrombocytopenia was not seen, except for disease-related indications.
With only three drugs in the NHL-15, it was our goal to use each to maximum advantage within the confines of acceptable limits of toxicity. Neurotoxicity was generally more common among older patients, but was not restricted to this group. Whether the level of dose intensification of vincristine attained here is essential remains uncertain. Nonetheless, one must be acutely sensitive to early changes in voice or peripheral motor neuropathy. Discontinuation of vincristine at this point should minimize these troubling toxicities. With careful application of the prescribed dose modifications, however, this regimen can be administered with acceptable toxicity.
With a median follow-up of 6.9 years, late toxicities of the NHL-15 regimen could be assessed. These are listed in Table 4 and appear to be no different than might be expected with CHOP.
Recently, Coiffier [5] summarized all prospective, randomized trials demonstrating a statistically significant benefit of the dose-intense experimental arm as compared with CHOP. He concluded that the weight of these studies suggests that CHOP may be improved by increasing doses and by reducing intervals between cycles. However, Coiffier also cautioned that the R-CHOP regimen [4
] appears to provide similar benefit without exposing patients to the more toxic experimental arm. The non-myeloablative experimental data are summarized in Table 7 and provide a benchmark comparison for the NHL-15 protocol. Our mature data appear to compare favorably with the experimental arms of these prospective trials. In the future it will be of interest to determine the additional benefit of rituximab to such dose-intense regimens.
|
The NHL-15 protocol is a novel, dose-intense, dose-dense, sequential chemotherapy regimen that can be delivered safely at near ideal dose intensity. The 5%15% improvement in 5-year OS as compared with historical CHOP, according to the IPI/aaIPI model (in low-intermediate and high-intermediate risk groups), is encouraging. Further evaluation and prospective testing of the NHL-15 appears to be warranted; we are exploring the feasibility of combining Rituximab with the NHL-15 regimen.
![]() |
Acknowledgements |
---|
![]() |
Notes |
---|
Received for publication March 29, 2004. Revision received May 11, 2004. Accepted for publication May 19, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Fisher RI, Gaynor ER, Dahlberg S et al. Phase III comparison of CHOP vs m-BACOD vs ProMACE-CytaBOM vs. MACOP-B in patients with intermediate or high grade non-Hodgkin's lymphoma: results of SWOG 8516 (Intergroup 0067) the national high priority lymphoma study. N Engl J Med 1993; 328: 10021006.
3. Norton L, Simon R. The NortonSimon Hypothesis revisited. Cancer Treat Rep 1986; 70: 163169.[ISI][Medline]
4. Coiffier B, Lepage E, Briere J et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002; 346: 235242.
5. Coiffier B. Increasing chemotherapy intensity in aggressive lymphomas: a renewal? J Clin Oncol 2003; 21: 24572459.
6. Non-Hodgkin's Lymphoma Pathologic Classification Project. National Cancer Institute sponsored study of classifications of non-Hodgkin's lymphomas: summary and description of a working formulation for clinical usage. Cancer 1982; 49: 21122135.[ISI][Medline]
7. Cheson BD, Horning SJ, Coiffier B et al. Report of an International Workshop to Standardize Response Criteria for Non-Hodgkin's Lymphomas. J Clin Oncol 1999; 17: 12441253.
8. The International Non-Hodgkin's Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin's lymphoma. N Engl J Med 1993; 329: 987994.
9. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457481.[ISI]
10. Linch DC, Smith P, Hancock BW et al. A randomized British National Lymphoma Investigation trial of CHOP vs. a weekly multi-agent regimen (PACEBOM) in patients with histologically aggressive non-Hodgkin's lymphoma. Ann Oncol 2000; 11 (Suppl 1): 8790.[CrossRef][Medline]
11. Pfreundschuh M, Trümper L, Kloess M et al. 2-weekly vs. 3-weekly CHOP with and without etoposide for patients >60 years of age with aggressive non-Hodgkin's lymphoma: results of the completed NHL-B-2 trial of the DSHNHL. Blood 2002; 100 (Suppl 1): 774a.[CrossRef]
12. Pfreundschuh M, Trümper L, Schmits R et al. 2-weekly vs. 3-weekly CHOP with and without etoposide in young patients with low-risk (low LDH) aggressive non-Hodgkin's lymphoma: results of the completed NHL-B-1 trial of the DSHNHL. Blood 2002; 100 (Suppl 1): 110a.
13. Tilly H, Lepage E, Coiffier B et al. Intensive conventional chemotherapy (ACVBP regimen) compared with standard CHOP for poor-prognosis aggressive non-Hodgkin lymphoma. Blood 2003; 102: 42844289.
14. O'Brien JP, O'Keefe P, Roistacher N et al. NHL-15 protocol for diffuse aggressive lymphomas: a dose intense regimen of doxorubicin, vincristine and cyclophosphamide. Blood 1992; 80: 157a.
15. O'Brien JP, O'Keefe P, Alvarez A et al. The Memorial Sloan-Kettering NHL-15 protocol for intermediate grade and immunoblastic lymphoma. Hem Onc Annals 1993; 1: 4750.