Localized Aggressive Lymphoma: Combined Modality Therapy is Better than Chemothe
September 1, 1998
Localized Aggressive Lymphoma: Combined Modality Therapy is Better than Chemotherapy Alone
ABSTRACT & COMMENTARY
Historically, patients with apparently localized, intermediate- or high-grade non-Hodgkin's lymphoma (NHL) were treated with radiation therapy alone. However, cure rates were highly variable and largely disappointing. The use of combination chemotherapy either with or without radiation therapy produced more consistent and higher rates of long-term remission. In several series, 84-91% of patients were long-term disease-free survivors. The role of radiation therapy in producing these favorable results has been difficult to discern because of the absence of controlled data on its use. In most series, abbreviated courses and/or modified doses of chemotherapy were administered, and radiation therapy was viewed as permitting the use of less chemotherapy. However, Miller and Jones, on the basis of their experience at the University of Arizona, argued that radiation therapy did not appear to be necessary to achieve excellent results.1 Yet, their excellent results had been obtained by including radiation therapy under certain circumstances, such as for patients with bulky disease or those who responded more slowly to treatment or had underlying illnesses prompting dose modification. Thus, it could be (and was) argued that the patients in whom radiation therapy was used had a poorer prognosis and that fact was responsible for the observation that retrospective analysis of patients treated with chemotherapy alone or combined modality therapy showed no significant differences in outcome.2
Southwest Oncology Group (SWOG) study 8736 undertook to address the question of the role of radiation therapy in a prospective randomized trial. Patients were enrolled with biopsy-proven intermediate- or high-grade NHL (excluding patients with lymphoblastic lymphoma). The dominant histology was diffuse large B-cell lymphoma. All previously untreated, ambulatory patients (except those with congestive heart failure, history of other cancer, or those with CNS involvement) with stage I, IE (including bulky disease), non-bulky stage II, or non-bulky stage IIE disease were eligible. Bulky disease was defined as a mediastinal mass greater than one-third the chest diameter or any mass greater than 10 cm in diameter. To normalize at least some of the risk factors, patients were stratified by age (65 years or over vs younger than 65 years), stage (I vs II) and histologic subtypes, and whether all visible disease had been resected. Treatment consisted of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) for three (21 day) cycles followed by involved-field radiation therapy (4000-5500 cGy) or CHOP for eight cycles without radiation therapy. Of 442 patients registered, 41 were excluded from analysis, primarily because of histological reclassification by the study pathologists. Of the remaining patients, 200 received three cycles of CHOP followed by radiation therapy and 201 received eight cycles of CHOP without radiation therapy.
With a median follow-up of 4.4 years, those treated with three cycles of CHOP and radiation therapy had significantly better progression-free survival (77% at 5 years vs 64% for chemotherapy alone; P = 0.03) and overall survival (82% vs 72%; P = 0.02) than patients treated with eight cycles of CHOP. Furthermore, although there were only two deaths in the study, life-threatening toxicity occurred in 30% of those on the combined modality arm, but in 40% of those receiving CHOP alone. Prognostic factor analysis demonstrated that patients with three risk factors (age > 60 years, stage II disease, LDH elevation, poor performance status) had only a 34% five-year progression-free survival. This subset accounted for less than 10% of all patients. Miller and colleagues concluded that three cycles of CHOP followed by radiation therapy is superior to eight cycles of CHOP alone in the treatment of localized intermediate- or high-grade NHL. (Miller TP, et al. N Engl J Med 1998;339:21-26.)
COMMENTARY
This study addressed an important question in the subset of aggressive lymphoma patients with localized disease-the role of radiation therapy. Radiation therapy is more effective in the setting of low-volume disease. Aggressive histology lymphoma tends to spread hematogenously early in its course. Thus, it fits with our preconceived notions that a brief course of chemotherapy reduces the tumor volume in the primary site and eradicates micrometastatic disease, while the radiation therapy sterilizes the primary site.
However, several questions remain to be answered and not all the information that one would desire was in the paper. The first concern deals with differences in the pattern of relapse based upon the treatment approach. If radiation therapy was truly making an impact on survival, one would expect that patients treated with chemotherapy alone would have higher rates of recurrence in the primary site of disease than would patients treated with combined modality therapy. However, no data on sites of relapse were presented. If the rate of recurrence in the primary site was the same on both arms, it would be difficult to understand the role of radiation therapy in the treatment outcome.
Second, it would be of great value to know the dose intensity with which CHOP was given on each arm for the first three cycles. A potential problem is that physicians are more likely to modify doses and schedules of administration when a patient needs to get through eight cycles of chemotherapy than when the patient is facing only three cycles. Data demonstrating that the dose intensity of CHOP was identical on both arms during the first three cycles would go a long way toward ruling out a dose-intensity related explanation for the observed differences. The choice of eight cycles of CHOP seems odd. Nearly 30% of the patients in this study were rendered free of disease by their diagnostic excisional biopsy. I don't think many people would consider eight cycles of CHOP appropriate treatment for a patient without measurable disease. Thus, a persistent question is whether three cycles of CHOP plus radiation therapy are more effective than 4-6 cycles of CHOP alone.
Another unresolved issue is the risk of second solid tumors related to the use of radiation therapy in these patients. The duration of follow-up is not sufficient to assess the magnitude of the second malignancy risk. Of course, the radiation fields were not large and even if late second tumors appear in such patients, the early advantage for combined modality therapy (10% survival advantage at 5 years) is not going to be influenced by second cancers that develop in the second and third decades after treatment.
It is exceedingly unlikely that additional controlled trials will be undertaken in patients with localized aggressive histology lymphoma on the basis of the questions that are currently not resolved. Thus, it is important that this group of 400 patients be reported in greater detail than is permitted within the space constraints of the New England Journal of Medicine. I hope that the next follow-up of this group of patients includes data on dose-intensity, sites of relapse, and second malignancies.