Elizabeth A. Chrischilles, PhD, Brian K. Link, MD, Shane D. Scott, PharmD, David J. Delgado, PhD, Moshe Fridman, PhD 
 
Abstract and Introduction
Abstract
Background: Six to eight cycles of CHOP therapy (cyclophosphamide, doxorubicin, vincristine, and prednisone) is standard for intermediate-grade non-Hodgkin's lymphoma (NHL) but is associated with toxicity that may cause premature termination of therapy.
Methods: We studied factors associated with premature termination of CHOP therapy (receiving <6 cycles) and the relationship of premature termination with survival. Subjects consisted of a population-based sample of Iowa residents with intermediate-grade NHL who were planned to receive >/= 6 cycles of CHOP and who were chemosensitive (ie, achieved a documented partial or complete response to CHOP).
Results: In a comparison with patients 18-59 years of age, the odds of premature termination of CHOP therapy was 2.6 (95% CI, 0.7-9.2) for those aged 60-74 and 6.2 (95% CI, 1.7-23.3) for those aged >/=75. Patients with cycle 1 febrile neutropenia hospitalization (FNH) were 4.4 times (95% CI, 1.4-13.8) more likely to terminate CHOP prematurely than those without cycle 1 FNH. Among patients aged 60-74, but not those aged >/= 75 premature termination appeared to be associated with decreased 5-year survival (hazard ratio [HR] = 6.0; 95% CI, 2.4-15.2) compared with those completing CHOP therapy (HR = 2.1; 95% CI, 1.0-4.2). Findings for overall survival were similar.
Conclusions: First-cycle FNH and age >/=60 years were associated with premature termination of CHOP therapy. The association of premature termination with survival among chemosensitive patients differed by age.

Introduction
Intermediate-grade non-Hodgkin's lymphoma (NHL), as classified by the Working Formulation with diffuse large-cell lymphoma as the prototype, represents a group of potentially curable diseases.^[1] Anthracycline-based multi-agent chemotherapy has been the mainstay of curative treatment strategies for NHL.^[2] The impact of dose intensity (DI; the amount of chemotherapy delivered per unit of time) on cure rates in NHL has been a subject of widespread interest. Retrospective data suggest that diminished DI, with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) as a standard,^[3] may be associated with lower response rates.^[4,5] Less is known about the importance of duration of chemotherapy in patients with chemosensitive disease. Debate concerning the duration of chemotherapy in advanced NHL centers on whether 6 or 8 cycles of CHOP should be considered standard for patients with advanced-stage disease. For patients with limited-stage disease, 3 or 4 cycles of chemotherapy followed by involved-field radiation therapy comprise an acceptable alternative to more-prolonged chemotherapy.^[6]

Emerging data suggest that delivery of planned chemotherapy is often incomplete, especially in the elderly, due to toxicities encountered early in the chemotherapy course.^{[7- 9]} In a previous analysis of this study population, the incidence of febrile neutropenia among those 65 years of age and older was 34% compared with 21% among patients under age 65.^[10] Febrile neutropenic hospitalization rates were 28% (95% confidence interval [CI], 26% to 30%) among patients >/= 65 years of age and 16% (95% CI, 14% to 18%) among patients <65 years of age.

The survival advantage or disadvantage of completing a full 6 cycles of chemotherapy is unknown in patients who respond to chemotherapy but experience significant toxicity that prompts consideration of dis-continuation. Our retrospective, historical study links observed practice patterns with survival data, describes risk factors associated with receiving <6 cycles of CHOP chemotherapy among those planned to receive at least 6 cycles, and assesses the relationship of administering <6 cycles of CHOP chemotherapy with survival.

Next 
 
Elizabeth A. Chrischilles, PhD, Brian K. Link, MD, Shane D. Scott, PharmD, David J. Delgado, PhD, and Moshe Fridman, PhD

From the University of Iowa, Iowa City, Iowa (EAC, BKL, SDS), Amgen Inc, Thousand Oaks, California (DJD), and AMF Consulting, Los Angeles, California (MF).

 
Cancer Control 10(5):396-403, 2003. © 2003 H. Lee Moffitt Cancer Center and Research Institute, Inc.