The Oncologist, Vol. 12, No. 12, 1416-1424, December 2007; doi:10.1634/theoncologist.12-12-1416 © 2007 AlphaMed Press
Elderly Cancer Patients Receiving Chemotherapy Benefit from First-Cycle PegfilgrastimaH. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, USA; bCabrini Center for Cancer Care at Christus St. Frances Cabrini Hospital, Alexandria, Louisiana, USA; cPacific Cancer Medical Center, Anaheim, California, USA; dGeriatric Oncology Consortium (GOC), Baltimore, Maryland, USA; eAmgen Inc., Thousand Oaks, California, USA; fClinical Research Branch, National Institute on Aging, and Institute for Advanced Studies in Aging, Washington, District of Columbia, USA Key Words. Elderly • Neutropenia • Chemotherapy • Cancer Correspondence: Lodovico Balducci, M.D., H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, Florida 33612, USA. Telephone: 813-979-3822; Fax: 813-972-8359; e-mail: Balducci{at}moffitt.usf.edu Received March 12, 2007; accepted for publication November 6, 2007. Disclosure: This study was sponsored by Amgen Inc., Thousand Oaks, CA. L.B., H.Al-H., V.C., and W.B.E. have received research support from Amgen Inc., and have acted as consultants on advisory boards for Amgen Inc. L.B. and W.B.E. have served on a speakers bureau for Amgen Inc. L.D. and S.S. are employees of and stockholders of Amgen Inc. V.C. holds stock in Amgen Inc. No other potential conflicts of interest were reported by the authors, planners, reviewers, or staff managers of this article.
Background. There is a misconception that elderly cancer patients cannot tolerate standard doses of chemotherapy because of the frequency and severity of myelosuppressive complications. The reactive use of colony-stimulating factors (i.e., in response to severe neutropenia) commonly observed in this setting contributes to the frequency and severity of these complications. This study evaluated the incidence of febrile neutropenia and related events in elderly cancer patients receiving pegfilgrastim beginning with cycle 1 (proactive) in comparison with pegfilgrastim initiated after cycle 1 at the physician's discretion (reactive).
Methods. Patients ( Results. There were 852 patients enrolled (median age, 72 years). Proactive pegfilgrastim use resulted in a significantly lower incidence of febrile neutropenia for both solid tumor and NHL patients compared with reactive use. Proactive pegfilgrastim use also led to fewer hospitalizations resulting from neutropenia and febrile neutropenia by approximately 50%. Antibiotic use was lower for solid tumor patients receiving proactive pegfilgrastim and equivalent in the two NHL groups. Conclusions. This is the largest, randomized, prospective trial evaluating growth factor support in typical elderly cancer patients. Proactive pegfilgrastim use effectively produced a lower incidence of febrile neutropenia and related events in elderly patients with either solid tumors or NHL receiving an array of mild to moderately neutropenic chemotherapy regimens. Pegfilgrastim should be used proactively in elderly cancer patients to support the optimal delivery of standard chemotherapy.
Approximately two thirds of cancers occur in people 65 years of age [1], but many in this age group receive dose reductions from standard chemotherapy regimens [2–5], potentially reducing response and survival rates [6–9]. Although some studies have demonstrated that elderly patients may obtain treatment benefits comparable to those of younger patients [10–12], there is a misconception that they, particularly those with pre-existing comorbidities, cannot tolerate standard doses of chemotherapy because of the risk for toxicity, including that of febrile neutropenia [13]. This risk may be exaggerated because colony-stimulating factors are often used reactively, rather than proactively, in clinical practice [14]. The use of pegfilgrastim from the first cycle has been shown to reduce the incidence of febrile neutropenia in patients receiving moderately myelosuppressive chemotherapy [15]. The initiation of colony-stimulating factors from the first cycle onwards to minimize neutropenic events may be particularly beneficial in elderly patients [2, 16–18], for whom neutropenic infections are generally more serious than in other populations and result in longer, more frequent, and costlier hospitalizations [19–22]. Therefore, the aim of this study was to compare the proportion of elderly patients experiencing febrile neutropenia when receiving pegfilgrastim from the first cycle (proactive) with the proportion experiencing febrile neutropenia when treated with the current practice of administering pegfilgrastim in response to observed severe neutropenia or neutropenia-related events (reactive). This randomized, controlled trial was designed in collaboration with the Geriatric Oncology Consortium (GOC), an organization established to improve the representation of typical elderly patients with cancer in clinical trials. With the exception of some studies of patients with non-Hodgkin's lymphoma (NHL) [23–25], there are few clinical trials studying the effectiveness of standard cancer treatment regimens in older individuals. This is a result of the restrictive eligibility criteria of many clinical trials, which do not accommodate the diversity (in terms of function and comorbidities) of this population [26]. The inclusion criteria for this study were specifically designed to be inclusive of elderly patients and, unlike most traditional clinical studies, allowed patients to enroll with comorbidities.
Study Design This phase IV, open-label, randomized, multicentered trial compared first (and subsequent) cycle pegfilgrastim use (proactive) with the current community clinical practice (which may include pegfilgrastim in later cycles) (reactive) in elderly patients. The study was designed by Amgen Inc. (Thousand Oaks, CA) in collaboration with the GOC. The research protocol was approved by the institutional review board of each participating institution, and all patients provided written informed consent prior to receiving treatment with pegfilgrastim.
Patients
Randomization and Treatment The randomization schedules, which were stratified by solid tumor type (breast, ovary, lung) and NHL, were generated using permuted blocks by statisticians not involved with the studies. Site personnel called a central randomization center to receive patient treatment assignment. Patients were assigned in a 1:1 ratio to either proactive pegfilgrastim or secondary prophylaxis with pegfilgrastim. Neither the site personnel, the patients, nor the study team was blinded to patient treatment. Following screening, patients were randomly assigned to receive a s.c.injection of 6 mg pegfilgrastim once per cycle 24 hours after chemotherapy completion, beginning either with cycle 1 (pegfilgrastim all-cycles arm) or after cycle 1 (in any cycle) at the physician's discretion (physician-discretion arm). Commercial pegfilgrastim was used in this study. After cycle 1, options in the physician-discretion arm were: no intervention, chemotherapy dose reduction, chemotherapy delay, or administration of pegfilgrastim. Blood samples for complete blood counts with five-part differential were to be drawn on day 1 of each cycle of chemotherapy; during cycle 1 on day 8, once between day 11 and 13, and on day 15 to determine approximate nadir; and at approximate nadir in subsequent cycles.
Endpoints and Assessments
Statistical Analysis The primary analysis set included all randomized patients who had signed informed consent before invasive, protocol-specified procedures were carried out and had received at least one cycle of chemotherapy. Safety data were tabulated and presented descriptively for all patients who received at least one cycle of chemotherapy.
Rates of febrile neutropenia between the two treatment groups were compared using Fisher's exact test (
Patients and Treatment Between June 2002 and November 2004, 852 patients were enrolled at 192 community cancer centers nationwide. As the study progressed, many participating physicians expressed a reluctance to randomize NHL patients to a non–first cycle pegfilgrastim arm, because of the associated higher rates of febrile neutropenia. As a result, recruitment proceeded rapidly for all tumor types except NHL. Enrollment was therefore halted once the expected number of patients with solid tumors had been enrolled. Of 701 solid tumor patients enrolled, 349 were randomized to the pegfilgrastim all-cycles arm and 352 were randomized to the physician-discretion arm; 15 were excluded from the primary analysis set (11 received no chemotherapy, four initiated chemotherapy before signing informed consent), leaving 343 patients in each arm. Nine solid tumor patients randomized to the physician-discretion arm of the study received first-cycle pegfilgrastim as a result of administrative errors. These patients were analyzed as randomized for the primary analysis, but as treated for the safety analysis. Of the 151 NHL patients enrolled, 75 were randomized to the pegfilgrastim all-cycles arm and 76 to the physician-discretion arm; five were excluded from the primary analysis set (two received no chemotherapy, three initiated chemotherapy before signing informed consent), leaving 73 in each arm (online supplementary Fig. 1).
Baseline characteristics were similar between groups and are summarized in Table 2. The median age was 72 years (range, 65–88 years), with approximately two thirds of patients
Among solid tumor patients, 198 (58%) in the pegfilgrastim all-cycles arm and 175 (51%) in the physician-discretion arm completed their planned chemotherapy, with 42% of patients in the physician-discretion arm receiving pegfilgrastim during the study period. The most common reason for growth factor initiation was grade 3 or 4 neutropenia in a preceding cycle. Patients with breast cancer had the highest study completion rate (78% in the pegfilgrastim all-cycles arm and 67% in the physician-discretion arm), followed by patients with ovarian cancer (64% in the pegfilgrastim all-cycles arm and 53% in the physician-discretion arm), and patients with lung cancer (46% in the pegfilgrastim all-cycles arm and 43% in the physician-discretion arm). Overall, for patients with solid tumors, the most common reasons for discontinuation were disease progression (10%), administrative/investigator decision (10%), and adverse events (9%). For NHL patients, 38 (52%) in each arm completed their planned chemotherapy, with 64% of patients in the physician-discretion arm receiving pegfilgrastim during the study period, mostly for grade 3 or 4 or febrile neutropenia in a preceding cycle. The attrition rate for NHL patients in cycle 1 was noticeably different between the two arms: only four patients discontinued study in the pegfilgrastim all-cycles arm compared with 14 patients in the physician-discretion arm. Over all cycles, the most common reasons for discontinuation were adverse events (16%), administrative/investigator decision (7%), and death (7%). Overall, the reasons for discontinuation were similar between the two treatment groups for both solid tumor and NHL patients. An increasing trend in pegfilgrastim use was observed by chemotherapy cycle (Fig. 1). Although the protocol stated that no growth factor was to be administered in the first cycle for patients in the physician-discretion arm, 3% of solid tumor patients received pegfilgrastim (as a result of administrative errors), and 7% of solid tumor patients and 27% of NHL patients received filgrastim in response to neutropenia in cycle 1.
Efficacy Incidence of Febrile Neutropenia As most neutropenic events usually occur in cycle 1, we examined febrile neutropenia rates in this cycle alone: 3% (95% CI, 1%–5%) of solid tumor patients in the pegfilgrastim all-cycles arm experienced febrile neutropenia versus 7% (95% CI, 4%–10%) in the physician-discretion arm. Similarly, 7% (95% CI, 2%–15%) of NHL patients in the pegfilgrastim all-cycles arm experienced febrile neutropenia versus 25% (95% CI, 15%–36%) in the physician-discretion arm (Fig. 2). The use of pegfilgrastim from the first cycle resulted in a lower incidence of febrile neutropenia in patients treated with the most common chemotherapy regimens (those received by approximately 90% of the study participants) (Fig. 3).
Incidence of Neutropenia and Neutropenia-Related Events The incidence of grade 4 neutropenia was 58% (95% CI, 53%–64%) in the physician-discretion arm and 22% (95% CI, 18%–27%) in the pegfilgrastim all-cycles arm among solid tumor patients, and was 86% (95% CI, 76%–93%) in the physician-discretion arm and 75% (95% CI, 64%–85%) in the pegfilgrastim all-cycles arm among NHL patients. Considering cycle 1 alone, 26% (95% CI, 21%–31%) of solid tumor patients in the pegfilgrastim all-cycles arm experienced grade 3 or 4 neutropenia versus 68% (95% CI, 63%–71%) in the physician-discretion arm. Similarly, 69% (95% CI, 57%–79%) of NHL patients in the pegfilgrastim all-cycles arm experienced grade 3 or 4 neutropenia in cycle 1 versus 88% (95% CI, 78%–94%) in the physician-discretion arm. There were more hospitalizations resulting from neutropenia and febrile neutropenia across all cycles among patients randomized to the physician-discretion arm than among patients receiving pegfilgrastim from the first cycle for both the solid tumor (9%; 95% CI, 6%–12% versus 5%; 95% CI, 3%–7%) and NHL (37%; 95% CI, 26%–49% versus 17%; 95% CI, 10%–28%) subsets (Fig. 4A, B). NHL patients were hospitalized approximately three times as often as solid tumor patients in both treatment arms. For both the solid tumor and NHL subsets, the majority of hospitalizations occurred during cycle 1.
Among NHL patients in the physician-discretion arm hospitalized during cycle 1, the majority (10 of 18) received pegfilgrastim from the second cycle as an alternative to chemotherapy dose delays and/or dose reductions. Only one patient was dose delayed (in addition to receiving pegfilgrastim), while the remaining seven patients dropped out. As a result, chemotherapy dose delays and reductions overall occurred more frequently for patients receiving pegfilgrastim from cycle 1 (Fig. 4B, Table 3), who, when faced with chemotherapy side effects, had no option but to dose reduce and/or delay because they were already receiving growth factor support.
Solid tumor patients receiving pegfilgrastim from the first cycle experienced fewer chemotherapy dose delays and dose reductions than those who received pegfilgrastim per physician discretion (Fig. 4A, Table 3). Rates of antibiotic use associated with neutropenia-related events were lower for solid tumor patients receiving pegfilgrastim from the first cycle. For NHL patients, antibiotic use associated with neutropenia-related events was approximately equivalent for both treatment arms (Fig. 4B).
There was no significant difference in the benefit of pegfilgrastim for patients aged
Safety
This is the largest, randomized, prospective, community-based trial of typical elderly cancer patients. Several important clinical conclusions can be drawn from this study. We demonstrated that pegfilgrastim use from the first cycle reduced the incidence of febrile neutropenia, grade 3 or 4 neutropenia, febrile neutropenia-related hospitalizations, and antibiotic use in this population with solid tumors or NHL receiving an array of mild to moderately myelosuppressive chemotherapy regimens.
For solid tumor patients, we powered this study to demonstrate a 40% difference in the incidence of febrile neutropenia assuming a baseline rate of 20%. However, because we observed a greater than expected use of colony-stimulating factors in the physician-discretion arm, it is likely that this assumed baseline rate of febrile neutropenia was much lower than 20%. Even at an assumed baseline rate of 10%, we observed a Patients with solid tumors who received pegfilgrastim from the first cycle had fewer chemotherapy dose delays and reductions than patients in the physician-discretion arm. This is an important conclusion as the receipt of a fulldose on schedule may lead to better patient outcomes [8]. For NHL patients, chemotherapy dose delays and reductions appeared to be higher in the pegfilgrastim all-cycles arm. However, this finding may reflect the high use of pegfilgrastim in the physician-discretion arm from the second cycle, as an alternative to chemotherapy dose delays and reductions. Had there been no option to administer pegfilgrastim from the second cycle, dose reductions and delays would have, in all likelihood, been higher in the physician-discretion arm than in the pegfilgrastim all-cycles arm. The high use of pegfilgrastim in cycle 1 and the large dropout rate noted in cycle 1 may have also confounded the data.
As the population ages, the elderly will represent an increasing proportion of those treated for cancer. As such, it is important that these patients be adequately represented in clinical trials [28, 29]. The results of this study begin to address this and provide much-needed evidence that standard-dose mild to moderately myelosuppressive chemotherapy can be safely administered to elderly cancer patients when pegfilgrastim is used from the first cycle to reduce the incidence of febrile neutropenia and related events.
We thank the research nurses and the patients who participated in this study. Sadie Whittaker assisted with the writing of the article.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||