A retrospective study has revealed new potential risk factors for chemotherapy-induced febrile neutropenia (FN) in patients with solid tumors and non-Hodgkin lymphoma (NHL).
Researchers found the timing and duration of corticosteroid use were both associated with FN.
The team also observed “marginal” associations between FN and certain dermatologic and mucosal conditions as well as the use of intravenous (IV) antibiotics before chemotherapy.
On the other hand, there was no association between oral antibiotic use and FN or between radiation therapy (RT) and FN.
Chun Rebecca Chao, PhD, of Kaiser Permanente Southern California in Pasadena, and her colleagues reported these findings in JNCCN.
“Febrile neutropenia is life-threatening and often requires hospitalization,” Dr. Chao noted. “Furthermore, FN can lead to chemotherapy dose delay and dose reduction, which, in turn, negatively impacts antitumor efficacy. However, it can be prevented if high-risk individuals are identified and treated prophylactically.”
With this in mind, Dr. Chao and her colleagues set out to identify novel risk factors for FN by analyzing 15,971 patients who were treated with myelosuppressive chemotherapy at Kaiser Permanente Southern California between 2000 and 2009.
Patients had been diagnosed with NHL (n=1,617) or breast (n=6,323), lung (n=3,584), colorectal (n=3,062), ovarian (n=924), or gastric (n=461) cancers.
In all, 4.3% of patients developed FN during their first cycle of chemotherapy.
The researchers found corticosteroid use was associated with an increased risk of FN in a propensity score-adjusted (PSA) model (adjusted for age, sex, socioeconomic factors, comorbidities, etc.). The hazard ratio (HR) was 1.53 (95% CI, 1.17-1.98; P<0.01) for patients who received corticosteroids.
A longer duration of corticosteroid use was associated with a greater risk of FN. The adjusted HR (compared to no corticosteroid use) was:
- 1.78 for corticosteroid treatment lasting less than 15 days (P<0.01)
- 1.84 for treatment lasting 15 to 29 days (P<0.01)
- 2.27 for treatment lasting 30 to 44 days (P<0.01)
- 2.86 for treatment lasting 45 to 90 days (P<0.01).
More recent corticosteroid use was associated with a greater risk of FN as well. The adjusted HR was:
- 1.88 for corticosteroid treatment less than 15 days before chemotherapy (P<0.01)
- 1.13 for treatment 15 to 29 days before chemotherapy (P=0.72)
- 1.22 for treatment 30 to 44 days before chemotherapy (P=0.66)
- 1.41 for treatment 45 to 90 days before chemotherapy (P=0.32).
“One way to reduce the incidence rate for FN could be to schedule prior corticosteroid use and subsequent chemotherapy with at least 2 weeks between them, given the magnitude of the risk increase and prevalence of this risk factor,” Dr. Chao said.
Other potential risk factors
The researchers found a “marginally” increased risk of FN in patients with certain dermatologic conditions (dermatitis, psoriasis, pruritus, etc.) and mucosal conditions (gastritis, stomatitis, mucositis, etc.).
In the PSA model, the HR was 1.40 (95% CI, 0.98-1.93; P=0.05) for patients with these conditions.
IV antibiotic use was also found to be marginally associated with an increased risk of FN in a restricted analysis covering patients treated in 2008 and 2009. In the PSA model, the HR was 1.35 (95% CI, 0.97-1.87; P=0.08).
On the other hand, there was no association between FN and oral antibiotic use in the restricted analysis. In the PSA model, the HR was 1.07 (95% CI, 0.77-1.48; P=0.70) for patients who received oral antibiotics.
Dr. Chao and her colleagues said these results suggest IV antibiotics may have a more profound impact than oral antibiotics on the balance of bacterial flora and other immune functions. Another possible explanation is that patients who received IV antibiotics were generally sicker and more prone to severe infection than patients who received oral antibiotics.
As with oral antibiotics, the researchers found no association between FN and the following factors (with the PSA model):
- Prior surgery (HR=0.89; 95% CI, 0.72-1.11; P=0.30)
- Prior RT (HR=0.91; 95% CI, 0.64-1.27; P=0.61)
- Concurrent RT (HR=1.32; 95% CI, 0.69-2.37; P=0.37).
The researchers noted that they did not account for radiation field or dose in this study, so additional evaluation of RT as a risk factor is needed.
In closing, Dr. Chao and her colleagues said these results suggest corticosteroid use, IV antibiotics, and certain dermatologic and mucosal conditions should be taken into consideration when monitoring patients receiving myelosuppressive chemotherapy and when evaluating the need for prophylactic granulocyte colony-stimulating factor or chemotherapy dose reduction.
Dr. Chao and her colleagues received funding from Amgen, Inc., to perform this study.