Identifying Patients at Low Risk for FN Complications: Development and Validation of the MASCC Risk Index Score [Unpublished]
2015 Febrile neutropenia (FN) is a life-threatening complication of cancer chemotherapy and is considered a medical emergency. The classic treatment strategy is broad-spectrum antibiotics administered intravenously in a hospital setting, and most patients respond promptly with no complications.
Several instruments have been developed to identify patients at high risk for FN complications. One was the tool developed and validated by Talcott and colleagues (Talcott et al., 1988). However, it depended on patients’ hospitalization status, a measure that varies from country to country or institution to institution and can change with the development of new anticancer treatments. Kern et al. (1999) and Freifeld et al (1999) proposed other methods, but neither gained consensus.
Klastersky et al. conducted a survey between 1994 and 1997 to identify independent predictive factors that could be assessed at fever onset in adult febrile neutropenic cancer patients being treated with chemotherapy. The research team also aimed to develop an internationally validated scoring system. The identified factors were combined to produce a score indicating the probability of resolution of FN without serious medical complications. Secondary objectives of this research included assessing response to empiric therapy, infection documentation, bacteremia development, and fever duration.
This effort resulted in the MASCC Risk Index, a scoring system for identifying low-risk cancer patients with febrile neutropenia. Further research, published by Klastersky et al. in the Journal of Clinical Oncology (2000), showed that the Risk Index could accurately identify patients at low risk for complications and that it could be used to select patients for testing more convenient or cost-effective therapies. The MASCC Risk-Index Score comprises the following factors.
1 Burden of febrile neutropenia refers to general clinical status as influenced by the febrile neutropenic episode. It is evaluated in accordance with the following scale: no symptoms (5), mild symptoms (5), moderate symptoms (3), severe symptoms (0), moribund (0).
4 The points attributed to the variable "burden of febrile neutropenia" are not cumulative. Thus, the maximum theoretical score is therefore 26. A score of ≥ 21 is considered low risk and a score of < 21 as high risk (positive predictive value of 91%, specificity of 68%, and sensitivity of 71%).
This tool was endorsed in 2002 by the Infectious Diseases Society of America and has also been adopted by the European Society of Medical Oncology. This work also led to the development and publication of updated methodology guidelines for clinical trials involving patients with cancer and FN. The guidelines, jointly issued by MASCC and the Immunocompromised Host Society, suggest that response to empirical antibiotic therapy be determined after 72 hours and again on day 5, and the reasons for modification stated. The guidelines further specified that patients enrolled in outpatient studies be selected by use of a validated risk model and that they be carefully monitored after discharge from the hospital. Recommendations are included for blinding and stratification, as well as statistical consideration of trials specifically designed for showing equivalence, recording of response, safety parameters, and readmission rates. The guidelines were intended to simplify and improve comparisons between studies with the ultimate goal of improving therapy.
The publication and endorsement of the MASCC Risk Index was followed by numerous external validations. These are discussed by Klastersky and Paesmans (2013) in their review of 10 years of research using the MASCC index. The authors also discuss antimicrobial empiric therapy for FN, early hospital discharge of low-risk FN patients, and prediction and management of non-low-risk patients. Alternatives to the MASCC scoring system have been proposed — for example methods combining clinical and biological parameters — but these have not been validated in a prospective study or compared with the MASCC score. One proposed model (Eun Ha et al, 2011) would incorporate prediction of bacteremia, but such prediction remains uncertain at this time. Another recent model for predicting risk of neutropenic complications (Lyman et al., 2011) includes clinical and biological parameters as well as chemotherapy characteristics. Klastersky et al. point out that this would be difficult to use in many clinical settings, thus limiting its acceptance.
The MASCC Risk Index has been shown to be a reliable tool for identifying patients at low risk for complications of FN. As such, it is a valuable part of the selection of patients who can safely be treated at home. Specificity of prediction could still be improved, especially for patients with hematological tumors. There have been attempts to improve the MASCC score by incorporating biological variables (e.g., measurement of various interleukins TNF, procalcitonin, and CRP) or by incorporating a prediction of bacteremia. But these do not appear to improve predictive accuracy of the score. Patient characteristics, tumor characteristics, anticancer therapies, and management of FN continue to change over time, so it is important to continue monitoring predictive accuracy and searching for ways to improve it. Klastersky et al. point out that a prediction rule for high-risk patients is still needed and should be developed.
References
