Aims and objectives

The Prospective Observational European Neutropenia Study is designed to better define the relationship between risk factors and neutropenia. The main study objectives are:

 

• to estimate the incidence of grade 3/4 neutropenia following common myelosuppressive chemotherapy regimens;
• to assess the frequency and severity of FN and of neutropenia-induced chemotherapy dose delays and dose reductions;
• to identify associations between neutropenia risk factors (e.g. treatment characteristics, comorbidities) and neutropenic event occurrence, and between neutropenic event occurrence and impaired chemotherapy delivery;
• to contribute to the development of a clinically effective risk model, which will identify patients who are at an increased risk of experiencing neutropenia, in order to target prophylactic measures.

Pettengell R and Schwenkglenks M Leuk Lymphoma 2011; 52(6):1133-6

http://www.ncbi.nlm.nih.gov/pubmed/21314247

 

This subgroup analysis of non-Hodgkin lymphoma (NHL) patients enrolled in the INC-EU Prospective Observational European Neutropenia Study aimed to investigate the impact of age on the frequency of neutropenic events, chemotherapy delivery, and colony-stimulating factor (CSF) use. Patients with NHL were divided into three age groups: ‹60 years, 61-10 years, and ›70 years. Grade 4 chemotherapy-induced neutropenia (CIN) and febrile neutropenia (FN) occurred frequently in all age groups, but only the frequency of grade 4 CIN tended to increase with age. The proportion of patients with dose delays of more than four days increased with age whereas dose reductions were comparable between age groups. Discontinuation of planned chemotherapy treatment was more likely in the oldest age group (›70 years). The proportion of patients receiving primary CSF prophylaxis or reactive CSF was similar in all age groups, despite the fact that the European Organisation for Research and Treatment of Cancer (EORTC) guidelines recommended prophylactic CSF use for all patients ›65 years receiving chemotherapy. Patients ›70 years are at higher risk of a adverse outcome following a neutropenic event; therefore CSF prophylaxis may be particularly relevant for this population.

Schwenkglenks M, Pettengell R, Jackisch C, et al. Support Care Cancer 2011; 19(4):483-90

http://www.ncbi.nlm.nih.gov/pubmed/20306092

 

Risk factors of grade 4 chemotherapy-induced neutropenia (CIN), either during the first or in any cycle of adjuvant chemotherapy, have been evaluated based on data from 444 breast cancer patients enrolled into the INC-EU Prospective Observational European Neutropenia Study. Using multivariate logistic regression analysis, risk factors for grade 4 CIN were identified with the ultimate aim to better target colony-stimulating factor (CSF) prophylaxis and to prevent low relative dose intensity (RDI) in breast cancer patients.

 

Results

Most patients (70%) received anthracycline-based treatments; 20% received anthracycline-containing and taxane-containing sequential regimens, 4% received a combination of docetaxel, doxorubicin, and cyclophosphamide (TAC), and 2% received other taxane-containing regimens. Grade 4 CIN occurred in approximately one of three patients (34%), primary CSF prophylaxis was given to 9% of the patients, and secondayr CSF prophylaxis was provided to 24%. Patients who experienced grade 4 CIN were more likely to only achieve a reduced RDI (‹85%).

 

Predictors of CIN in any cycle

• older age
• lower weight
• increasing planned dose intensity of doxorubicin, epirubicin, or docetaxel
• higher number of planned chemotherapy cycles
• vascular comorbidity
• lower baseline white blood cell (WBC) count
• higher baseline bilirubin

 

Predictors of CIN in cycle 1

• lower weight
• higher planned dose intensity of doxorubicin, epirubicin, or docetaxel
• higher number of planned chemotherapy cycles
• vascular comorbidity

 

CSF use, dose reductions, and dose delays, before a neutropenic event occurred, protected against grade 4 CIN.

 

Test characteristics

The area under the receiver operating characteristic (ROC) curve was 0.82, sensitivity was 73%, specificity was 74%, and negative predictive value (NPV) was 84% in the any cycle model. Under 10-fold cross-validation conditions, predictive ability of the any cycle model was slightly decreased (area under the ROC curve, 0.78). The area under the ROC curve for the cycle 1 model was 0.73 and NPV was 86%.

 

Discussion and Conclusion

The model showed good predictive ability and performed well in 10-fold cross-validation, but remains to be validated further in independent datasets. It identified or confirmed several risk factors of potential clinical relevance. High baseline bilirubin level is indicative of impaired liver function and may be predictive of grade 4 CIN in patients receiving chemotherapy with doxorubicin, epirubicin, or docetaxel, because these agents are detoxified by the liver.

Pettengell R, Bosly A, Szucs TD, et al. Br J Haematol 2009; 144(5):677-85

http://www.ncbi.nlm.nih.gov/pubmed/19055662

 

A multivariate risk model of febrile neutropenia (FN) occurrence in the first cycle of chemotherapy was developed, based on a subgroup analysis of 240 non-Hodgkin lymphoma (NHL) patients from the INC-EU Prospective Observational European Neutropenia Study. To decrease the incidence of myelosuppression and enable full-dose-on-time chemotherapy delivery, such models intend to identify patients at high risk of FN and who may particularly profit from prophylactic colony-stimulating factor (CSF) treatment, and/or patients at low risk of FN who may not require prophylactic CSF treatment. 

 

Results

Three out of four patients (75%) received combination chemotherapy with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) and 82% of patients received rituximab. The incidence of FN across all cycles of chemotherapy was 22% and 9% of patients had a FN event during the first cycle. Overall, 28% of patients received primary CSF prophylaxis and 29% had other CSF use.

FN risk factors in the first cycle of chemotherapy

• older age
• low baseline albumin ‹35g/l

• lower glomerular filtration rate
• lower weight
• increasing planned cyclophosphamide or etoposide dose
• history of previous chemotherapy or recent infection

 

Additional risk factors associated with increased FN across all cycles of chemotherapy in NHL patients were:

• low baseline absolute neutrophil count (ANC) or white blood cell count (WBC)
• high baseline alkaline phosphatase
• increasing planned cytarabine dose
• cardiac comorbidity

 

Primary CSF prophylaxis and higher weight were protective factors against FN in the first cycle and across all cycles.

 

Test characteristics

At the optimal cut-point, the model of FN occurrence in cylce 1 had high sensitivity and specificity (81%/80%), the negative predictive value (NPV) was 98%, and the area under the receiver operating characteristic (ROC) curve was 0.86. Predictive ability was slightly lower under 10-fold cross-validation conditions (area under the ROC curve, 0.78).

The model of FN occurrence in any cycle had a sensitivity and specificity of 76%, the NPV was 92%, and the area under the ROC curve was 0.83. Its predictive ability was slightly lower under 10-fold cross-validation conditions (area under the ROC curve, 0.72).

 

Discussion and Conclusion

Other retrospective studies specific to NHL also reported increased risk for cycle 1 FN in patients with higher age and low baseline albumin, and a protective effect of CSF prophylaxis (Lyman & Delgado, 2003; Rabinowith et al, 2006; Teegala et al, 2007). Increasing planned chemotherapy dose was also predictive of FN in earlier published models (Voog et al, 2000; Shayne et al 2007). Favourable test characteristics indicate potential clinical utility but additional validation in independent datasets is clearly required.

Pettengell R, Schwenkglenks M, Leonard R, et al. Support Care Cancer 2008; 16:1299-1309

http://www.springerlink.com/content/l81q274364424uq9/

 

The first manuscript from the INC-EU Prospective Study reports data from breast cancer and lymphoma patients recruited from 66 practices in five European countries. Patients eligible for inclusion were starting a new myelosuppressive chemotherapy sequence with at least 4 cycles planned.

Data from 444 breast cancer patients were analysed, the majority of whom received anthracycline-based regimens. Primary CSF prophylaxis was provided to 9% of patients and the rate of FN was generally low at 6%, although grade 4 CIN occurred in 34% of patients. Over 20% of patients did not receive chemotherapy as planned, experiencing reductions and delays in their chemotherapy (Figure 1). Dose reductions and low RDI were more common in patients who experienced grade 4 neutropenia or FN. Of 305 lymphoma patients, 240 had a diagnosis of NHL and 65 had HL. The majority of NHL patients (74%) were treated with CHOP-21-like regimens and primary CSF prophylaxis was provided to 12% of these patients. The rate of FN with CHOP- 21-like regimens was 22%, with over 50% of patients experiencing grade 4 CIN. One third of NHL patients undergoing CHOP-21- like chemotherapy did not receive treatment as planned (RDI ≤ 85%). Fifteen percent of HL patients experienced FN, 40% developed grade 4 CIN, and 30% received RDI ≤ 85%. Across both breast cancer and lymphoma patients, the highest rates of CIN and FN were seen in the first cycle of chemotherapy.

Increased awareness of factors that are associated with chemotherapy dose  limitations can assist clinicians in identifying at-risk patients and facilitate patient management. In a multivariate analysis, age ≥ 65 years, ECOG > 1 and previous FN were identified as risk factors for RDI ≤ 85%. In lymphoma patients, primary CSF prophylaxis was associated with a decreased risk of low RDI.

The INC-EU has a good working relationship with their US counterpart, the Awareness of Neutropenia in Chemotherapy Study Group (ANC). Professor Gary Lyman and other members of the ANC provided advice on the initial set-up of the Prospective Study. Comparison of results from the INC-EU and the ANC reveals that FN and CIN not only remain frequent in European practice, but also in the US. In a prospective nationwide study¹, rates of FN in 1,138 breast cancer patients were slightly higher than in the European study, at 15% (Figure 2); although a broadly similar number of patients (42%) experienced neutropenia at some point during their treatment. Among 257 NHL and 55 HL patients in the US study, trends in FN and CIN rates generally mirrored those seen in the INC-EU study, but were slightly lower.

*The US study group define FN as fever/infection + ANC < 1.0 x 10⁹/L, the INC-EU define FN as temperature ≥ 38° C + ANC < 0.5 x 10⁹/L

 

1. Crawford J, Dale DC, et al.: Risk and Timing of Neutropenic Events in Adult Cancer Patients Receiving Chemotherapy: The Results of a Prospective Nationwide Study of Oncology Practice. J Natl Compr Canc Netw 2008; 6(2):109-118