Introduction

Febrile neutropenia is a common but serious consequence of myelosuppressive cytotoxic chemotherapy, which can delay or reduce doses of chemotherapy regimens and thereby compromise the anticancer treatment [1–3]. Febrile neutropenia can lead to potentially fatal infections that require antibiotic treatment and usually necessitate hospitalization [4], with a mortality rate of approximately 8% in the inpatient setting [5].

Granulocyte-colony stimulating factor (G-CSF) is a cytokine that acts upon hematopoietic progenitor cells to stimulate the formation and function of mature neutrophils [6]. G-CSF reduces the incidence and duration of febrile neutropenia, thereby decreasing the need to delay the chemotherapy cycles or to reduce the doses [7–9]. US [10] and European [11] guidelines recommend prophylactic G-CSF for chemotherapy regimens associated with a high risk of febrile neutropenia (≥ 20%) and consideration in patients at intermediate risk (10–20%). However, a number of studies have shown that G-CSF is underused in patients undergoing chemotherapy treatments associated with a high risk of febrile neutropenia [12].

Filgrastim (G-CSF) was first approved in 1991 in Europe [13] and US [14], under the trade name Neupogen^® (Amgen). Additionally, a pegylated recombinant human filgrastim with a longer half-life, pegfilgrastim (Neulasta^®, Amgen) [15], has been developed [16]. Another recombinant human G-CSF, lenograstim (Granocyte^®, Chugai Pharma), is also available [17]. In addition, several biosimilars of filgrastim have been developed [18]. One of them is Nivestim^® (Hospira Inc) which was subsequently approved for use in Europe in 2010. Nivestim demonstrated equivalent efficacy and a similar safety profile as the reference product Neupogen^® in a phase-III double-blind clinical trial that included patients with breast cancer undergoing myelosuppressive chemotherapy [19].

The aim of the present article is to describe the results of the non-interventional NEXT study which assessed the safety and efficacy of the biosimilar filgrastim Nivestim in real-life in patients undergoing chemotherapy for solid tumor or hematological malignancies.

Methods:

Study design

The NEXT (“Tolérance de Nivestim chez les patiEnts traités par une chimiothérapie anticancéreuse cytotoXique en praTique courante”) study was a prospective, non-interventional, longitudinal, national, multicentre study conducted across 160 sites in France by oncologists and hematologists. The study design has been previously published [20]. This study has the ClinicalTrials. gov identifier NCT01574235.

The patients were informed both orally and in writing on the objectives of the study. This study was conducted according to the current revision of the 1964 Helsinki declaration and with the French laws and regulations.

The primary objective was to evaluate the safety of Nivestim, administered either as prophylactic or curative treatment. The secondary objectives included information on patient characteristics, efficacy of Nivestim therapy, use of Nivestim, physician knowledge regarding filgrastim prescription and reasons for choosing Nivestim.

Patient data were recorded for 1–6 chemotherapy cycles following study inclusion, with three study visits; at inclusion (Visit 1), at one-month follow-up (Visit 2) and at the end of the last chemotherapy cycle (Visit 3).

Patient demographics, clinical characteristics and Nivestim treatment-related data on efficacy and safety, such as febrile neutropenia and adverse events, were recorded on case report forms. Physicians who took part in the survey completed a questionnaire concerning the treatment and prevention of chemotherapy-induced febrile neutropenia.

Patients

Patients who met the following criteria were included: patients with an age ≥ 18 years presenting with a solid tumor or a hematological malignancy; on-going or starting treatment with neutropenia-inducing chemotherapy (regardless of the cycle); treatment with Nivestim instituted for the purpose of reducing the duration of neutropenia and the incidence of chemotherapyinduced febrile neutropenia. Patients were not included if they presented with a chronic myeloproliferative syndrome or a myelodysplastic syndrome, had hypersensitivity to any of the ingredients of Nivestim or were not receiving chemotherapy.

Patients were classified for the risk of febrile neutropenia according to the European Organization for Research and Treatment of Cancer (EORTC) recommendations [11]: chemotherapy regimen with a low risk (<10%), medium risk (10–20%) or high risk (>20%) of febrile neutropenia.

Statistical methodology

The primary endpoint was the description of all treatmentemergent adverse events. The secondary endpoints were the occurrence of febrile neutropenia and infections, their impact on chemotherapy treatment, the hospitalizations due to any febrile neutropenia or infection, the characteristics of patients treated with Nivestim in real-life practice, the methods of treatment with Nivestim (curative or prophylactic) in routine practice and the assessment of the general practice of study physicians for G-CSF prescription.

The percentages of prophylactic patients who experienced febrile neutropenia and/or infection were calculated according to the modified Wald method.

Analyses were performed with SAS software version 9.2 (SAS Institute; Cary, NC, USA). Missing data were not replaced for the analyses.

Results

Study population

A total of 2114 patients were enrolled and 12 patients were excluded due to protocol deviations; 2102 patients were included in the analysis population (Figure 1). During the study, 516 patients prematurely withdrew from the study: 86 patients prior to V2, 297 patients prior to V3 and 133 patients at V3. The most frequent reasons of patient withdrawals were cessation of chemotherapy (52.2%), switch from Nivestim to another G-CSF (23.5%) and patient death (16.5%) (Table 1). In patients with premature withdrawal due to chemotherapy discontinuation, recurrent febrile neutropenia was reported for 4 patients (1.5%) and persistent febrile neutropenia for 3 patients (1.1%).

Figure 1: Study population.

*12 patients were excluded from the analysis due to no information, missing age, patient with a disorder outside of the protocol, patient already included and no biosimilar filgrastim prescription upon inclusion.

Table 1: Reasons for premature withdrawal from the observational NEXT study.

Results are given as n (%) G-CSF: granulocyte-colony stimulating factor. * Premature withdrawal could be due to more than one reason.

Patient demographics and clinical characteristics

Patient characteristics are summarized in (Table 2). Their mean age was 63.5 years and 50.2% were male. A total of 75.1% (n = 1579) of patients had a solid tumor and the most common were gynecological (37.2%), digestive (26.2%) and lung tumors (18.8%). Patients with hematological malignancies were 25.3% (n = 532), the most frequent being lymphoma (19.5%; n=410).

In the entire population, 34.4% of patients had received prior chemotherapy, 19.6% prior radiotherapy and 20.2% prior G-CSF therapy. In patients with available data, first-line chemotherapy was predominantly administered: 74.8% (n = 1082) of patients with solid tumors and 70.7% (n = 378) of patients with hematological malignancies.

The indication for Nivestim was mainly for prophylaxis (98.2%, n = 2065) rather than for curative treatment (1.8%, n = 37).

In the prophylactic population, Nivestim was prescribed as primary prophylaxis in 92.6% (n = 1912) of patients and as secondary prophylaxis in 7.4% (n = 153). The 2065 patients with prophylactic Nivestim received chemotherapy regimens with a high risk of febrile neutropenia for 19.4% (n=975), intermediate risk for 55.6% (n=752) and low risk for 25.0% (n=338). Among patients receiving chemotherapy regimens with an intermediary or low risk of febrile neutropenia, 95.4% and 96.5% had at least one patient risk factor for febrile neutropenia, respectively.

Among patients who received curative treatment, 32.4% (n = 12) had grade 3 neutropenia (500–1000 neutrophils/mm³) and 29.7% (n = 11) grade 4 neutropenia (<500 neutrophils/mm³). Five patients who were prescribed curative treatment presented with an infection at the inclusion visit (three cases were bacterial and one viral; missing data for one patient).

Table 2: Patient characteristics at inclusion in NEXT study.

* Some patients had missing data Patients could present with more than one type of malignancy

BMI, body mass index; CLL, chronic lymphoid leukemia; ECOG, Eastern Cooperative Oncology Group; ENT, ear, nose or throat.

Safety

During the study, treatment-emergent adverse events were reported in 20.4% (414/2034) of patients. Muscle and/or bone pain was the most commonly reported adverse event (12.1%), followed by nausea (3.0%). Allergic reactions to Nivestim occurred in 0.4% of patients (Table 3).

The percentages of patients who experienced adverse events were comparable in patients with solid tumors (21.1%, n = 321) or with hematological malignancies (18.5%, n = 97). Regarding adverse events of particular interest for G-CSF, one patient presented with cutaneous vasculitis, two patients with hypersensitivity and two patients with inefficacy during the course of the study (all received Nivestim as prophylactic treatment).

Table 3: Patients with at least one treatment-emergent adverse events experienced by >1% of patients during the study.

Prophylaxis with Nivestim

Treatment administration

In patients who received Nivestim as prophylactic treatment, the median time to initiation after onset of chemotherapy was 2 days (Table 4). The dose of Nivestim was 30 MIU for 79.9% (n = 1646) of patients and was administered subcutaneously in 99.4% (n = 2049) of patients; the mean (SD) treatment duration was 6.0 (3.8) days.

Anti-infective prophylaxis was also prescribed to 47.9% (n = 251) of patients with a hematological malignancy and 2.6% (n = 41) of patients with solid tumor. Among patients who were administered at least one treatment for anti-infective prophylaxis, 93.6% (n = 235) and 58.5% (n = 24) with hematological malignancies and solid tumors received an antibiotic, 9.2% (n = 23) and 43.9% (n = 18) received an antifungal, 94.8% (n = 238) and 14.6% (n = 6) received an antiviral and 1.6% (n = 4) and 7.3% (n = 3) received other anti-infective agents, respectively.

Febrile neutropenia and infections. Febrile neutropenia was reported in 98 patients (4.9%; 95% CI, 4.06–5.98) in the prophylactic group, occurring at a median time of 14.0 days after chemotherapy cycle. Among patients who were chemotherapy-naïve at study inclusion, febrile neutropenia occurred in 4.3% (n = 42) and 6.4% (n = 62) after the first chemotherapy cycle and throughout the study, respectively (Figure 2). Overall, prophylactic patients with hematological malignancies experienced more frequently febrile neutropenia (7.5%, n = 38) compared with patients with solid tumors (4.1%; n = 60).

Infections were reported in 61 patients (3.1%; 95% CI, 2.39–3.93) on prophylaxis during the study and occurred after a median time of 23.5 days following the first chemotherapy cycle. Among prophylactic patients who had been chemotherapy-naïve, 2.2% (n = 21) and 4.1% (n = 40) experienced infections after their first chemotherapy cycle and throughout the rest of the study, respectively (Figure 3).

Table 4: Characteristics of Nivestim treatment during the study.

Figure 2: Incidence of febrile neutropenia in patients receiving prophylactic treatment with Nivestim*. Overall, 4,9% (7,2% primary prophylaxis, 4,7% secondary prophylaxis) of patients presented with FN during study (CLL: chonic lymphocytic leukemia; FN: febrile neutropenia).

Figure 3: Efficacy of prophylactic Nivestim* at the first CT cycle. Mean duration of hospitalization for FN and/or infection after the first cycle was 8,7+/-10,9 days. (CT: chemotherapy; FN febrile neutropenia)