Oncol Ther (2020) 8:311–324 https://doi.org/10.1007/s40487-020-00115-3 BRIEF REPORT Mobilization of Hematopoietic Stem Cells into Peripheral Blood for Autologous Transplantation Seems Less Efficacious in Poor Mobilizers with the Use of a Biosimilar of Filgrastim and Plerixafor: A Retrospective Comparative Analysis Rocı́o Parody . Isabel Sánchez-Ortega . Christelle Ferrá . Ramon Guardia . Carme Talarn . Maite Encuentra . Eduard Fort . David López . Mireia Morgades . Eva Alonso . Sandra Ortega . Josep Sarrá . David Gallardo . Josep M. Ribera . Anna Sureda Received: March 6, 2020 / Published online: May 14, 2020 Ó The Author(s) 2020 ABSTRACT similar efficacy to originator filgrastim (Neu- pogenÒ [NEU]; Amgen Inc.) as prophylaxis in Introduction: Biosimilars of granulocyte col- neutropenia and in the mobilization of stem ony-stimulating factors (G-CSF) have shown cells in patients receiving combination chemotherapy with G-CSF. Digital Features To view digital features for this article Methods: This was a retrospective study in go to https://doi.org/10.6084/m9.figshare.12162039. which the characteristics of stem cell mobiliza- tion treated with a G-CSF alone were compared R. Parody (&)  I. Sánchez-Ortega  M. Encuentra  in 216 patients and 56 donors. The two G-CSF A. Sureda compared were NEU and the biosimilar filgras- Clinical Hematology Department, Institut Català tim ZarzioÒ (Sandoz GmbH) (referred to here- d’Oncologia, L’Hospitalet, Institut d’Investigacio Biomedica de Bellvitge (IDIBELL), Barcelona, Spain after as BIO). Primary objectives were e-mail: rparody@iconcologia.net mobilization rate (minimum of 10 9 103/ml CD34? on day 4 of treatment [day ?4]) and use C. Ferrá  M. Morgades  J. M. Ribera of the immunostimulant plerixafor (PLEX) in Hematology Department, Institut Català d’Oncologia-Badalona, Hospital Universitari each group. Germans Trias I Pujol, Josep Carreras Leukemia Results: The general characteristics of the Research Institute,, Universitat Autònoma de patients receiving NEU (n = 138) and those Barcelona, Barcelona, Spain receiving BIO (n = 78) did not differ signifi- R. Guardia  D. Gallardo cantly. PLEX was used in 24% of BIO patients Hospital Universitari Josep Trueta, Girona, Spain and in 25.7% of NEU patients. The median C. Talarn  J. Sarrá CD34? cell count on day ?4 was significantly Hospital Universitari Joan XXIII, Tarragona, Spain lower in BIO patients who needed PLEX than in those who did not (2.4 vs. 4.8 9 103/ml; p = M. Encuentra  M. Morgades 0.002), as was the final CD34? cell count (2.5 Institut Català d’Oncologia, L’Hospitalet, IDIBELL, Barcelona, Spain vs. 3.3 9 106/kg; p 0.03). Mobilization failure rate was higher in the BIO group than in the E. Fort  D. López NEU group (20 vs. 0%; p = 0.01). With respect to Pharmacy Department, Institut Català d’Oncologia, L’Hospitalet, IDIBELL, Barcelona, Spain donors, more than one apheresis was needed in three BIO donors, one of them with PLEX. The E. Alonso  S. Ortega use of BIO was the only risk factor for mobi- Banc de Sang I Teixits de Catalunya, Barcelona, lization failure in patients who needed PLEX Spain 312 Oncol Ther (2020) 8:311–324 (hazard ratio 10.3; 95% confidence interval neutropenia and the mobilization of peripheral 1.3–77.8). blood (PB) stem cells [1, 2]. The main advantage Conclusion: The study revealed that BIO had a of biosimilars over the originator G-CSF is the lower efficacy for stem cell mobilization when reduction in final costs, up to 80%, without a the only treatment was G-CSF, especially in theoretical loss of efficacy, while preserving the poor mobilizers needing PLEX. posology and administration route [3–5]. Nev- ertheless, there is still an ongoing debate on the Keywords: Biosimilars; Plerixafor; Stem cell quality, efficacy, and safety of biosimilar G-CSF. mobilization Initial phase I–II prospective clinical trials and more recently published phase III ones indicate that both the originator G-CSF and biosimilar Key Summary Points G-CSF do demonstrate equivalence in terms of efficacy and safety profile when being used to Biosimilars for filgrastim have become reduce chemotherapy-induced neutropenia and widely used in place of originator infectious-related complications in patients filgrastim (NeupogenÒ) following study with oncohematological malignancies [6–10]. results showing no significant differences Biosimilars have also been used as a mobi- in efficacy between both formulations. lizing agent of hematopoietic stem cells (HSC) No data have been published to date in PB in the setting of autologous stem cell comparing the efficacy of biosimilar transplantation (auto-hematocrit [HCT]), and filgrastim with originator filgrastim for studies have examined equivalence between stem cell mobilization in the setting of biosimilars and the original filgrastim (Neu- mobilization with granulocyte colony- pogenÒ [NEU]; Amgen Inc.) in patients and stimulating factors (G-CSF) only. healthy donors [11–16]. Although comparative studies with a historical control of NEU did not We have respectively reviewed a series of show significant statistical differences with 216 consecutive patients and 56 donors, respect to peak CD34? cells in PB on day 4 of analyzed separately, who were treated treatment, the total number of CD34? cells in with only a G-CSF and compared the the final apheresis product, and the median biosimilar ZarzioÒ and originator number of apheresis per patient, most of these Neupogen in terms of efficacy of studies were non-randomized and focused on mobilization rate. chemotherapy-based mobilization strategies Our results show that biosimilar Zarzio and not on mobilization with G-CSF alone might be less efficacious than originator [17–19]. filgrastim when used with plerixafor in Plerixafor (PLEX; MozobilÒ), an antagonist patients who are poor mobilizers. of CXCR4 chemokine receptor, is indicated in combination with G-CSF to mobilize HSC into PB in patients with non-Hodgkin’s lymphoma (NHL) or multiple myeloma (MM) (USA) and in patients with lymphoma or MM (EU) who are poor mobilizers (insufficient CD34? cells after INTRODUCTION apheresis, or low PB CD34? cell counts pre- cluding apheresis) [20–23]. Two randomized, Biosimilars of the granulocyte colony-stimulat- double-blind, multicenter trials demonstrated ing factor (G-CSF) filgrastim were approved by that PLEX used in combination with a G-CSF the European Medicines Agency (EMA) in 2008 mobilized HSC more efficiently than placebo and by the US Food and Drug Administration used with a G-CSF in adult patients with NHL or (FDA) in 2015 as substitutes for treating regis- MM [24, 25]. tered indications of the originator G-CSF, Our institute, the Catalan Institute of including the prevention and treatment of Oncology (ICO), switched in May 2016 from Oncol Ther (2020) 8:311–324 313 using the originator filgrastim NEU to using Procedure biosimilar filgrastim ZarzioÒ (Sandoz GmbH), not only as primary/secondary prophylaxis of The G-CSF dose was 10 lg/kg (for patients) and chemotherapy-induced neutropenia but also as 5 lg/kg (for healthy donors, unless significant a mobilizing agent of HSC for patients with MM weight differences) twice daily, for 5 consecu- or lymphoma candidates for auto-HCT, as well tive days. The minimum level of CD34? cells in as for healthy donors. We hypothesized the PB (monitored on day 4 of treatment [day ?4]) efficacy of G-CSF (NEU vs. Zarzio [hereafter to proceed to apheresis the next day was referred to as BIO]) might be different for 10 9 103/ml. If the minimum level was not mobilization procedures in the setting of no reached on day ?4, G-CSF was continued toge- prior chemotherapy and analyzed the results of ther with a single dose of PLEX 0.24 mg/kg a retrospective series of both patients and administered subcutaneously that evening, healthy donors in terms of PLEX use and col- after the G-CSF injection. On day ?5, if the lection results. circulating CD34? cells target was not reached, a second dose of PLEX would be administered; failure of this second dose to achieve the cir- METHODS culating CD34? cells target was interpreted as mobilization failure. Patients The apheresis procedure was generally per- formed through a central venous catheter in all The ICO is a comprehensive cancer center that patients, but peripheral veins were used in all acts as an umbrella organization for four dif- previous healthy donors, with a Spectra Optia ferent Clinical Hematology Departments each Apheresis System (Terumo Corporation, Tokyo, integrated in one of four different university Japan). The number of blood volumes per hospitals. These four departments share the patient to be processed followed standard rec- same general clinical procedures, pharmacy ommendations [23]. policies, blood bank, apheresis procedures, and HSC collection procedures, which are central- Definitions: Primary and Secondary ized in the Blood and Tissue Bank of Catalunya Objectives (BST). The study included consecutive adult patients (age C 18 years) with the underlying A poor mobilizer was defined as a patient who diagnosis of MM and lymphoma who were was unable to reach the minimum number of candidates for auto-HCT, and healthy adult CD34? cells to proceed to the collection proce- donors (age C 18 years), from December 2013 dure (10 9 103/ml). Mobilization failure was to November 2017. Only HSC mobilization defined as the inability to collect at least procedures performed with G-CSF as single 2 9 106 CD34? cells/kg body weight in the final agent were included in the study. product, regardless of the use of PLEX. The scientific proposal was reviewed and The primary objective of the study was the approved by the ethical committee of the comparison of the mobilization efficacy of NEU institution (ICO-IDIBELL), and all procedures versus BIO alone (not in combination with were in accordance with the ethical standards of chemotherapy), defined as a minimum number the 1964 Declaration of Helsinki and its later of CD34? cells in the PB on day ?4 (10 9 103/ amendments. All participants included in the ml) and/or the collection of at least 2 9 106 study provided written informed consent. CD34? cells/kg for patients and 4 9 106 CD34? cells/kg for healthy donors at the end of the apheresis procedure. Secondary objectives of the study were the use of PLEX following the indications of the recently published Catalano- Balear consensus [26] in both groups of patients 314 Oncol Ther (2020) 8:311–324 as well as the analysis of independent prog- RESULTS nostic factors predicting the use of PLEX and mobilization failure. Finally, a cost-effectiveness Patients analysis was conducted. A total of 216 patients diagnosed with lym- Statistical Analysis phoma (n = 102, of whom 88 had NHL) or MM candidates to be consolidated with auto-HCT Descriptive statistics were used to summarize (n = 114), from December 2013 to November the clinical data. Qualitative data were com- 2017, were included in the analysis. Of these pared using the Chi-square test. A non-para- patients, 138 received NEU and 78 received BIO. metric test (Mann–Whitney U test) was used to There was no statistically significant differences compare medians for quantitative data. in the general characteristics of the two patient A cost-effectiveness analysis (CEA) quantifies groups (Table 1). the relationship between intervention/product A total of 53 patients received PLEX, with no costs and the consequences of this, evaluated as significant difference in the number of patients specific pre-defined units. The relative value is between the BIO and NEU groups (25.7 vs. 24%, expressed as a ratio:total costs divided by total respectively; p = 0.7). Of these 53 patients, eight effectiveness and is called the median cost-ef- received PLEX outside of the established indi- fectiveness (MCE). In general, a procedure with cation, with a CD34? level [ 10 9 106/ml on a low MCE is more cost-effective (lower cost by day ?4 (range 10.7–13.5 9 106/ml) and were unit of benefit produced), whereas a higher subsequently excluded from the final analysis, MCE shows less efficiency in comparison to the which was based on the remaining 45 patients costs [27]. (Table 2). The median number of CD34? cells Patient costs were calculated taking into on day ?4 was significantly lower in patients in consideration the total number of doses of the BIO group receiving PLEX (n =15) than in G-CSF received per patient, PLEX if required, those in the NEU group receiving PLEX (n = 30) and each apheresis procedure (C 1). If a second (2.4 [range 0.7–9.8] vs. 4.8 [0.7–9.8] 9 103/ml; mobilization was needed (all cases with 1 dose p = 0.02), with a higher proportion of patients of cyclophosphamide 1 g/m2), the costs of this in the BIO group within the range \ 5 9 103/ml second mobilization procedure were also inclu- (80 vs. 56.7%; p = 0.1). Moreover, a significantly ded (chemotherapy, G-CSF dose, and apheresis). higher proportion of patients in the BIO group Both Pharmacy and BST Deparment reported compared to the NEU group needed more than corresponding prices as follows: NEU 300 mcg one dose of PLEX (26.7 vs. 3.3%; p = 0.02) and (€28,060; €93.5/mcg); NEU 480 mcg (€44,916); required more than one apheresis (28.6 vs. BIO Zarzio 300 mcg (€6396; €21.3/mcg), BIO 16.7%), although this latter difference was not 480 mcg (€10,233); PLEX single dose (24 mg; statistically significant. Regarding patients not €5473). Chemotherapy for a second mobiliza- needing PLEX (Table 3), it is noteworthy the tion with cyclophosphamide 1 g/m2 carried a median number of CD34? cells on day ?4 was cost of €5.37/g. Each apheresis procedure cost also significantly lower for patients in the BIO €1096.24. group (23.7 vs. 33.4; p = 0.03). Binary logistic regression was performed Mobilization failure was seen in five patients using SPSS v19 software (IBM Inc., Armonk, NY, in total (2.3%), all of whom were diagnosed USA). Variables included in multivariate analy- with NHL: three patients were in the BIO group sis were sex, age, weight, basal disease, pre-mo- and two were in the NEU group (3.8 vs. 1.4%; bilization status, prior treatment lines, and type p = 0.2). PLEX was unsuccessfully used in the of G-CSF. three BIO patients. With respect to the 45 patients who received PLEX, mobilization fail- ure occurred three patients in the BIO group versus no patient in the NEU group (20 vs. 0%; Oncol Ther (2020) 8:311–324 315 Table 1 General characteristics of patients Variablesa All patients (n = NEUb group (n = 138 BIOc group (n = 78 p 216) patients) patients) Age (years) 56 (19–72) 55 (19–68) 57 (20–72%) ns Sex Male 126 (58.3%) 78 (56.5%) 48 (61.5%) ns Female 90 (41.7%) 60 (43.5%) 30 (38.5%) Weight (kg) 73.5 (44–121) 73 (44–117) 75 (47–121) ns Underlying disease Lymphoma 102 (47.2%) 59 (42.8%) 42 (53.8%) 0.1 Multiple myeloma 114 (52.8%) 79 (57.2%) 36 (46.2%) Disease status at auto-HCT CR 123 (56.9%) 74 (53.6%) 49 (62.8%) ns PR 87 (40.3%) 60 (43.5%) 27 (34.6%) SD/active 6 (2.8%) 4 (2.9%) 2 (2.6%) Number of prior lines 0–1 130 (60.2%) 88 (63.8%) 42 (54%) ns 2 76 (35.2%) 45 (32.6%) 31 (39.7%) C3 10 (4.6%) 5 (3.6%) 5 (6.3%) Peripheral blood counts prior to mobilization: Leucocytes (9 106/ml) 5.6 (1–17.8) 5.5 (1–17.8) 5.6 (2.2–14.2) ns 6 Platelets (9 10 /ml) 221 (57–525) 217 (57–525) 234 (85–423) ns Dose of G-CSF (lg) 1560 (600–2040) 1560 (600–1920) 1560 (600–2040) ns Price G-CSF (EUR/day) 112,240 145,952 33,258 \ 0.01 (12,792–179,664) (56,120–179,664) (12,792–43,491)) Plerixafor 53 (24.5%) 33 (24%) 20 (25.6%) ns 1 dose 47 (21.8%) 32 (23.2%) 15 (19.2%) 2 doses 6 (2.8%) 1 (0.7%) 5 (6.4%) 0.05 Number of apheresis 0 2 (0.9%) 1 (0.7%) 1 (1.3%) ns 1 181 (83.3%) 113 (81.9%) 68 (87.2%) 2 31 (14.4%) 23 (16.7%) 8 (10.3%) 3 2 (0.9%) 1 (0.7%) 1 (1.3%) CD34? cells on day ?4 (9 103/ 18.8 (0.7–185) 20.2 (0.7–185) 17.3 (0.7–152) 0.1 ml) 316 Oncol Ther (2020) 8:311–324 Table 1 continued Variablesa All patients (n = NEUb group (n = 138 BIOc group (n = 78 p 216) patients) patients) \ 5 9 103/ml 30 (13.9%) 17 (12.3%) 13 (16.7%) 3 5–10 9 10 /ml 25 (11.6%) 20 (14.5%) 5 (6.4%) 0.07 3 10–15 9 10 /ml 40 (18.5%) 20 (14.5%) 2 0(25.6%) 3 [ 15 9 10 /ml 121 (56%) 81 (58.7%) 40 (56%) 3 CD34? cells on day ?5 (9 10 / 34.2 (2.5–244) 34.6 (4.1–244) 29 (2.5–223) ns ml) Total CD34 ? cells collected 4.3 (0.6–20.7) 4.3 (0.8–20.7) 4.13 (0.6–20.6) ns (9 106/kg) Mobilization failure (%) 5 (2.%) 2 (1.4%) 3 (3.8%) ns Auto-HCT yes/no 214 (99.1%)/ 2 137 (99.3%) 76 (97.4%) ns CR Complete remission, day ?4, ?5 days 4, 5, respectively, of treatment, G-CSF granulocyte colony-stimulating factor, HCT hematocrit, NS not significant, PR partial remission, SD stable disease a Variables in table are reported as the median with the range in parenthesis or as the number of patients with the percentage in parenthesis b NEU refers to NeupogenÒ (Amgen Inc.), the originator filgrastim c BIO refers to biosimilar filgrastim ZarzioÒ (Sandoz GmbH) p = 0.01). A second mobilization attempt was receiving BIO were significantly younger than indicated in one NEU patient (the other one those receiving NEU and were predominantly progressed and auto-HCT was not subsequently females, but no differences were found in terms indicated) and in two of the three patients in of mobilization outcomes (CD34? cells on the BIO group, after prior cyclophosphamide days ?4 and ?5). However, three donors (1 g/m2). These two BIO patients once again receiving BIO needed more than one apheresis needed PLEX and successfully mobilized after to achieve the minimum target, and one of one apheresis procedure. Auto-HCT could be these even needed PLEX. done in all but two cases (due to progression and failure mobilization, respectively). Cost-Effectiveness Analysis in the Patients’ Multivariable analysis (Table 4) showed three Cohort variables were independent risk factors for PLEX use: age, basal disease (lymphoma), and the The cost-effectiveness analysis based on the number of pre-mobilization therapies. On the CD34? cell levels in PB on day ?4 was favor- other hand, the use of BIO was the only risk able for BIO since the MCE was 4.41-fold lower factor associated to mobilization failure in for BIO than for NEU (193.7991 vs. 856.3112, patients receiving PLEX (hazard ratio 10.3, 95% respectively). With respect to the final number confidence interval 1.3–77.8, p = 0.02). of CD34? cells collected in the apheresis pro- duct, the analysis also favored BIO over NEU, Donors with the MCE being 4.2-fold lower for BIO (152.4128 vs. 654.6695, respectively). A total of 56 related donors were mobilized during the study period, of whom 33 received NEU and 23 received BIO (Table 5). Donors Oncol Ther (2020) 8:311–324 317 Table 2 General characteristics and mobilization data for patients who received plerixafor with CD34?\10 9 103/ml on days ?4 or ?5 Variablesa All patients Patients in NEU group Patients in BIO group p receiving PLEX (n = receiving PLEX (n = 30) receiving PLEX (n = 15) 45) Age (years) 54 (20–68) 59 (20–68) 55 (20–67) ns Weight (kg) 75 (53–117) 77 (53–110) 74 (55–117) ns Sex Male 29 (64.4%) 19 (63.3%) 10 (66.7%) ns Female 16 (35.6%) 11 (36.7%) 5 (33.3%) Underlying disease 0.3 Lymphoma 15 (77.8%) 2 (73.3%) 13 (86.7%) Multiple myeloma 10 (22.2%) 8 (26.7%) 2 (13.3%) Disease status at auto-HCT ns CR 30 (66.7%) 21 (70%) 9 (60%) PR 14 (31.1%) 9 (30%) 5 (33.3%) SD/active 1 (2.2%) 0 1 (6.7%) Number of prior lines ns 0–1 15 (33.3%) 12 (40%) 3 (20%) 2 23 (51.1%) 14 (46.7%) 9 (60%) C3 7 (15.6%) 4 (13.3%) 3 (20%) Peripheral counts prior to ns mobilization Leucocytes (9 106/ml) 4.7 (1.8–9.3) 4.6 (1.8–9.3) 5.1 (2.2–8.4) 6 Platelets (9 10 /ml) 181 (57–340) 183 (57–263) 182 (85–340) CD34? cells on day ?4 (9 4.5 (0.7–9.8) 4.8 (0.7–9.8) 2.4 (0.7–9.8) 0.02 103/ml) \ 5 9 103/ml 29 (64.4%) 17 (56.7%) 12 (80%) 0.1 3 5–10 9 10 /ml 16 (35.6%) 13 (43.3%) 3 (20%) CD34? cells on day ?5 22.1 (2.5–82.1) 20.7 (4.1–82.1) 13.8 (2.5–58.8) 0.08 (9 103/ml) Total number of mobilized 3.7 (0.5–14.7) 3.3 (1.6–14.7) 2.5 (0.5–4.9) 0.03 CD34? cells (9 106/ kg) More than 1 dose of PLEX 5 (11.1%) 1 (3.3%) 4 (26.7%) 0.02 [ 1 Apheresis procedure 9 (20%) 5 (16.7%) 4 (28.6%) ns 318 Oncol Ther (2020) 8:311–324 Table 2 continued Variablesa All patients Patients in NEU group Patients in BIO group p receiving PLEX (n = receiving PLEX (n = 30) receiving PLEX (n = 15) 45) Mobilization failure 3 (6.7%) 0 3 (20%) 0.01 PLEX Plerixafor a Variables in table are reported as the median with the range in parenthesis or as the number of patients with the percentage in parenthesis DISCUSSION significant lower number of CD34? cells col- lected (2.5 vs. 3.3 9 106 CD34?/kg; p = 0.03), The introduction of biosimilars into clinical and more BIO patients needed a second dose of practice is based on the finding of biological PLEX and, eventually, a second apheresis. The products being equivalent to originators in only independent prognostic factor that was terms of efficacy and toxicity profile, with the associated to a mobilization failure in the con- same posology and administration and the text of PLEX was the use of BIO, although this advantage of a significant decrease in total costs finding must be considered carefully because [28]. In the study reported here, the originator there were only five cases of mobilization filgrastim NEU is significantly more expensive failure. than the biosimilar filgrastim Zarzio (here Based on these results, PLEX would appear to referred to as BIO) by approximately 4.3-fold. be less efficacious when associated to BIO than The EMA approved the use of Zarzio in 2008 for to NEU, at least in the setting of G-CSF alone. shortening neutropenia days after chemother- Previous studies in patients are all based on apy [29, 30]. By extrapolation, indications were chemotherapy plus G-CSF-based mobilization, broadened to other areas, such as HSC mobi- rather than G-CSF alone [11–13]. One can lization and post-transplant use of G-CSF. eventually speculate that the use of In the setting of HSC mobilization, all ret- chemotherapy may overcome the detrimental rospective studies conducted to date have effect of BIO compared to NEU, although this combined the use of BIO and chemotherapy; hypothesis only applies to patients, and not to similar outcomes to those of NEU were found healthy donors. both in terms of the CD34? cell peak on day ?4 The use of BIO also seemed to be detrimental and total number of CD34 ?cells harvested for HSC collection in healthy donors. We found after apheresis, supporting the use of BIO that 13% of the donors who received BIO nee- instead of NEU in all indications. ded more than one apheresis to reach the With all the weaknesses and pitfalls associ- minimum level, and one of them also required ated to a retrospective analysis, our study is the PLEX. first one that compares the results of mobiliza- At the present time little data are available tion characteristics of NEU and BIO alone, out- from comparative studies on the use of PLEX in side of chemotherapy-based regimens, in combination with biosimilar G-CSF versus patients with MM or lymphoma candidates for originator G-CSF in poor mobilizers or patients auto-HCT. The major differences were seen in at high risk of mobilization failure. An Italian those patients who fulfilled the criteria for being study reported 296 patients mobilized preemp- poor mobilizers and who were candidates to tively with biosimilar G-CSF or originator G-CSF receive PLEX. In these 45 patients, the median plus PLEX, as standard practice [31], 40% fol- number of CD34? cells was significantly lower lowing chemotherapy. In contrast to our find- in patients receiving BIO than in those receiv- ings, the combination of biosimilar plus PLEX ing NEU (2.4 vs 4.8 9 103/ml; p = 0.02). More- appeared to be more efficient in terms of pri- over, the use of BIO was associated to a mary endpoints (CD34? cell threshold in blood Oncol Ther (2020) 8:311–324 319 Table 3 General characteristics of ‘ No Plerixafor’’ group Variablesa All patients not Patients in NEU group not Patients in BIO group not p receiving PLEX (n = receiving PLEX (n = 105) receiving PLEX (n = 58) 163) Age, years (median 54 [19–72] 55 [19–66] 56 [29–72] ns [min–max]) Weight (kg) 75 (44–121) 72.2 (44–112) 75.4 (52–121) ns Sex ns Male 80 (49%) 58 (55.2%) 22 (37.9%) Female 83 (51%) 47 (44.8%) 36 (62.1%) Basal disease 0.02 Myeloma 104 (63.8%) 70 (66.7%) 34 (58.6%) Lymphoma 59 (36.2%) 35 (33.3%) 24 (41.4%) Disease status: ns CR 87 (53.3%) 51 (48.6%) 36 (62.1%) PR 71 (43.5%) 50 (47.6%) 21 (36.2%) SD/active 5 (3.2%) 4 (3.8%) 1 (1.7%) Number of prior lines ns of chemotherapy 0–1 113 (69.3%) 74 (70.5%) 39 (67.2%) 2 48 (29.4%) 30 (28.5%) 18 (31%) 3 2 (1.3%) 1 (1%) 1 (1.7%) Peripheral blood counts prior mobilization Leucocytes (9 106/ 6.7 (0–17.8) 6.9 (0–17.8) 6.2 (2.9–14.2) ns ml) Platelets (9 106/ml) 247 (78–525) 224 (78–525) 253 (106–423) ns G-CSF dose (ug) 1380 (600–2040) 1560 (600–2014) 1560 (600–2040) ns CD34? cells on 39.8 (10.5–185) 33.4 (10.5–185) 23.7 (10.6–152) 0.03 day ?4 (9 103/ml) 10.1–15 9 103/ml 28 (17.1%) 14 (15.6%) 14 (26.4%) 0.1 3 [ 15 9 10 /ml 115 (82.9%) 76 (84.4%) 39 (73.6%) CD34? cells on 55.6 (8–244) 49.5 (8–244) 42 (9.1–223) ns day ?5 (9 103/ml) Total CD34 ? cells 5.5 (1.1–20.7) 4.9 (1.1–20.7) 4.79 (2.1–20.6) ns mobilized (9 106/ kg) 320 Oncol Ther (2020) 8:311–324 Table 3 continued Variablesa All patients not Patients in NEU group not Patients in BIO group not p receiving PLEX (n = receiving PLEX (n = 105) receiving PLEX (n = 58) 163) [ 1 Apheresis 19 (11.6%) 16 (15.2%) 3 (5.2%) 0.05 procedure Mobilization failure 2 (1.2%) 2 (1.9%) 0 ns a Variables in table are reported as the median with the range in parenthesis or as the number of patients with the percentage in parenthesis, unless indicated otherwise Table 4 Multivariable analysis significantly increases the final total of collected Variables Hazard ratio (95% p CD34? cells, thus assuring the minimum confidence interval) required and avoiding further apheresis proce- dures (and associated risks as thrombocytopenia Risk factors for the use of PLEX (CD34? [ 10 9 103/ or citrate reactions). However, this did not ml in peripheral blood) translate into improved long-term graft func- tion or clinical outcomes [30], and more cost- Age 0.96 (0.93–0.99) 0.04 effective analyses are needed to potentially Basal disease 0.24 (0.1–0.56) 0.001 change future guidelines for mobilization. In (reference our clinical setting, PLEX is administered myeloma) according to specific criteria based on CD34? level in PB and final CD34 ? cells yiel- Number of prior 0.32 (0.17–0.61) 0.001 ded after apheresis, with some differences lines (\ 2) among the groups [26, 31, 32, 36]. Risk factors for mobilization failure (only in PLEX Previous studies on cost-effective analyses in patients) the setting of HSC mobilization are mainly based on the use of PLEX against chemotherapy Type of GSCF 10.3 (1.3–77.8) 0.02 plus G-CSF [35, 37]. With respect to BIO versus (reference BIO) NEU, most studies have been designed in the setting of neutropenia after chemotherapy, with the results favoring the use of the biosimilar and CD34 target finally collected in apheresis),. since non-inferiority of the biosimilar was pro- The authors of the Italian study concluded that ven [38]. On the other side, as overall outcomes their data supported the standard inclusion of in previously published studies have been biosimilar filgrastim in mobilizing protocols equivalent between BIO and NEU, the universal since a stronger efficacy of the biosimilar G-CSF use of BIO in this setting can represent a sig- plus PLEX was found and, consequently, sig- nificant cost saving [12, 13]. Our cost-effective nificant cost saving would be possible. Other analysis also favored the use of BIO, in spite of studies have also found no differences with apparent lower efficacy, and we believe this is different GCSG plus PLEX [32], but in these attributable to the expensive cost of NEU (4-fold studies PLEX was used in a pre-emptive manner [ BIO) and the small sample size. and not, as in our study, only in cases of poor The present study has certain limitations due mobilization. to its retrospective design, small sample size, It is debatable if a pre-emptive use of PLEX and the different number of patients in the should be worldwide indicated in predicted groups being compared. However, these pre- poor mobilizers [33–35]. Previous studies have liminary results deserve some attention since shown that PLEX mobilization in these patients Oncol Ther (2020) 8:311–324 321 Table 5 General characteristics of donors Variablesa All donors (n = 56) Donors receiving NEU Donors receiving BIO p (n = 33) (n = 23) Age, years (median 49.5 [17–75] 53 [21–73] 46 [17–75] 0.04 [min–max]) Weight (kg) 72 (50–119) 73 (50–104) 67 (54–119) ns Sex 0.05 Male 33 (58.9%) 23 (69.7%) 10 (43.5%) Female 23 (41.1%) 10 (30.3%) 13 (56.5%) Peripheral counts prior ns mobilization Leucocytes (9 106/ml) 6.5 (4.1–13.7) 6.5 (4.2–13.7) 6.5 (4.1–10) 6 Platelets (9 10 /ml) 221 (134–369) 210 (134–297) 238 (161–369) G-CSF dose (ug) 960 (600–1560) 960 (600–1560) 1080 (600–1560) 0.1 3 CD34? on day ?4 (9 10 / 73.6 (15.3–166.2) 70 (13.3–166.2) 71.5 (22.3–156.7) ns ml) CD34? on day ?5 (9 103/ 100 (10–243) 97.7 (10–243) 100 (15.3–239.7) ns ml) Total CD34? mobilized 8.3 (0.7–23.2) 7.7 (0.7–12.9) 10.7 (3–23.2) 0.02 (9 106/kg) [ 1 Apheresis 3 (5.3%) 0 3 (13%) 0.03 Use of plerixafor 1 (1.7%) 0 1 (4.3%) 0.2 Mobilization failure 1 (1.7%) 1 (3%) 0 0.1 a Variables in table are reported as the median with the range in parenthesis or as the number of patients with the percentage in parenthesis, unless indicated otherwise there are no prior publications, to our knowl- ACKNOWLEDGEMENTS edge, reporting a comparison of these two for- mulations in the setting of only G-CSF-based We thank the participants of the study, who HSC mobilization. were responsible of the study design, data report, statistical analysis, and final review CONCLUSION Funding. No funding or sponsorship was Based on our experience and within the setting received for this study or publication of this of G-CSF-based mobilization with no prior article. chemotherapy, the combination of PLEX ? BIO Authorship. All named authors meet the might be less efficacious than PLEX ? NEU in International Committee of Medical Journal patients defined as being poor mobilizers. These Editors (ICMJE) criteria for authorship for this findings should be confiirmed in larger samples, article, take responsibility for the integrity of ideally in prospective randomized clinical trials. 322 Oncol Ther (2020) 8:311–324 the work as a whole, and have given their visit http://creativecommons.org/licenses/by- approval for this version to be published. nc/4.0/. Disclosures. Anna Sureda has received honoraria from Takeda, BMS, Sanofi, Roche, Celgene, Janssen, Novartis, Gilead-Kite; support REFERENCES for travel from Takeda, Roche, Celgene, Sanofi; fees for participating in review activities from 1. US Food and Drug Administration. Scientific con- Takeda, BMS, Celgene, Janssen, Gilead-Kite, siderations in demonstrating biosimilarity to a ref- Novartis; and payment for lectures including erence product: guidance for industry. 2015. https://www.fda.gov/downloads/Drugs/ service on speakers bureaus from Takeda, GuidanceComplianceRegulatoryInformation/ Sanofi, Celgene, Gilead-Kite, Novartis. The Guidances/UCM291128.pdf. remaining authors Rocı́o Parody, Isabel Sán- chez-Ortega, Christelle Ferrá, Ramon Guardia, 2. European Medicines Agency. Guideline on similar biological medicinal products containing biotech- Carme Talarn, Maite Encuentral, Eduard Fort, nology-derived proteins as active substance: non- David López, Mireia Morgades, Eva Alonso, clinical and clinical issues. 2014. https://www.ema. Sandra Ortega, Josep Sarrá, David Gallardo, and europa.eu/docs/en_GB/document_library/ Josep M. Ribera have nothing to disclose. Scientific_guideline/2015/01/WC500180219.pdf. 3. Mellstedt H, Niederwieser D, Ludwig H. The chal- Compliance with Ethics Guidelines. The lenge of biosimilars. Ann Oncol. 2008;19:411–9. scientific proposal was reviewed and approved by the ethical committee of the institution 4. Waller CF, Bronchud M, Mair S, Challand R. Com- parison of the pharmacodynamic profiles of a (ICO-IDIBELL), and all procedures were in biosimilar filgrastim and Amgen filgrastim: a ran- accordance with the ethical standards of the domized, phase I trial. Ann Hematol. 2010;89: 1964 Declaration of Helsinki and its later 971–8. amendments. All participants included in the 5. Waller CF, Bronchud M, Mair S, Challand R. Phar- study provided written informed consent. macokinetic profiles of a biosimilar filgrastim and Amgen filgrastim: results from a randomized, phase Data Availability. All data generated or I trial. Ann Hematol. 2010;89:927–33. analyzed during this study are included in this published article. 6. Waller CF, Semiglazov VF, Tjulandin S, Bentsion D, Chan S, Challand R. A phase III randomized equivalence study of biosimilar filgrastim versus Open Access. This article is licensed under a Amgen filgrastim in patients receiving myelosup- Creative Commons Attribution-NonCommer- pressive chemotherapy for breast cancer. Onkolo- cial 4.0 International License, which permits gie. 2010;33:504–11. any non-commercial use, sharing, adaptation, 7. Verpoort K, Möhler TM. A non-interventional study distribution and reproduction in any medium of biosimilar granulocyte colony-stimulating factor or format, as long as you give appropriate credit as prophylaxis for chemotherapy-induced neu- to the original author(s) and the source, provide tropenia in a community oncology centre. Ther Adv Med Oncol. 2012;4:289–93. a link to the Creative Commons licence, and indicate if changes were made. The images or 8. Nahon S, Rastkhah M, Abdelghani MB, Soumou- other third party material in this article are dronga RF, Gasnereau I, Labourey JL. ZarzioÒ, included in the article’s Creative Commons biosimilar of filgrastim, in prophylaxis of chemotherapy-induced neutropenia in routine licence, unless indicated otherwise in a credit practice: a French prospective multicentric study. line to the material. If material is not included Support Care Cancer. 2016;24:1991–8. in the article’s Creative Commons licence and your intended use is not permitted by statutory 9. Gascón P, Aapro M, Ludwig H, Bokemeyer C, Boc- cadoro M, Turner M. Treatment patterns and out- regulation or exceeds the permitted use, you comes in the prophylaxis of chemotherapy-induced will need to obtain permission directly from the (febrile) neutropenia with biosimilar filgrastim (the copyright holder. To view a copy of this licence, Oncol Ther (2020) 8:311–324 323 MONITOR-GCSF study). Support Care Cancer. 20. Jantunen E, Varmavuo V, Valtola J. Plerixafor 2016;24:911–25. injection: a hematopoietic stem cell mobilizer in non-Hodgkin lymphoma and multiple myeloma. 10. Waller CF, Ranganna GM, Pennella EJ, et al. Ran- Expert Rev Hematol. 2016;9:723–32. domized phase 3 efficacy and safety trial of pro- posed pegfilgrastim biosimilar MYL-1401H in the 21. Sheppard D, Bredeson C, Huebsch L, Allan D, Tay J. prophylactic treatment of chemotherapy-induced A plerixafor-based strategy allows adequate neutropenia. Ann Hematol. 2019;98:1217–24. hematopoietic stem cell collection in poor mobi- lizers: results from the Canadian Special Access 11. Lefrère F, Brignier AC, Elie C, et al. First experience Program. Bone Marrow Transplant. 2014;49:751–5. of autologous peripheral blood stem cell mobiliza- tion with biosimilar granulocyte colony-stimulat- 22. Mohty M, Azar N, Chabannon C, et al. Plerixafor in ing factor. Adv Ther. 2011;28:304–10. poor mobilizers with non-Hodgkin’s lymphoma: a multi-center time-motion analysis. Bone Marrow 12. Publicover A, Richardson DS, Davies A, et al. Use of Transplant. 2018;53:246–54. a biosimilar granulocyte colony-stimulating factor for peripheral blood stem cell mobilization: an 23. Giralt S, Costa L, Schriber J, et al. Optimizing analysis of mobilization and engraftment. Br J autologous stem cell mobilization strategies to Haematol. 2013;162:107–11. improve patient outcomes: consensus guidelines and recommendations. Biol Blood Marrow Trans- 13. Pahnke S, Egeland T, Halter J, et al. Current use of plant. 2014;20:295–308. biosimilar G-CSF for haematopoietic stem cell mobilisation. Bone Marrow Transplant. 2019;54: 24. DiPersio JF, Stadtmauer EA, Nademanee A, et al. 858–66. Plerixafor and G-CSF versus placebo and G-CSF to mobilize hematopoietic stem cells for autologous 14. Antelo ML, Zabalza A, Sánchez Antón MP, et al. stem cell transplantation in patients with multiple Mobilization of hematopoietic progenitor cells myeloma. Blood. 2009;113:5720–6. from allogeneic healthy donors using a new biosimilar G-CSF (ZarzioÒ). J Clin Apher. 2016;31: 25. DiPersio JF, Micallef IN, Stiff PJ, et al. Phase III 48–52. prospective randomized double-blind placebo-con- trolled trial of plerixafor plus granulocyte colony- 15. Bonig H, Becker PS, Schwebig A, Turner M. stimulating factor compared with placebo plus Biosimilar granulocyte–colony-stimulating factor granulocyte colony-stimulating factor for autolo- for healthy donor stem cell mobilization: need we gous stem-cell mobilization and transplantation for be afraid? Transfusion. 2015;55:430–9. patients with non-Hodgkin’s lymphoma. J Clin Oncol. 2009;27:4767–73. 16. Schmitt M, Hoffmann JM, Lorenz K, Publicover A, Schmitt A, Nagler A. Mobilization of autologous 26. Sancho JM, Duarte R, Medina L, Querol S, Marı́n P, and allogeneic peripheral blood stem cells for Sureda A. Mobilization of peripheral blood stem transplantation in haematological malignancies cells with plerixafor in poor mobilizer patients. Med using biosimilar G-CSF. Vox Sang. 2016;111: Clin (Barc). 2016;147:223.e1–.e7. 178–86. 27. Drummond MF, O’Brien BJ, Stoddart GL, Torrance 17. Manko J, Walter-Croneck A, Jawniak D, et al. A GW. Methods for the economic evaluation of clinical comparison of the efficacy and safety of health care programs. 2nd ed. Oxford: Oxford biosimilar G-CSF and originator G-CSF in University Press; 1997. haematopoietic stem cell mobilization. Pharmacol Rep. 2014;66:239–42. 28. Berghout A. Clinical programs in the development of similar biotherapeutic products: rationale and 18. Nasillo V, Paolini A, Riva G, et al. Effectiveness of general principles. Biologicals. 2011;39:293–6. originator (Neupogen) and biosimilar (Zarzio) fil- grastim in autologous peripheral blood stem cell 29. Raedler LA. Zarxio (filgrastim-sndz): first biosimilar mobilization in adults with acute myeloid leuke- approved in the United States. Am Health Drug mia: a single-center retrospective study. Leuk Benefits. 2016;9:150–4. Lymphoma. 2018;59:225–8. 30. Visram A, Bredeson C, Allan D, et al. Long-term 19. Maul JT, Stenner-Liewen F, Seifert B, et al. Effica- graft function following autologous hematopoietic cious and save use of biosimilar filgrastim for cell transplantation and the impact of preemptive hematopoietic progenitor cell chemo-mobilization plerixafor in predicted poor mobilizers. Blood with vinorelbine in multiple myeloma patients. Cancer J. 2018;8:14. J Clin Apher. 2017;32:21–6. 324 Oncol Ther (2020) 8:311–324 31. Lanza F, Saraceni F, Pezzi A,et al. A comparative 35. Greil C, Ihorst G, Kiote-Schmidt C, et al. Stem cell analysis of biosimilar vs originator filgrastim in mobilization in poor mobilizers with multiple combination with plerixafor for stem cell mobi- myeloma or lymphoma before and after introduc- lization in lymphoma and multiple myeloma: a tion of plerixafor: a single-center comparative propensity-score weighted multicenter approach. analysis using a cost-efficient single fixed-dose Am J Hematol. 2017;92:557–9. schedule. Leuk Lymphoma. 2018;59:1722–5. 32. Bhamidipati PK, Fiala MA, Grossman BJ, et al. 36. Smith VR, Popat U, Ciurea S, et al. Just-in-time Results of a prospective randomized, open-label, rescue plerixafor in combination with chemother- noninferiority study of Tbo-Filgrastim (Granix) apy and granulocyte-colony stimulating factor for versus Filgrastim (Neupogen) in combination with peripheral blood progenitor cell mobilization. Am J plerixafor for autologous stem cell mobilization in Hematol. 2013;88:754–7. patients with multiple myeloma and non-Hodgkin lymphoma. Biol Blood Marrow Transplant. 37. Antar A, Otrock ZK, Kharfan-Dabaja MA, et al. 2017;23:2065–9. G-CSF plus preemptive plerixafor vs hyperfraction- ated CY plus G-CSF for autologous stem cell mobi- 33. Greil C, Kiote-Schmidt C, Fink G, et al. Successful lization in multiple myeloma: effectiveness, safety peripheral blood stem cell mobilization with a cost- and cost analysis. Bone Marrow Transplant. efficient single fixed-dose plerixafor schedule in 2015;50:813–7. poor mobilizers. Leuk Lymphoma. 2017;58: 1849–58. 38. Aapro M, Cornes P, Abraham I. Comparative cost- efficiency across the European G5 countries of var- 34. Ogunniyi A, Rodriguez M, Devlin S, et al. Upfront ious regimens of filgrastim, biosimilar filgrastim, use of plerixafor and granulocyte-colony stimulat- and pegfilgrastim to reduce the incidence of ing factor (GCSF) for stem cell mobilization in chemotherapy-induced febrile neutropenia. J Oncol patients with multiple myeloma: efficacy and Pharm Pract. 2012;18:171–9. analysis of risk factors associated with poor stem cell collection efficiency. Leuk Lymphoma. 2017;58:1123–9.