Rituximab biosimilars for lymphoma in Europe

ABSTRACT

Introduction: The approval of rituximab, a monoclonal antibody targeting CD20, revolutionized the treatment of B-cell non-Hodgkin lymphomas and became an undisputed standard of care. However, as with all biologic medicines, the complex development and manufacturing process for rituximab have meant that the medicine attracts high treatment costs. Approved rituximab biosimilars have been comprehensively demonstrated to match the reference medicine. With the potential to increase access to biologic therapy, they have a key role in helping to improve patient outcomes in lymphoma care.

Areas covered: In this review, we discuss the role of rituximab in the treatment of lymphoma. We explore development and regulatory requirements for biosimilar development and the potential impact of these medicines on access and sustainability. Focusing on biosimilars of rituximab, we examine in detail the evidence for biosimilarity for the two rituximab biosimilars that are approved in Europe and provide an overview of rituximab biosimilars currently in development.

Expert opinion: We foresee a wider uptake of biosimilar medicines for lymphoma treatment over the next 5 years. The associated cost savings should be invested in broadening patient access to biological therapies, enabling wider use of more expensive treatment strategies and driving innovation in cancer care.

KEYWORDS:

• Biosimilar
• chronic lymphocytic leukemia
• diffuse large B-cell lymphoma
• extrapolation
• follicular lymphoma
• rituximab
• non-Hodgkin lymphoma

1. Introduction

Rituximab (MabThera®/Rituxan®, Roche/Genentech) is a monoclonal antibody approved by the European Medicines Agency (EMA) for the treatment of non-Hodgkin lymphoma (NHL), chronic lymphocytic leukemia (CLL), rheumatoid arthritis, granulomatosis with polyangiitis (Wegener’s syndrome), and microscopic polyangiitis [1]. Rituximab was the first monoclonal antibody and the first B-cell-targeted therapy to be approved for therapeutic use in oncology [2–4]. Since its first approval, rituximab has revolutionized the treatment of B-cell NHL, significantly improving therapeutic responses, as indicated by enhanced response rates [3,5]. Rituximab is also associated with longer progression-free survival (PFS) than previously available regimens [3,5], and remains a fundamental component of treatment regimens for B-cell NHL and CLL today [6].

Biosimilars of rituximab are now available in Europe and other regions [7,8]. A biosimilar is a biologic medicine that matches an existing approved biologic, known as the reference medicine, in terms of safety, efficacy, and quality [9]. Since 2006, when a biosimilar of the growth hormone somatropin became the first biosimilar medicine to be approved by the EMA, a total of 58 biosimilars have been approved for use across Europe to date, including human growth hormone, erythropoiesis-stimulating agents, insulins, monoclonal antibodies, fusion proteins, and monoclonal antibody fragments [10]. Similar regulatory pathways have also seen the approval of biosimilars in other countries and territories including Australia, New Zealand, Japan, Switzerland, Canada, and the USA. It has been proposed that rituximab biosimilars have the potential to enhance patients’ access to this valuable medicine [11].

The first rituximab biosimilars CT-P10 (Truxima®; developed by Celltrion) and GP2013 (Rixathon®; developed by Sandoz) were approved by the EMA in 2017 [7,8]. The approvals were based on the totality of evidence for biosimilarity derived from a comprehensive comparability exercise with the reference medicine [7,8,12]. Several proposed biosimilars of rituximab are also in development.

In this review, we will discuss EMA-approved and proposed rituximab biosimilars in relation to their use for the treatment of lymphoma and CLL.

2. Body

2.1. Rituximab in lymphoma

Rituximab is a genetically engineered chimeric mouse/human monoclonal antibody [1]. The mechanism of action of rituximab is based on its targeted binding to CD20 on the surface of B-cells, and subsequent mediation of B-cell depletion [1]. It is understood that several effector mechanisms of rituximab contribute to this effect, including complement-mediated cytotoxicity (CDC), antibody-dependent cell-mediated cytotoxicity (ADCC), apoptosis, and antibody-dependent cellular phagocytosis (ADCP), although the extent of involvement of each of these mechanisms in different indications remains unclear [13–16].

Initial pivotal trials of rituximab in the early 2000s investigated the addition of rituximab to the standard-of-care regimen of cyclophosphamide, vincristine, and prednisone (CVP) chemotherapy for first-line induction treatment of follicular lymphoma (FL). The addition of rituximab resulted in major improvements in all clinical endpoints, including increased time to treatment failure and doubling in time to disease progression, as well as improvements in response rates, duration of response, and disease-free survival, while the safety profile remained broadly similar to CVP alone [17,18]. Similar results were observed when rituximab was added to other chemotherapy regimens – e.g. bendamustine or CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) – for the induction and maintenance treatment of untreated or relapsed/refractory FL [19–26]. Most recently, results from 10 years’ follow up of the PRIMA study found that 2 years of rituximab maintenance treatment increased median PFS to 10 years, as compared with a median PFS of 4 years for patients who did not receive maintenance treatment [27].

For first-line treatment of patients with diffuse large B-cell lymphoma (DLBCL), the addition of rituximab to the standard-of-care chemotherapy regimen (CHOP) for eight cycles improved response rate by 13% and prolonged event-free and overall survival (OS) [28]. Improvements in survival outcomes were maintained after 5 and 10 years of follow-up [29,30]. The addition of rituximab did not result in increased toxicity compared with CHOP alone [28–30]. Further studies confirmed the utility of rituximab treatment when added to a variety of different chemotherapy regimens in both older and younger patients with DLBCL [31–38].

For the treatment of previously untreated patients with CLL and those with relapsed/refractory disease, the addition of rituximab to standard fludarabine plus cyclophosphamide chemotherapy provided significant improvements in PFS, OS, overall response rate (ORR), and complete response [39,40]. Improvements in survival and response outcomes have also been observed when rituximab was added to other chemotherapy regimens, including bendamustine and chlorambucil [5].

Based on these favorable clinical data, rituximab plus chemotherapy is recommended as the standard of care for first-line treatment of FL and DLBCL and offers a treatment option in CLL [6,41].

2.2. Development and approval of biosimilars of monoclonal antibodies

The goal of development for biosimilars is to conclusively demonstrate that the biosimilar matches the reference medicine. Stringent regulatory pathways for biosimilar approval are established in several regions (EU: EMA; USA: Food and Drug Administration [FDA]; Japan: Pharmaceuticals and Medical Devices Agency [PMDA]; Australia: Therapeutic Goods Administration [TGA]). Common to all is a requirement for proof of similarity across multiple levels of evidence, which together form the totality of evidence for biosimilarity [42–45]. The development process, therefore, involves a stepwise comparability exercise where each step removes further residual uncertainty regarding potential clinical differences between the two molecules [46].

Regulatory authorities agree that the best possible basis of demonstrating comparability for a biosimilar and reference medicine is to use extensive analytical testing. Owing to continuing improvement, currently available analytics have much greater sensitivity, resolution, and throughput compared with those accessible to the developers of the first generation of biologic medicines [47,48]. By focusing on molecular attributes known to have an impact on efficacy and safety, analytical data can confirm a lack of clinically relevant differences between biosimilar and reference medicine within stringent and tightly controlled limits [47,49–52].

For now, regulatory pathways require additional testing to confirm that the biosimilar and reference medicine behave in a similar way in living systems. Therefore, comparative preclinical and clinical PK/PD testing and immunogenicity assessments are also performed. Currently, the final step in the process is to conduct at least one confirmatory Phase III clinical trial in a sensitive indication – one in which clinically meaningful differences in safety, immunogenicity, and efficacy are most likely to be detected.

Once biosimilarity is established by the totality of evidence, safety and efficacy data gathered for the reference medicine may be applied to the biosimilar for all approved indications of the reference medicine. This process is known as extrapolation [51].

2.3. Considerations for clinical testing of biosimilars in oncology

Selection of a suitable patient population and suitable study endpoints are important considerations for demonstration of clinical comparability between a biosimilar and reference medicine [42]. The indication must be sensitive enough to detect clinically meaningful differences between the biosimilar and reference medicine in terms of safety, immunogenicity, and efficacy. Ideally, the patient population should be homogeneous with known sensitivity to the effects of the reference medicine [42,53]. Extensive experience with rituximab in untreated, advanced FL provides a vast body of comparative data in populations with known sensitivity to the effects of rituximab, making this a suitably sensitive population for the assessment of rituximab biosimilars.

When selecting suitable endpoints, consideration needs to be given to the relevance of the endpoints to the disease, as well as their sensitivity to detect any clinically relevant differences between the biosimilar and reference medicine [42,51,54–57]. Endpoints traditionally used for the assessment of novel drugs in FL, such as PFS and OS, may not provide sufficient sensitivity for the detection of clinically meaningful differences in biosimilar trials [53] and can be influenced by factors that are independent of the drugs’ effects [42,58,59]. To power a lymphoma biosimilar trial for the detection of clinically meaningful differences in PFS or OS, a much larger sample size (several thousand patients) and a much longer study duration (6–8 years for PFS and >10 years for OS) would be needed [60]. Regulatory authorities recommend that the response rate is a suitable endpoint for biosimilar trials in oncology as it provides a direct measure of the antitumor activity of the evaluated agents [42]. For lymphoma, ORR is a good choice for comparing the efficacy of biosimilars as it is more sensitive to potential clinical differences and provides relatively rapid results compared with survival endpoints [42].

2.4. EMA-approved rituximab biosimilars

Two rituximab biosimilars are currently approved by the EMA for the treatment of lymphoma: GP2013 (Sandoz; brand names Rixathon®/Riximyo®) [7] and CT-P10 (Celltrion; brand names Truxima®/Ritemvia®/Rituzena®/Blitzima®) [8]. For both biosimilars, several different brand names are used owing to between-country differences in licensing and patent protection, as well as alternative distribution arrangements. Both biosimilars were approved according to the stringent regulatory pathway described above, following the demonstration that they are similar to reference rituximab (MabThera®/Rituxan®). Both biosimilars are approved for use in all indications of reference rituximab, including the hemato-oncology indications FL, DLBCL, and CLL by the EMA [7,8].

For GP2013, comprehensive preclinical data have been published, demonstrating similarity to reference rituximab [61,62]. An extensive array of analytical tests demonstrated that the two molecules had matching amino acid sequences and were indistinguishable in terms of molecular mass, 3D structure, disulfide bridging patterns, charge variants, amino acid modifications, and post-translational modifications, such as glycosylation and sialylation. GP2013 was shown to have similar size heterogeneity, and aggregate and particle impurity levels, when compared with reference rituximab [61]. Functional characterization included cell-based CD20 binding assays, and cell-based potency assays for central rituximab effector mechanisms including ADCC, CDC, and apoptosis. For all assays, activity for GP2013 was within the range of the activity for reference rituximab. Pre-clinical comparisons included evaluation of PK/PD in cynomolgus monkeys and antitumor activity assessment in mouse xenograft models. PK/PD and antitumor activity profiles were similar for GP2013 and reference rituximab [62].

Preclinical data have also been published for CT-P10 [8,63]. Although the exact methods used for assessment differed slightly, similar parameters were compared during the development of each biosimilar. Primary and higher-order structures of CT-P10 and reference rituximab were found to be indistinguishable in all analyses. Analysis of charge and size variants and glycosylation patterns found that these parameters for CT-P10 were within acceptable ranges defined for reference rituximab. CD20 and Fc receptor binding, as well as functional bioactivity, were also comparable for CT-P10 and reference rituximab. Functional bioactivity, in terms of ADCC, CDC, and apoptosis, was comparable for CT-P10 and reference rituximab. Unlike for GP2013, CT-P10 experiments also evaluated ADCP, which was found to be within the activity range for reference rituximab. PK/PD profiles, evaluated in cynomolgus monkeys, were within similar ranges for CT-P10 and reference rituximab. Overall, preclinical data showed high similarity between CT-P10 and the reference medicine [8,63].

The clinical development programs for the two approved rituximab biosimilars are summarized in Table 1. Overall, the clinical development programs for GP2013 and CT-P10 were highly similar. Both manufacturers selected similar indications (FL, rheumatoid arthritis) in which to test clinical performance. FL represented the most sensitive lymphoma indication for the assessment of biosimilarity because the largest effect size for rituximab when added to standard chemotherapy was observed in this indication (24% difference in response rate with rituximab plus CVP compared with CVP alone, compared with a 6% difference in response rate for rituximab plus CHOP compared with CHOP alone [17,19]). Other lymphomas represent less sensitive indications for assessment of biosimilarity; for example, the effect size with rituximab plus CHOP compared with CHOP alone in DLBCL was only 13% [28].

Table 1. Clinical development program for GP2013 and CT-P10.

Study	Design	Indication	Primary endpoint	Number of patients
CT-P10
CT-P10 1.1 [63,64]	Phase I RCT (2:1)* Double blind	RA	PK equivalence between biosimilar and reference rituximab	154
CT-P10 1.3 [65] (1.1 follow-on study)	Phase I Single arm Open label	RA	Long-term efficacy and safety of biosimilar rituximab	58
CT-P10 1.2 [8,66]	Phase I Single arm Open label	DLBCL	Initial evidence of safety for biosimilar rituximab	N/A
CT-P10 3.2 [67]	Phase III RCT (1:1:1) Double blind	RA	PK and therapeutic equivalence between biosimilar and reference rituximab	372
CT-P10 3.3 [68]	Phase I/III RCT (1:1) Double blind	Advanced FL	PK equivalence and therapeutic noninferiority between biosimilar and reference rituximab	140
CT-P10 3.4 [69]	Phase III RCT (1:1) Double blind	LTB FL	Therapeutic equivalence between biosimilar and reference rituximab	258
GP2013
JP-trial [70]	Phase I Single arm Open label	Indolent LTB NHL	Safety and PK of biosimilar rituximab	6
ASSIST-RA [71]	Phase II RCT^† Double blind	RA	PK equivalence between biosimilar and reference rituximab	312
ASSIST-FL [11]	Phase III RCT (1:1) Double blind	Advanced FL	Therapeutic equivalence between biosimilar and reference rituximab	629
ASSIST-RT [72]	Phase III RCT (1:1) Double blind	RA	Safety and immunogenicity	107

DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; LTB, low tumor burden; N/A, not available;

NHL, non-Hodgkin lymphoma; PK, pharmacokinetics; RA, rheumatoid arthritis; RCT, randomized controlled trial.

*CT-P10 : reference rituximab; ^†Randomization ratio 1:1 in Part 1, 1:2 (GP2013 : reference rituximab) in Part 2.

FL is also a sensitive indication for safety evaluations, including immunogenicity and infusion-related reactions. By contrast, rheumatoid arthritis represented a more sensitive indication for the assessment of PK/PD because between-patient variability in PK, PD, and baseline B-cell counts is much lower than in oncology indications [11]. In addition, the 24-week dosing regimen enabled capture of the complete concentration–time profile. The independent yet convergent decisions on studied indications suggest that these were indeed the most appropriate indications for clinical assessment of rituximab biosimilars.

Some differences exist between the two clinical development programs, including the selection of patient populations. The confirmatory Phase III study for GP2013 was conducted in patients with FL [11], whereas the confirmatory study for CT-P10 was conducted in rheumatoid arthritis [67]. For GP2013, two supportive studies were conducted in rheumatoid arthritis [71,72] and for CT-P10, two supportive studies were conducted in advanced and low tumor burden (LTB) FL [68,69].

The study designs for the key lymphoma trials for the two biosimilars (GP2013 study name: ASSIST-FL; CT-P10 study name: CT-P10 3.3) are compared in Table 2 and the patient populations for the two trials are compared in Table 3 [11,68]. It should be noted that there are always limitations when comparing results from different clinical studies. Indeed, small differences in study design and patient populations can be observed between ASSIST-FL and CT-P10 3.3, emphasizing that any comparison between the studies should be interpreted with caution. Both trials included patients with advanced FL and used CVP as chemotherapy. ASSIST-FL had a larger patient population than CT-P10 3.3, and the trials used different versions of reference rituximab as comparator (ASSIST-FL: EU-sourced reference rituximab; CT-P10 3.3: US-sourced reference rituximab). The site of WHO grading assessment also differed between trials (ASSIST-FL: central assessment; CT-P10 3.3: local assessment). A higher dose of prednisone and longer duration between maintenance doses were permitted in ASSIST-FL than in CT-P10 3.3.

Table 2. Comparison of study design for Phase III clinical trials in FL for GP2013 and CT-P10.

	ASSIST-FL [11] Phase III	CT-P10 3.3 [68] Phase III
Number of patients randomized	N = 629	N = 140 (N = 121 in part 1)
Patient characteristics	Advanced FL (Ann Arbor Stage III–IV) WHO Grade I–IIIa (confirmed by central pathological testing)	Advanced FL (Ann Arbor Stage III–IV) WHO Grade I–IIIa (based on local laboratory review)
Induction therapy	R-CVP (8 x 3-week cycles)	R-CVP (8 x 3-week cycles)
	IV cyclophosphamide (750 g/m^2) on Day 1	IV cyclophosphamide (750 g/m^2) on Day 1
	IV vincristine (1.4 mg/m^2) on Day 1	IV vincristine (1.4 mg/m^2) on Day 1
	Oral prednisone (100 mg) on Days 1–5	Oral prednisone/prednisolone (40 mg/m^2) on Days 1–5
	IV GP2013 or Ref-RTX-EU (375 mg/m^2) on Day 1	IV CT-P10 or Ref-RTX-US (375 mg/m^2) on Day 1
Maintenance therapy	IV GP2013 or Ref-RTX-EU (375 mg/m^2) Every 3 months for 2 years (every 2 months in Italy)	IV CT-P10 or Ref-RTX-US (375 mg/m^2) Every 2 months for 2 years
Primary objective(s)	To demonstrate equivalent efficacy between GP2013 and Ref-RTX-EU in terms of centrally assessed best overall response (CR+PR) up to 24 weeks	Part 1 – To demonstrate PK equivalence between CT-P10 and Ref-RTX-US in terms of AUCtau and CmaxSS over induction Cycle 4 (Weeks 9–12) Part 2 – To demonstrate noninferior efficacy of CT-P10 compared with Ref-RTX-US in terms of centrally assessed best overall response (CR+CRu+PR) up to 24 weeks
Secondary objectives	PK: Cmax, Ctrough, AUC0–21d, AUCall PD: Median B-cell count, AUEC Efficacy: PFS, OS Safety and immunogenicity	PK: Cmax, Ctrough, Cav, Tmax, VSS, T1/2, MRT, PTF PD: Median B-cell count Efficacy: PFS, TTP, TTF, DFS, OS Safety and immunogenicity

AUC_0–21d, area under the serum concentration–time curve from time zero to Day 21; AUC_all, area under the serum concentration–time curve from Day 0 to the time of last observation; AUC_tau, area under the serum concentration–time curve at steady state; AUEC, area under the effect curve of peripheral CD19 + B-cell counts; C_av, average serum concentration; C_max, maximum serum concentration; C_maxSS, maximum serum concentration at steady state; CR, complete response; CRu, unconfirmed complete response; C_trough, concentration before infusion; DFS, disease-free survival, FL, follicular lymphoma; IV, intravenous; MRT, mean residence time; OS, overall survival; PD, pharmacodynamics; PFS, progression-free survival; PK, pharmacokinetics; PR, partial response; PTF, peak-to-trough fluctuation ratio; R-CVP, rituximab + cyclophosphamide + vincristine + prednisone; Ref-RTX-EU/US, EU/US-sourced reference rituximab;; TTF, time to treatment failure; TTP, time to progression; T_1/2, terminal elimination half-life; T_max, time to maximum serum concentration after infusion; WHO, World Health Organization; V_SS, volume of distribution at steady state.

Table 3. Patient demographics and baseline disease characteristics for the Phase III studies in advanced FL for GP2013 and CT-P10.

		ASSIST-FL [11] (N = 627)		CT-P10 3.3 [68] (N = 140)
		GP2013 (N = 312)	Reference rituximab (N = 315)	CT-P10 (N = 70)	Reference rituximab (N = 70)
Age, years		57.5 (SD: 11.86)	56.4 (SD: 11.72)	57.0 (IQR: 45–66)	58.5 (IQR: 47–66)
Gender – female, n (%)		181 (58)	169 (54)	40 (57)	37 (53)
FLIPI score, n (%)	0–1	30 (10)	35 (11)	8 (11)	6 (9)
	2	106 (34)	103 (33)	25 (36)	21 (30)
	>3	176 (56)	177 (56)	37 (53)	43 (61)
ECOG performance score, n (%)	0	179 (57)	175 (56)	44 (63)	47 (67)
	1	125 (40)	123 (39)	25 (36)	22 (31)
	2	5 (2)	13 (4)	1 (1)	1 (1)
	Missing	3 (1)	4 (1)	0	0
Bulky disease (tumor >10 cm [ASSIST-FL]/ 7 cm [CT-P10 3.3]), n (%)		44 (14)	56 (18)	11 (16)	14 (20)

FL, follicular lymphoma; FLIPI, follicular Lymphoma International Prognostic Index; IQR, interquartile range; SD, standard deviation.

Patient demographics and baseline disease characteristics were broadly similar between the treatment arms in each trial. In the CT-P10 3.3 trial, a higher proportion of patients in the CT-P10 arm had Ann Arbor Stage IV disease compared with the reference rituximab arm (bone marrow involvement at baseline 64% vs. 47%, respectively). Both studies met the primary efficacy endpoints of equivalent or noninferior ORR between biosimilar and reference rituximab (Figure 1) [11,68]. In meeting the primary efficacy endpoint, both trials demonstrated that the efficacy of biosimilar rituximab was within the acceptable range of reference rituximab.

Figure 1. Primary efficacy results of the Phase III studies of GP2013 and CT-P10 in patients with advanced FL [11,68].

CI, confidence interval; FL, follicular lymphoma.

Notably, the approach to statistical demonstration of equivalence differed slightly between the trials. In ASSIST-FL, the comparison between the biosimilar and reference medicine was based on an equivalence margin of ±12%, with two-sided confidence intervals (CIs). This meant that the upper and lower boundaries of the CIs for the estimate of ORR for GP2013 could be no more than 12% above or below the ORR estimate for reference rituximab. In fact, a difference of −0.4%, with 95% CI of −5.94, 5.14, was observed between GP2013 and reference rituximab. As the 95% CI was well within the predefined equivalence margins, these results confirmed that GP2013 was neither better nor worse than reference rituximab in terms of ORR. In CT-P10 3.3, the comparison was based instead on a noninferiority margin of −7%, with a one-sided CI. This approach meant that the boundary of the CI for CT-P10 could be no more than 7% below the ORR for reference rituximab. In fact, a difference of 4.3%, with one-sided 97.5% CI –4.25, was noted between CT-P10 and reference rituximab. As the 97.5% CI was on the positive side of the predefined noninferiority margin of 7%, this result confirmed that CT-P10 was not worse than reference rituximab in terms of ORR. In ASSIST-FL, 90% and 95% CIs were used for the equivalence margins to satisfy the differing preferences of the EMA and FDA regarding CIs. In CT-P10 3.3, a 97.5% CI was used. It is important to note that the percentage chosen for CI is somewhat arbitrary and does not influence the statistical significance of the results. For the same trial data, a lower percentage will provide tighter intervals around the estimated value, whereas a higher percentage will provide wider intervals.

Both studies included survival endpoints; however, neither study was powered for statistical assessment of PFS or OS. Safety and immunogenicity were broadly similar between biosimilar and reference rituximab in both trials (Table 4) [11,68]. Grade 3 neutropenia was higher in the CT-P10 arm compared with the reference rituximab arm in CT-P10 3.3. This may have been driven by the higher level of bone marrow involvement at baseline in the CT-P10 arm. PK/PD profiles for both rituximab biosimilars were also found to be within the similarity range for reference rituximab.

Table 4. Summary of safety and immunogenicity results from the induction treatment phase of Phase III studies of GP2013 and CT-P10 in patients with advanced FL.

	ASSIST-FL [11] (N = 627)n, %		CT-P10 3.3 [68](N = 140)n, %
	GP2013 (N = 312)	Referencerituximab(N = 315)	CT-P10 (N = 70)	Referencerituximab(N = 70)
Any adverse event	289 (93)	288 (91)	58 (83)	56 (80)
Grade ≥3 neutropenia	55 (18)	65 (21)	20 (29)	12 (17)
Serious adverse event	71 (23)	63 (20)	16 (23)	9 (13)
Infusion-related reaction	41 (13)	37 (12)	15 (21)	17 (24)
Deaths	4 (1)	7 (2)	1 (1.4)	0
Anti-drug antibodies	5/268 (2)	3/283 (1)	3 (4)	2 (3)

FL, follicular lymphoma.

2.5. Rituximab biosimilars in development

In addition to the two approved biosimilars, five proposed rituximab biosimilars are in an advanced stage of development (Table 5). One was submitted to the EMA for consideration in 2018: MabionCD20 (Mabion) [73]. A further four are in Phase III clinical trials: PF-05280586 (Pfizer), RTXM83 (mAbxience), ABP 798 (Amgen), and SAIT101 (Archigen). The availability of published data differs for each biosimilar.

Table 5. Overview of rituximab biosimilars, proposed biosimilars, and biomimics.

		Published supporting data
					Clinical
Medicine	Company	Physicochemical	Functional	Preclinical	Indolent NHL/FL	DLBCL	RA	First approval/regulatory status
Biosimilars (approved by EMA)
GP2013	Sandoz (Germany)	✓	✓	✓	✓	✖	✓	EU (2017)
CT-P10	Celltrion (South Korea)	✓	✓	✖	✓	✖	✓	EU (2017)
Proposed biosimilars (not yet approved by the EMA, but under clinical development)
PF-05280586	Pfizer (US)	✓	✓	✓	✓	✖	✓
RTXM83	mAbxience (Spain)	✓	✓	✓	✖	✓	✖	Argentina (2014)
ABP 798	Amgen (US)	✓	✓	✖	✖	✖	✖
MabionCD20	Mabion (Poland)	✖	✖	✖	✖	✖	✖	Submitted to EU EMA (2018)
SAIT101	Archigen (UK)	✖	✖	✖	✖	✖	✖
Biomimics (distributed in some countries, but not approved to the regulatory standards of a biosimilar for the EMA)
BCD-020	Biocad (Russia)	✖	✖	✖	✓	✖	✓	Russia (2014)^†
Reditux^TM	Dr Reddy’s (India)	✖	✖	✖	✖	✓	✖	India (2007)^‡

^†Distributed in Eastern Europe, Africa, and Asia; ^‡Distributed in Asia and Latin America.

DLBCL, diffuse large B-cell lymphoma; EMA, European Medicines Agency; FL, follicular lymphoma; NHL, non-Hodgkin lymphoma; RA, rheumatoid arthritis.

For PF-05280586, published data are available from all stages of development [74–78]. Physicochemical and functional similarity was demonstrated using an array of comparability assessments similar to those described for GP2013, as well as in vitro CDC activity assays [74]. Preclinical testing confirmed the comparability of PK/PD profiles in cynomolgus monkeys [74]. Phase I clinical testing compared PK/PD profiles of biosimilar and reference rituximab in patients with active rheumatoid arthritis and included an extension period during which patients who originally received reference rituximab were switched to the biosimilar after the initial study period (first treatment course) [75–77]. PK/PD, immunogenicity, safety, and efficacy were similar between biosimilar and reference rituximab during the extension phase [77]. A Phase III confirmatory study was conducted in patients with LTB FL [78]. The primary endpoint of equivalent ORR between the biosimilar and reference rituximab was met. In addition, the estimated PFS at 1 year was also found to be highly similar. Additional response rate endpoints, PK/PD, safety, and immunogenicity profiles were also similar for biosimilar and reference rituximab over 26 weeks.

RTXM83 is approved in Argentina as a rituximab biosimilar. Published data are available for analytical and preclinical testing, as well as from a Phase III clinical trial in DLBCL [79–81]. Preclinical comparisons for RTXM83 and reference rituximab confirmed matching primary and higher-order structures, and post-translational modifications. CD20 binding assays, in vitro CDC potency, and in vivo PK/PD in cynomolgus monkeys confirmed comparable functional bioactivity between RTXM83 and reference rituximab [79]. CD20-specific cell binding and ADCC were also assessed in human cell lines [80]. Overall, preclinical data showed that RTXM83 and reference rituximab are highly similar in terms of structure and function. In a Phase III confirmatory trial in DLBCL, the primary endpoint of noninferiority in ORR between RTXM83 and reference rituximab was met [81]. PK/PD profiles were also shown to be similar between the two treatment arms.

Limited published data are available regarding analytical assessments undertaken for ABP 798 [82]. Comparable receptor binding affinity, as well as ADCC and CDC activity, was demonstrated in cell-based assays. Phase III trials are ongoing in LTB FL and rheumatoid arthritis; results were not available at the time of this review [83,84].

No published preclinical or clinical data are available for MabionCD20, which was accepted for regulatory review by the EMA in June 2018. Phase III clinical testing was conducted in both rheumatoid arthritis and DLBCL [85,86]. No published preclinical or clinical data are available for SAIT101. Phase III clinical testing is ongoing in LTB FL [87].

2.6. Potential pharmacoeconomic impact of rituximab biosimilars in lymphoma

Monoclonal antibodies such as rituximab are complex to develop and manufacture, resulting in high treatment costs, which are a possible factor in limitations in access to cancer medicines in Europe and elsewhere [88–90]. The abbreviated development pathway for biosimilars outlined in the previous sections, with a reduced emphasis on clinical evaluation, greatly reduces the costs of development [91,92]. There is emerging evidence that biosimilars are likely to be available at lower costs versus reference medicines in many health-care systems [93,94]. A recent European assessment of biosimilar pricing in rheumatoid arthritis reported an average 86% reduction in the retail price of biosimilar rituximab versus the reference product [95].

Such cost reductions may lead to increased patient access to biologic therapies [96–98]. However, despite Europe representing a substantial proportion of the global biosimilars market, global data evaluating the economic impact of biosimilar use in oncology and hematology versus reference products are limited. It has been proposed that reduced prices for biosimilars may have the potential to prompt competitive price reductions for reference medicines [99,100]. Further economic analysis will be needed to appreciate how the pricing of biosimilars will be affected by the entry of more biosimilars to the market [95].

2.7. Biomimics

A biomimic is a copy of a biologic medicine that has not been developed, assessed, or approved according to the same stringent regulatory standards that are applied for approval of biosimilars by the EMA and FDA. Biomimics are claimed to have high similarity with the reference medicine but the stepwise comparability exercise required for US and EU approval is not completed in full or in sufficient depth for these medicines. The lack of rigor in the development process of biomimics could mean that there are differences in the efficacy or safety of the drug and therefore they cannot be considered biosimilars [101]. Some countries (e.g. China, Mexico, India) have different requirements for the approval and licensing of biological medicines. Biomimics are available in these markets but may be associated with different efficacy or toxicity profiles to the reference medicine [102].

Reditux^TM (Dr. Reddy’s Laboratories) was approved for use in India in 2007 and is also distributed in markets in Asia and Latin America. Available analytical data for Reditux^TM include structural and functional comparisons with reference rituximab (MabThera®/Rituxan®/Ristova® [alternative brand name used by Roche in India]) [103–105]. These data show similarities between the biomimic and reference medicine in terms of size heterogeneity, higher-order structure, and intact and reduced mass. Differences were observed in terms of charge heterogeneity, glycan profiles, and reduced stability, as well as significant differences in biological activities, including reduced ADCC and CDC activity for the biomimic. No published data are available from randomized controlled clinical trials with Reditux^TM. A single-arm study in 21 patients with DLBCL suggested that PK/PD profiles were similar between Reditux^TM and published data for reference rituximab, although no direct statistical comparison was conducted [106]. Published clinical experience with Reditux^TM includes safety and effectiveness data in patients with DLBCL and rheumatoid arthritis but limited comparisons between the biomimic and reference medicine are available [106–112].

BCD-020 (Biocad) is available in markets in Eastern Europe, Asia, and Africa. Limited published preclinical data are available; similarities between BCD-020 and reference rituximab in terms of quality characteristics, in vitro biological activity, toxicology, and PK/PD profiles in nonhuman primates have been demonstrated [113]. Phase III clinical testing was conducted in patients with FL or marginal zone lymphoma, and in rheumatoid arthritis [113,114]. Both studies showed comparable efficacy, PK/PD, and safety profiles for reference rituximab and BCD-020.

3. Conclusion

Biologic treatments, including monoclonal antibodies such as rituximab, have greatly improved patient outcomes in several types of cancers and are now regarded as the standard of care. However, access to biologics can be limited by high treatment costs [115], which pose a challenge to patients, families, providers, and insurers. The development and introduction of biosimilars that match a reference medicine in terms of structure, function, immunogenicity, efficacy, and safety have the potential to provide savings for health-care systems, broaden patient access to biologic therapies, and support the sustainability of cancer care [116–118].

Two biosimilars of rituximab have already been approved by the EMA for use in the same indications as reference rituximab. Additional proposed rituximab biosimilars are in Phase III trials or under consideration by the EMA. The FDA has so far taken a more cautious approach to regulation of biosimilars, and only one biosimilar rituximab, CT-P10, is currently approved in the USA. The increased availability of rituximab biosimilars for the treatment of lymphoma is expected to further decrease the costs associated with monoclonal antibody treatment, thereby improving patient access to rituximab, as well as enabling health-care systems to invest in innovative therapies.

4. Expert opinion

Just as generic medicines before them, biosimilars have proved to be controversial upon their first introduction. When the first generic aspirin was introduced, it was widely considered to be an inferior product to the branded version. However, with improved education and understanding, generic medicines have since gained wide acceptance and are part of daily clinical practice. Unfortunately, biosimilars have faced similar barriers to uptake. Common themes emerge when physicians discuss their reluctance to prescribe biosimilars, often centering around the idea of ‘similar but not identical’. Concerns remain about the efficacy and safety of biosimilars relative to their reference medicines, particularly in extrapolated indications. Since biosimilar monoclonal antibodies are still relatively new to market, many physicians are unfamiliar with the regulatory concepts that underpin them. Improved education on these concepts is therefore required.

Interestingly, the stepwise comparability exercise and extrapolation process used in biosimilar development are not new ideas, nor are they specific to biosimilar medicines. Rather, they are established principles that have been applied historically to biologic and other medicines. The very nature of biologic medicines as large, complex molecules that are manufactured in living cells means that they are associated with inherent variability and are particularly sensitive to changes in the manufacturing process, which occur frequently. These changes require that the critical quality attributes of biologic medicines are tightly controlled within defined reference ranges during manufacture. These are the same attributes and ranges that are compared between reference and biosimilar medicines. Indeed, the basis for the comprehensive biosimilar comparability exercise required by regulatory authorities lies in the process for confirming similarity following manufacturing process changes for reference biologics.

Similarly, the concept of extrapolation is a common regulatory process. Extrapolation may be applied when approval is granted for a medicine to be used in different indications than that in which it was tested; for example, when a medicine that was tested in adults is licensed for use in children. Extrapolation is also applied, specifically to biologic medicines, after manufacturing process changes. Depending on the type of change, different levels of evidence are required to remove uncertainty regarding the similarity of the pre- and post-change molecules. While minor changes require only analytical comparisons, major changes may require clinical testing to confirm similarity of efficacy and safety. In these cases, clinical data are gathered in just one sensitive indication, one in which clinical differences would be most easily detectable, rather than in all indications for which the medicine is approved. In this way, extrapolation based on sound scientific justification avoids the need for unnecessary and potentially unethical clinical trials.

Since the first biosimilar approval in Europe, more than 10 years of data have been gathered from thousands of patients treated in biosimilar clinical trials and real-world clinical practice. No concerns relating to similarity of efficacy to the reference medicine have been reported, either in tested or extrapolated indications, and there have been no safety issues.

Looking to the future, we believe that biosimilars will play an increasing role in cancer treatment. This will apply not only to biosimilars that are already available but also to the introduction of new biosimilar medicines as we see patents expiring on increasing numbers of biologic medicines. As these new biosimilars enter the market, we hope that increasing familiarity with biosimilar concepts and trust in the development and regulatory processes will lead to swifter uptake than has occurred so far. With exciting novel medicines also in development, we would hope that increasing the use of biosimilars wherever possible will help to reduce health-care costs, not only enabling wider access to current therapies but also enabling access to higher-cost novel treatments, such as CAR-T cell therapy. However, while cost issues are of great concern across oncology, drug manufacturing practices that do not respond to accepted and stringent development criteria cannot be uncritically reported or described. Therefore, it is extremely important to differentiate stringently regulated biosimilars from less highly regulated biomimics, which have a high potential for altered efficacy and safety profiles compared with the reference medicine. Overall, we hope that savings borne by the use of biosimilar medicines could be reinvested in tackling the expected increase in cancer prevalence in the future and in driving pharmaceutical innovation, to see more novel, efficacious treatments become available for our patients. In our view, biosimilars are likely to play a key role in helping to achieve the sustainability of cancer care.

Article highlights

• Biologic therapies are associated with high development and treatment costs, which can result in restrictions in patient access to these therapies.

• Two rituximab biosimilars have been approved for lymphoma treatment, meeting the stringent regulatory guidelines for biosimilar development; these medicines are approved for use in all indications of reference rituximab.

• Several additional biosimilars of rituximab are in development and are likely to be available in the near future.

• Biomimics of rituximab, which are not developed with the same rigor as approved biosimilars, are also available in some countries.

• Increased availability of rituximab biosimilars for the treatment of lymphoma can improve patient access to rituximab treatment and is, in our opinion, expected to result in cost savings and contribute to achieving sustainability of cancer care. Furthermore, improving global access to established biologics, such as rituximab, may also facilitate access to new treatment modalities that are possible with more recently approved drugs.

This box summarizes the key points contained in the article.

Declaration of interest

W Jurczak has received research funding from and participated in advisory boards for Celltrion, Roche, and Sandoz. M Długosz Daneka has received research funding from Celltrion, Roche, and Sandoz. C Buske has received consultancy/honoraria from AbbVie, Celgene, Celltrion, Janssen, Pfizer, Roche, and Sandoz. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer Disclosures

Peer reviewers on this manuscript have no relevant financial relationships or otherwise to disclose.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Acknowledgments

Under the authors’ direction, medical writing support was provided by Amanda Hatton, MSc of Spirit, a division of Spirit Medical Communications Group Ltd., Manchester, UK, and funded by Sandoz, a Novartis Division.

Additional information

Funding

This paper was funded by Sandoz, a Novartis Division.