CD19 expression is maintained in DLBCL patients after treatment with tafasitamab plus lenalidomide in the L-MIND study

Introduction

CD19 is an important target for novel anti-lymphoma treatments as it is broadly and homogenously expressed across many B-cell malignancies [1,2]. Approximately 30–50% of patients with diffuse large B-cell lymphoma (DLBCL) who do not respond to first-line therapy with R-CHOP have a poor prognosis and need effective treatment options, especially those ineligible for autologous stem cell transplant (ASCT) [2]. With the emergence of cellular- and antibody-based therapies targeting CD19 [3–5], it is of scientific interest to study the expression of CD19 in this patient population with few treatment options. Salles et al. recently reported durable complete responses in a significant proportion of patients with relapsed or refractory (R/R) DLBCL from the phase II study (L-MIND; NCT02399085) of tafasitamab, an Fc-modified, anti-CD19 monoclonal antibody, in combination with lenalidomide, an immunomodulatory agent [2]. Results from L-MIND led to the US FDA approval of the tafasitamab plus lenalidomide combination as a second and subsequent line treatment option for ASCT-ineligible patients with R/R DLBCL [5]. However, changes in CD19 expression after tafasitamab treatment may impact subsequent CD19-targeted approaches, such as CAR T-cell therapy; understanding expression changes could inform optimal sequencing of treatment options and identify treatment feasibility. Available data on this question remain limited and appear to vary by drug class and indication. An anti-CD19 (antibody–drug conjugate) immunotherapy did not preclude subsequent responses to CD19-directed CAR T-cell therapy, with maintenance of CD19 expression [6]. This contrasts with reports of CD19 expression loss after bispecific anti-CD19 treatment and CD19-directed CAR T-cell treatments [7]. For approved anti-CD19 therapies, more data are needed to understand treatment impact on target expression. This is the first report on CD19 expression analyzed in tumor biopsies in R/R DLBCL patients before and after tafasitamab treatment.

Methods

We analyzed 12 tumor lymph node biopsies (four pre- and eight post-tafasitamab treatment) of six patients with R/R DLBCL from three centers in Germany and the Czech Republic, who were treated with tafasitamab plus lenalidomide in the L-MIND study [2]. Samples were obtained from formalin-fixed and paraffin-embedded core needle lymph node biopsies; L-MIND participation required patients to be willing to undergo biopsy [2]. CD19 expression was analyzed by immunohistochemistry (IHC; anti-CD19 clone LE-CD19 (Zytomed; 1:400 dilution; citric acid pH 6.0) used according to standard procedures) in all samples, as well as DNA whole exome and RNA exome sequencing – permitting sufficient sample material/quality (Twist Human Core Exome, Twist Bioscience Corp, South San Francisco, CA; and TrueSeq RNA Exome, Illumina Inc., San Diego, CA).

DNA and RNA extraction was of sufficient quantity and quality from six samples (one pre- and five post-tafasitamab treatment (including one paired sample)), allowing subsequent DNA and RNA exome sequencing. DNA reads were mapped using the Illumina DRAGEN Bio-IT Platform 3.4 (Illumina Inc., San Diego, CA) and variants annotated using VarSeq software 2.2.1-SP3 (Golden Helix Inc., Bozeman, MT). CD19 variants from unmatched DNA whole exome sequencing (WES) were assessed using Ingenuity Variant Analysis (Qiagen GmbH, Hilden, Germany). RNA sequencing data were mapped to human reference genome GRCh38.p using CLC Genomics Workbench 20.0.3 (CLC bio, Qiagen GmBH, Hilden, Germany). Log2 transformed reads per kilobase of transcript, per million mapped reads (RPKM) values were used to assess gene expression. Splice junctions for CD19 exons were determined for samples with sufficient depth of coverage using Arraystudio v.11.0 (OmicsSoft^®, Qiagen GmBH, Hilden, Germany).

Results and discussion

Of the six patients, four had paired pre- and post-tafasitamab treatment biopsies. For the remaining two patients, only post-tafasitamab treatment biopsies were available. Patients received a median 14.5 (range 1–60) doses of tafasitamab (Table 1). Of eight lymph node biopsies taken after tafasitamab treatment, three were taken within the timeframe of potential residual exposure to tafasitamab (i.e. 85 days or five tafasitamab half-lives (5 × 17 days)), and five were taken outside of this timeframe (>85 days). Of the eight post-tafasitamab biopsies, seven were taken before next line of therapy and one was taken post-chemotherapy; chemotherapy was not expected to impact CD19 expression. All 12 biopsies, including those post-tafasitamab treatment, demonstrated distinct CD19 expression by IHC independent of the duration of tafasitamab therapy, response to study treatment, or any potential residual exposure to tafasitamab (Figure 1(A)).

Figure 1. CD19 expression. (A) IHC data from serial core needle lymph node FFPE biopsies of six L-MIND patients; (B) CD19 exon skipping analysis from bulk RNA sequencing data from evaluable post-baseline samples. (A) Pre-tafasitamab and post-tafasitamab (during and after residual tafasitamab exposure) treatment. Shown are images (×400) after IHC staining of cytoplasmic CD19 epitope (anti-CD19 clone LE-CD19). Biopsies after tafasitamab discontinuation were categorized as taken within or after the timeframe of potential residual tafasitamab exposure (cut-off: five half-lives after last tafasitamab dose or 85 days (∼12 weeks)). Distinct CD19 expression can be observed despite any response-related selective pressure or presence of tafasitamab. Differences in staining intensity are considered to be caused by differences in sample quality. (B) Exon skipping was observed for exons 2 (E1–E3), 5 (E4–E6), and 5/6 (E4–E7). Frequency corresponds to the number of splice junction reads observed for each 3′-exon relative to total 5′-exon junction reads. The percentage of canonical E1–E2 and E4–E5 junction reads is provided for each patient. BOR: best objective response; FFPE: formalin-fixed paraffin-embedded; IHC: immunohistochemistry; PD: progressive disease; PR: partial response; SD: stable disease; W: week.

Table 1. Patient demographics, disease characteristics, and CD19 analyses.

Patient	#1	#2	#3	#4	#5	#6
Age (years)^a	72	59	62	55	76	82
Sex (M/F)^a	M	F	F	M	M	M
Biopsy
Pre-tafasitamab, Y/N	Y	Y^e	Y	Y	N	N
Post-tafasitamab	8 days (≤85 days post-tafasitamab)/18 weeks (>85 days post-tafasitamab)^e^,^f	14 days (≤85 days post-tafasitamab)^e^,^f	15 weeks (>85 days post-tafasitamab)/25 weeks (>85 days post-tafasitamab)	2 years, 3 months (>85 days post-tafasitamab)^e^,^f	10 months (>85 days post-tafasitamab)^e^,^f	7 weeks (≤85 days post-tafasitamab)^e
Prior therapies before L-MIND	1. R-CHOP	1. R-CHOP 2. R-DHAP	1. R-CHOP	1. R-CHOP2. R-DHAP  + HD BEAM	1. R-CHOP 2. GemOx	1. R-CHOP
Therapies after L-MIND	1. GemOx 2. Odronextamab 3. Polatuzumab vedotin + atezolizumab	–	1. R-DHAP 2. Polatuzumab vedotin 3. Odronextamab	1. R-GemOx 2. Axicabtagene ciloleucel	1. R-PACEBO 2. Palliative Rx	1. R mono 2. Local Rx
Prior CD20 therapy^a	Y	Y	Y	Y	Y	Y
Primary refractory^a^,^d	N	Y	N	N	Y	N
Prior ASCT^a	N	N	N	Y	N	N
Ann Arbor staging at screening^a	IV	IV	IV	II	IV	I
ECOG PS^a	1	1	0	0	1	1
Bulky disease^a	N	N	N	N	N	N
Cell of origin	Non-GCB	GCB	GCB	GCB	N/A	Non-GCB
Number of tafasitamab doses^a	15	1	10	16	60 (30 × 2 tafasitamab doses)	14
Tafasitamab duration of treatment (first to last dose), weeks	16.7	0.1	8.1	16.0	109.1	62.2
Best objective response to tafasitamab + LEN^a^,^b	PR	PD	PD	SD	CR	PR
CD19 IHC expression, post-treatment	Y	Y	Y	Y	Y	Y
CD19 DNA mutational status	NM_001770.5: c.1580-4C>A SNP (neutral)	WT	WT	WT	WT	WT
CD19 RNA expression, log2 (RPKM)^c	4.7	4.6/5.6	N/A	6.2	6.6	3.6

ASCT: autologous stem cell transplant; CR: complete response; ECOG PS: Eastern Cooperative Oncology Group Performance Status; F: female; GCB: germinal center B-cell; GemOx: gemcitabine plus oxaliplatin; HD: high dose; IHC: immunohistochemistry; LEN: lenalidomide; M: male; mo: month; mono: monotherapy; N/A: not available; N: no; OR: objective response; PACEBO: prednisolone, doxorubicin, cyclophosphamide, etoposide; PD: progressive disease; PR: partial response; SD: stable disease; R: rituximab; R-CHOP: rituximab, cyclophosphamide, hydroxydaunorubicin, oncovin, and prednisone; R-DHAP: rituximab, dexamethasone, cytarabine, and cisplatin; RPKM: reads per kilobase of transcript, per million mapped reads; Rx: radiotherapy; SNP: single nucleotide polymorphism; W: week; WT: wild type; Y: yes.

a Baseline assessment before starting tafasitamab in the L-MIND study.

b Investigator assessed, as per Cheson et al. criteria [8].

c Two samples (baseline and relapse) were available for patient #2, while no sample was available for patient #3.

d Patients had progressed during or within 6 months after completion of their first line of therapy.

e Samples used for DNA sequencing and RNA expression analyses.

f Samples used for RNA splicing analyses.

In cases with available baseline samples (n = 4), CD19 expression post-tafasitamab treatment was comparable to baseline levels prior to tafasitamab therapy in terms of staining intensity and cellular distribution. As a cytosolic CD19 epitope was stained, no conclusion on CD19 integrity could be drawn. To obtain further evidence of the integrity of the CD19 protein and its mutational status, we performed respective DNA and RNA sequencing (at baseline, during, and/or after potential residual tafasitamab exposure) for six samples (5/6 patients; two samples were taken from the same patient). Analysis of DNA WES data revealed no likely somatic mutations in the CD19 gene. Analysis of CD19 exons from bulk RNA sequencing data from evaluable post-baseline samples (n = 4) showed no evidence of exon skipping as a dominant escape mechanism, suggesting expression of mainly full-length CD19 mRNA (Figure 1(B)). Preexisting low-level CD19 exon skipping has been previously reported for non-leukemia donors and B-cell acute lymphoblastic leukemia tumor cells prior to anti-CD19 treatment [9,10]. It remains unclear whether the observed limited exon skipping originates from subclones. Furthermore, the presence of CD19-negative subclones before initiation of therapy may not preclude CAR T-cell activity [11], potentially due to their elimination by FAS death receptor signaling [12]. CD19 mRNA expression post-tafasitamab was detectable in all analyzed biopsies. The patient with paired biopsy sequencing data (pre- and post-tafasitamab) showed consistent CD19 mRNA expression levels (Table 1). These observations further corroborate the comparable CD19 protein expression detected by IHC.

As tafasitamab targets CD19, it is important to elucidate the effect of this antibody on CD19 expression at relapse, and potential implications on subsequent lines of therapy if tumors have antigen loss or a reduction in antigen density [13]. There has been little clinical evidence to direct the sequencing of anti-CD19 CAR T-cell therapy following a prior therapy targeting CD19. Our data provide the first early evidence to support the potential therapeutic sequencing of tafasitamab and subsequent anti-CD19 CAR T-cell therapy. Encouragingly, a recent case report provided the first clinical evidence showing sustained remission in a DLBCL patient receiving anti-CD19 CAR T-cell therapy after prior treatment with tafasitamab [14]. Apart from these data, preliminary evidence has shown favorable outcomes of CAR T-cell therapy after treatment with another investigational anti-CD19 immunotherapy, loncastuximab tesirine – an antibody–drug conjugate with a humanized anti-CD19 monoclonal antibody component [6]. In the study in question, no cases of CD19 antigen-negative relapse were observed where expression was reassessed after relapse or upon progression [6]. Our data, considered in context, support that patients receiving tafasitamab treatment may retain potential to benefit from subsequent anti-CD19 CAR T-cell therapy.

Limitations of this work include sample size. Furthermore, DNA and RNA analyses were performed on unmatched samples, such that baseline CD19 expression was not determined in most patients and patient-specific germline variants could not be confidently filtered. Moreover, the study used exome capture-based RNA sequencing that may not detect all possible mechanisms of transcriptome changes, such as previously reported CD19 intron 2 retention [15]. Finally, IHC provides limited information on antigen levels, without any conclusions on CD19 surface expression. We hope to build upon this research and report findings at a later date.

In conclusion, IHC analysis showed a comparable, distinct CD19 expression before and after tafasitamab therapy in this subset of L-MIND study patients. DNA and RNA analyses did not find evidence for CD19 mutations, dominant exon skipping or loss of CD19 mRNA expression, which would be indicative of resistance to further CD19-targeted therapy. These findings indicate a maintained CD19 expression after tafasitamab therapy and may provide a rationale for subsequent CD19-directed therapies in patients with R/R DLBCL.

Authors contributions

All authors contributed to data acquisition, manuscript development, and approval. All authors interpreted the results and agree on accountability for all study aspects, including accuracy, integrity, and protocol adherence. All authors contributed to study design or conduct, data analyses, or manuscript writing.

Acknowledgements

The authors would like to thank the patients and their families, as well as clinical researchers and their teams who have participated in this study. The authors thank Tim Burn, PhD, for his input on this manuscript. Medical writing assistance was provided by Ori Bowen, PhD, of Syneos Health.

Disclosure statement

J. Duell’s institution has received patient fees from MorphoSys. A. Obr reports consultancy fees from Janssen and honoraria from Roche. M. Augustin reports consultancy fees from Bristol-Myers Squibb, MSD, Pfizer, PharmaMar, IPSEN, AstraZeneca; research funding (institution) from Bristol-Myers Squibb, MSD, Morphosys, AstraZeneca, Pfizer, PharmaMar; travel expenses from Lilly, Novartis, Bristol-Myers Squibb, PharmaMar and IPSEN. J. Endell, S. Geiger, and S. Ambarkhane are employees of MorphoSys AG. H. Liu and I.M. Silverman are employees and stockholders of Incyte Corporation.

About tafasitamab

Tafasitamab is a humanized Fc-modified cytolytic CD19-targeting monoclonal antibody. In 2010, MorphoSys licensed exclusive worldwide rights to develop and commercialize tafasitamab from Xencor, Inc.

Tafasitamab incorporates an XmAb^® engineered Fc domain, which mediates B-cell lysis through apoptosis and immune effector mechanism including antibody-dependent cell-mediated cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP).

In January 2020, MorphoSys and Incyte entered into a collaboration and licensing agreement to further develop and commercialize tafasitamab globally. Following accelerated approval by the U.S. Food and Drug Administration in July 2020, tafasitamab is being co-commercialized by MorphoSys and Incyte in the United States. Incyte has exclusive commercialization rights outside the United States.

XmAb^® is a trademark of Xencor, Inc.

Additional information

Funding

This work is funded by MorphoSys AG.