Abstract
Introduction.
SIRPα+ macrophages can mediate resistance to lenalidomide and rituximab (R2) in patients with B-cell non-Hodgkin lymphoma (B-NHL). Evorpacept (ALX148) is a novel CD47 blocker that abrogates interactions between lymphoma cells and SIRPα+ macrophages.
Methods.
Adult patients with B-NHL who had received at least 2 prior lines of systemic therapy were included in this single arm phase I study (NCT05025800). Evorpacept was administered intravenously (IV), in a 28-day cycle, until progression, at two dose levels (DL): 30 mg/Kg on day (D) 1 and D15 (DL1), or 60 mg/Kg on day 1 (DL2); rituximab 375 mg/m2 IV was given weekly during cycle 1, and on D1 during cycles 2–6; lenalidomide 20 mg was given orally on D1–21 during cycles 1–6. Single-cell RNA sequencing was performed on tumor biopsies collected before treatment and during cycle 1.
Results.
Twenty patients were included in this study. Median age was 61 (27–85) years and 18 (90%) had indolent B-NHL. Three patients were treated at DL1, 17 at DL2, and no dose limiting toxicity was observed. The most common grade 3–4 adverse events included: neutropenia (60%), infections (30%), and alanine transferase increase (15%). Sixteen (80%) patients achieved complete response and after a median follow-up of 28 months 2-year progression-free survival rate was 69%. During treatment, a significant increase in T cells and macrophages was observed, and macrophages pathways associated with anti-tumoral activity were upregulated.
Conclusions.
ER2 has a safe toxicity profile, promising anti-tumoral activity, and induces favorable biological effects on the tumoral immune microenvironment.
Introduction
The combination of lenalidomide and rituximab, also referred to as R2, is an active regimen for patients with previously untreated or relapsed refractory B-cell non-Hodgkin lymphoma (B-NHL), particularly indolent B-NHL (iNHL).(1–3) However, only one third of iNHL patients who receive R2 for relapsed disease achieve complete response (CR).(3) Of interest, CR is significantly associated with prolonged progression-free survival (PFS) for patients with iNHL, suggesting that modifying R2 to increase CR rates may translate into improved outcomes.(4)
Our group and others have previously demonstrated that pro-tumoral macrophages, mainly positive for signal regulatory protein α (SIRPα), increase at time of progression and may mediate resistance to R2 in patients with iNHL.(5,6) Evorpacept (former ALX-148) is a novel high-affinity CD47 blocker that abrogates the ‘do-not-eat-me signal’ provided to SIRPα+ macrophages, without targeting red blood cells.(7,8) In addition, it increases antibody-dependent cellular phagocytosis (ADCP) when combined with monoclonal antibodies.(7,9)
We hypothesized that evorpacept and R2 (referred to as ER2) can be safe and effective for the treatment of patients with relapsed refractory B-NHL. The safety of the combination of evorpacept and rituximab was investigated in a phase I study including 33 patients with relapsed refractory B-NHL, and was shown to be well tolerated and effective, with more than 50% of patients with iNHL achieving CR.(10) We present here the results of a phase I study investigating the safety and efficacy of ER2 in patients with relapsed refractory B-NHL.
Methods
Patients and Study Design
This is a single center, single arm, open label phase I study, conducted at The University of Texas MD Anderson Cancer Center (MDACC) from November 2021 to September 2023, with a data cut-off for reported follow-up of January 2025 (NCT05025800). All patients enrolled in the study provided written informed consent. The study was approved by the Institutional Review Board of MDACC and conducted in accordance with our institutional guidelines and the principles of the Declaration of Helsinki. Adult patients with relapsed refractory B-NHL who had received at least 2 prior lines of systemic therapy were included in this study. Only 1 prior line of systemic treatment was required for patients with iNHL, including follicular lymphoma (FL) and marginal zone lymphoma (MZL). Patient with iNHL also had to meet GELF criteria to be eligible.(11) Patients previously exposed to agents targeting the CD47-SIRPα axis were excluded. Detailed eligibility criteria are provided in Supplementary Protocol Information. Evorpacept was administered intravenously (IV), in a 28-day cycle, until progression (or for 12 cycles in case of CR), at two dose levels (DL): 30 mg/Kg on day (D) 1 and D15 (DL1), or 60 mg/Kg on day 1 (DL2); rituximab 375 mg/m2 IV was given weekly during cycle 1, and on D1 during cycles 2–6; lenalidomide 20 mg was given orally on D1–21 during cycles 1–6. Prophylactic acetylsalicylic acid was used based on age and cardiovascular risk as per institutional guidelines.
End points and Assessment
The primary endpoint was the safety and tolerability of ER2 in patients with relapsed refractory B-NHL. Secondary endpoints included overall response rate (ORR) and CR rate, PFS, and overall survival (OS). Exploratory endpoints included impact of ER2 on the tissue tumoral immune microenvironment, with focus on macrophage function.
For patients with iNHL, the FL international prognostic index (FLIPI), FLIPI-2 and PRIMA-prognostic index (PI) scores were calculated as previously described.(12–14) Adverse events (AEs) were prospectively graded according to the Common Toxicity Criteria for Adverse Events (CTCAE) version 5. Response status was prospectively assessed by the Lugano 2014 classification, using positron emission tomography-computed tomography (PET-CT) after 3 and 6 cycles, at the end of treatment, every 3 months for the first 2 years, every 6 months for the subsequent 3 years, and then yearly until progression.(15) In cases of complete metabolic response, use of CT with contrast was allowed instead of PET-CT during long-term monitoring. Bone marrow biopsy was performed for confirmation of CR in iNHL cases where it was positive for lymphoma involvement before initiation of treatment.
Patient samples and tumor tissue processing
Viably cryopreserved tissue samples were stored in −80C until processing. Frozen tissue samples were thawed by gently swirling tubes in a 37C water bath. Thawed contents were strained using a 70um cell strainer (Fisher Scientific) and transferred to gentleMACS™ C Tubes (Miltenyi Biotec; Cat.# 130-096-334). Thawed tissue was dissociated using the Tumor Dissociation Kit (Miltenyi Biotec; Cat.# 130-095-929) in tandem with the gentleMACS™ Octo Dissociator with Heaters (Miltenyi Biotec; Cat.# 130-134-029) according to manufacturer instructions. A custom program was designed to gently dissociate tissue by slowly increasing rotational speed and temperature up to 200rpm and 37C, respectively. Upon program completion, dissociated samples were strained through a 70um cell strainer, rinsed with a neutral buffer and centrifuged at 300 RCF for 5 min. After centrifugation, the supernatant was removed and samples were resuspended in complete media and counted with AO/PI dye on the Nexcelom cell counter. Quality of samples was assessed according to live cell counts and percent viability. If viability was less than 70%, a Ficoll separation clean-up step was performed. In 1mL of completed media, such samples were added to tubes of 6mL of Ficoll-Paque PLUS (Cytiva, cat# 17144003) and centrifuged for 30 minutes at 400rpm. The enriched sample was transferred to a new tube and again counted with AO/PI dye on the Nexcelom cell counter.
Library preparation and sequencing
Dissociated tissue samples were enumerated using AO/PI on the Nexcelom Automated Cell Counter. Samples with greater than 50,000 live cells were fixed following the “Fixation of Cells & Nuclei for Chromium Fixed RNA Profiling” protocol from 10X Genomics using the Chromium Next GEM Single Cell Fixed RNA Sample Preparation Kit (10X genomics, PN# 1000414) according to manufacturer instructions. Cells isolated from the tissue were normalized according to the sample with the lowest number of cells per pools, with a minimum of 5,000 cells/uL and loaded on to 10X Genomics Chromium X/iX Single Cell Analyzer (RRID:SCR_024537) according to the steps outlined in the Chromium Fixed RNA Profiling Reagent Kits user guide (10X Genomics). Barcoded nucleotides were used to construct libraries and RNA sequencing was performed on the Illumina NovaSeq X Plus (approximately 100,000,000 reads/sample).
Data processing
For single cell RNA sequencing (scRNA-seq) data processing, FASTQ files were processed using the standard CellRanger (v8.0.1)(RRID:SCR_017344) workflow to generate the gene expression count matrix for each sample, using the human genome (hg38) as the reference. Filtered count matrices were merged into an AnnData object.
Quality control
Quality control was performed using Scanpy (v1.10.4)(RRID:SCR_018139) (16). Gene filtering excluded genes expressed in fewer than 300 cells. For cell filtering, the following criteria were applied: the number of expressed genes per cell was required to range between 200 and 5000, mitochondrial RNA content had to be less than 8%, and total counts per cell had to be within 5 median absolute deviations (MAD). Out of thirty three samples, twenty-eight scRNA-seq samples passed QC step.
Potential doublets were identified and removed using Scrublet (v0.2.3)(RRID:SCR_018098) with automatically set doublet score thresholds for each sample (17). To address potential batch effects during clustering, the Harmony (v0.0.10)(RRID:SCR_022206) algorithm was applied, ensuring robust data integration across samples.
Dimensionality reduction
For scRNA-seq data analysis, we first normalized the gene expression of each cell to 10,000 counts and applied a logarithmic transformation. Then, the top 4000 most variable genes were selected for dimensionality reduction and clustering. For visualization, a UMAP was calculated by computing the single-cell neighborhood graph (kNN-graph) on the specific principal components using 20 neighbors. The number of principal components used in the neighborhood graph was determined based on the standard deviations of the top 16 principal components. The Leiden graph-clustering method was applied to cluster the neighborhood graph of cells.
InferCNV Analysis
We employed the inferCNV approach, utilizing infercnvpy (v.0.5.)(RRID:SCR_025804)(18), to estimate chromosomal copy number variations (CNVs). The analysis was performed as per the instructions available at https://github.com/broadinstitute/infercnv.
Single cell RNA sequencing and clustering
Single cell RNA sequencing (scRNA-seq) was performed on tumoral tissue samples collected at baseline (BL) and during cycle 1 (OT) to analyze the biological impacts of this combination on the phenotype and functions of the tissue-resident immune microenvironment, with focus on myeloid cells. In the first round of clustering, in addition to B cells, four main cell types were identified: T cells (TRAC, CD3D, CD2, IL32, CD3E), myeloid cells (CD68, MZB1, LYZ, CD14, LAMP3, CLEC9A, IRF7), NK cells (NKG7, KLRK1, TRDC, GNLY, GZMA), and endothelial cells (IGFBP7, ENG, PECAM1, VWF, SPARCL1). These classifications were based on canonical cell type markers.(19,20) A second round of clustering was performed within each major cell type to identify subclusters; markers used to further type the cells included CD4+ T cells (CD4, IL7R, CCR7), CD8+ T cells (CD8A, CCL5, GZMK, GZMA), Tregs (FOXP3, CTLA4, IL32, TNFRSF4), macrophages (LYZ, CST3, CD68, FCER1G, PSAP), and dendritic cells (CLEC4C, ITM2C, GZMB, IL3RA, IRF8).(20–22) A subsequent round of clustering focused on identifying myeloid cells, and five main subtypes were identified: the plasmacytoid dendritic cells (pDC) were characterized by the expression of GZMB, IL3RA, COBLL1, and TCF4; conventional dendritic cells (cDC) were identified by markers such as CLEC9A, CADM1, CST3, and COTL1; monocyte-derived dendritic cells (mo-DC) expressed CD86, CD40, IRF4, CD80, CCR7, and CD40; M1 macrophages were marked by S100A8, S100A9, MARCO, IL1B, TNF, TLR2, and IL6; M2 macrophages were identified with markers including CD68, MRC1, SIGLEC1, CD163, APOE, MMP9, IL10, CSF1, CCL18, C1QC, C1QB, C1QA, PLTP, and SELENOP.
Differential expression analysis
Differential gene expression (DEG) analysis was performed by scRNA-seq to identify macrophage genes that were differentially expressed between two groups of samples, using the Wilcoxon rank-sum test (RRID: SCR 108139). Scanpy was jused to perform downsampling of the data, adjusting for differences in sequencing depth and cell composition. DEG analysis was carried out with the thresholds of an adjusted p value of ≤ 0.05 and a fold change of ≥ 1. P values were corrected using the Benjamini–Hochberg method to control for multiple testing.
Pathways enrichment analysis
Gene Set Enrichment Analysis (GSEA) by scRNA-seq was conducted using GSEApy,(23) with gene sets obtained from the Molecular Signatures Database (MSigDB)(RRID: SCR 016863).(24,25) For pathway analysis, Scanpy was used to perform down-sampling (RRID: SCR 108139)
Statistical Analysis
The final analysis was conducted at the point when all patients could be evaluated for at least 30 days after ER2 initiation. A Bayesian Optimal Interval design with a target dose limiting toxicity (DLT) rate of 0.25 was used, and DLT was evaluated during the first 30 days after ER2 initiation. Details regarding DLT definitions are provided in Supplementary Protocol Information. PFS time was defined as the interval from the start of therapy to progression of disease, death, or last follow-up (whichever occurred first). OS time was defined as the interval from the start of therapy to death or last follow-up. PFS and OS were calculated for all patients in the study and for subgroups of patients using Kaplan-Meier estimates. A p-value of ≤ 0.05 (two-tailed) was considered statistically significant. Wilcoxon signed-rank test, with multiple comparisons corrected by the Benjamini-Hochberg procedure, was utilized for comparisons of immune cells and pathways. Statistical software SAS 9.4 (SAS, Cary, NC)(RRID: SCR 008567) and TIBCO Spotfire S+ 8.2 (TIBCO Software Inc., Palo Alto, CA)(RRID: SCR 008858) were used for all the analyses.
Data availability
The data reported in this article are not publicly available, as they are either confidential or proprietary. To the extent allowed, the authors will provide access to de-identified participant-level data underlying the data presented in this article to researchers who provide a methodologically sound proposal for academic purposes to interpret, verify and extend research in the article that does not violate privacy, data encumbrance, intellectual property or other legal, regulatory or contractual confidentiality obligations, immediately after article publication. Data provided will be subject to a data use agreement. Researchers should contact the corresponding author when applying for data access. Response to external data requests will be within a reasonable timeframe of a few weeks to months depending on the nature of the request. Use of data will be restricted to agreed purpose.
Results
Patients
Twenty patients were included in the study. All patients were evaluable for safety and efficacy. Median age was 61 (range, 27–85 years), 10 (50%) were male, and self-reported race/ethnicity included 13 (65%) White patients. Most patients had iNHL: 15 (75%) had FL, 3 (15%) MZL, 1 (5%) mantle cell lymphoma, and 1 (5%) Richter Syndrome (RS); all had previously received an anti-CD20 monoclonal antibody, 13 (72%) chemoimmunotherapy (CIT), and 16 (80%) had progressed within 24 months of frontline CIT. Remaining baseline characteristics are shown in Table 1. Representativeness of Study Participants is shown in Supplementary Table 1.
Table 1.
Baseline characteristics
| Patients (N=20) | Number (%); Median [Range] |
|---|---|
| Age | 61 [27–85] |
| Self-reported race/ethnicity: White | 13 (65) |
| Male | 10 (50) |
| Hemoglobin (g/dL) | 12.8 [9.2–15.2] |
| β2-microglobulin (mg/L) | 2.3 [0.8–6.2] |
| Lactate dehydrogenase (U/L) | 222 [129–338] |
|
Follicular lymphoma (FL)
Marginal zone lymphoma (MZL) Mantle cell lymphoma Richter syndrome |
15 (75) 3 (15) 1 (5) 1 (5) |
| Indolent B-NHL (FL and MZL) | 18 (90) |
| FL Grade 3A | 3/15 (20) |
| Bone marrow, involved | 4 (20) |
| B-symptoms, present | 3 (15) |
| Ann Arbor Stage III-IV | 18 (90) |
| Involved nodal areas (n) | 3 [1–5] |
| Largest lymph node (cm) | 2.9 [1.5–5.6] |
| Extra-nodal disease, present | 11 (55) |
| SUVmax | 15.8 [3.9–53.7] |
|
FLIPI score, low Intermediate high |
3/15 (20) 3/15 (20) 9/15 (60) |
|
FLIPI-2 score, low Intermediate high |
4/15 (27) 7/15 (46) 4/15 (27) |
|
PRIMA PI, low Intermediate high |
11/15 (73) 3/15 (20) 1/15 (7) |
| Previous systemic therapies (n) | 1 [1–3] |
| Previous anti-CD20 antibody | 20 (100) |
| Previous chemotherapy | 13 (72) |
| Previous POD24 | 16 (80) |
SUV, standardized uptake volume; FLIPI, follicular lymphoma international prognostic index; PI, prognostic index; POD24, progression of disease within 24 months
The median number of delivered cycles of ER2 was 12 (range, 6–12). Twelve patients received the planned 12 cycles; 3 patients stopped earlier due to progressive disease, 3 because of toxicity/other medical conditions (COVID-19 infection, lenalidomide-associated myocarditis, and accidental diagnosis of gastric neuroendocrine carcinoma); and 2 due to patient’s choice, once achieving CR after 6 cycles. Eight (40%) patients experienced a cycle delay, mainly due to transient and low-grade liver function test (LFT) elevation. Toxicity-related dose reduction in lenalidomide was required in 6 (30%) patients, and 1 discontinued ER2 after 1 cycle due to biopsy-proven lenalidomide-associated myocarditis (this was not considered a DLT due to lack of macrophage/histiocyte infiltration on tissue biopsy, and attributed to lenalidomide use). The toxicities that led to dose reduction of lenalidomide included G3 LFT elevation (1), G2 musculoskeletal pain (1), and creatinine clearance < 60 mL/min (4). None required toxicity-related dose reduction or discontinuation of evorpacept nor rituximab.
Safety and tolerability
Three patients were treated at DL1, 17 at DL2, and no DLT was observed. The most frequently reported AEs of any grade during treatment were neutropenia (55%), infections (30%), and ALT increase (15%). While most patients (85%) experienced neutropenia of any grade, 55% needed growth factor support, a median of 2 times (range, 1–5), and none had neutropenic fever. Overall, infections of any grade were experienced by 53% of patients throughout the treatment, including COVID-19 in 25% of patients. Anemia occurred in 70% of patients, was G1–2 in 60%, and transient (trends of hemoglobin during treatment during treatment are shown in Supplementary Figure 1).
LFT elevation of any grade included alanine aminotransferase (ALT) elevation (75%) and aspartate aminotransferase (AST) elevation (70%), with only low-grade bilirubin elevation observed (20%). Given the low-grade and transient nature of majority of observed LFT elevations, likely due to hepatic macrophages activation, protocol was amended to allow continuation of treatment for G1 ALT and/or AST increase. All cases resolved within 2 weeks, and only 1 patient required dose reduction in lenalidomide due to a second recurrence of grade ≥ 3 (trends of ALT and AST during treatment during treatment are shown in Supplementary Figure 1).
Skin rash of any grade was observed in 50% of patients. It was loco-regional and was managed with topical corticosteroids, oral antihistamines, and oral lysine in all cases. No patient required use of systemic corticosteroids nor dose reduction of lenalidomide due to skin rash.
A single case of myocarditis was observed after 1 cycle, and tissue biopsy showed lymphocytic infiltrate, with no macrophage nor histocytes, compatible with lenalidomide-associated myocarditis. Treatment was subsequently discontinued, and the patient remains in remission 2 years later, with complete recovery of cardiac function. Two patients developed a second primary malignancy (localized prostate adenocarcinoma and localized gastric neuroendocrine carcinoma), deemed not related to treatment. In both cases, lesions were visible on PET-CT before initiation of treatment, thought to represent lymphoma, and were biopsied subsequently due to persistence despite overall lymphoma resolution. Additional details regarding incidence and grade for all AEs are shown in Table 2.
Table 2.
Treatment-emergent adverse events
| Patients (N=20) | Grade 1–2 | Grade 3–4 |
|---|---|---|
| Neutropenia | 6 (30) | 11 (55) |
| Other infections | 5 (23) | 6 (30) |
| ALT increased | 12 (60) | 3 (15) |
| Skin rash | 8 (40) | 2 (10) |
| Anemia | 12 (60) | 2 (10) |
| AST increase | 12 (60) | 2 (10) |
| ALP increased | 4 (20) | 1 (5) |
| Infusion-related reaction | 6 (30) | 1 (5) |
| Myocarditis | 0 (0) | 1 (5) |
| Fatigue | 13 (65) | 0 (0) |
| Thrombocytopenia | 10 (50) | 0 (0) |
| Creatinine increase | 10 (50) | 0 (0) |
| Musculo-skeletal pain | 9 (45) | 0 (0) |
| Constipation | 7 (35) | 0 (0) |
| Nausea | 5 (25) | 0 (0) |
| COVID infection | 5 (25) | 0 (0) |
| Hyponatremia | 4 (20) | 0 (0) |
| Diarrhea | 4 (20) | 0 (0) |
| Dizziness | 4 (20) | 0 (0) |
| Bilirubin increase | 4 (20) | 0 (0) |
| Peripheral neuropathy | 3 (15) | 0 (0) |
| Headache | 3 (15) | 0 (0) |
| Hypercalcemia | 3 (15) | 0 (0) |
| Xerostomia | 2 (10) | 0 (0) |
| Mucositis | 2 (10) | 0 (0) |
ALT, alanine transferase; AST, aspartate transferase; ALP, alkaline phosphatase
Efficacy
OR rate was 90% (95% confidence interval [CI] 68.3–98.8%) after 3 cycles, 85% (95% CI 62.1–96.8%) after 6 cycles, and 75% (95% CI 50.9–91.3%) at the end of treatment, with a best ORR of 90% (95% CI 68.3–98.8%). CR rate was 80% (95% CI 56.3–94.3%) after 3 cycles, 75% (95% CI 50.9–91.3%) after 6 cycles, and 70% (95% CI 45.7–88.1%) at the end of treatment, with a best CR rate of 80% (95% CI 56.3–94.3%) (Figure 1). Among the 18 patients with iNHL, best ORR was 94% (95% CI 72.7–99.9%) and best CR rate 83% (95% CI 58.6–96.4%). No improvement in response, including conversion from partial response to CR, was observed beyond 3 cycles. Only two patients were primary refractory, including a patient with RS and a patient with a maximum standardized uptake volume of 16 and lactate dehydrogenase levels twice above the upper limit of normal, raising clinical radiological concern for transformed FL (despite biopsy evidence of low-grade FL); responses were durable in all other patients (Figure 2). When comparing baseline characteristics shown in Table 1 between the 4 patients who did not achieve CR (which included the one patient with RS) to the 16 who did, the only variable showing significant difference was the fraction of patients exposed to 2 or more prior lines of systemic therapy (100% vs 18.8%, p=0.0072)
Figure 1. Response rates.
Response after 3, 6 and 12 cycles (or end of treatment). All patients were evaluable for response. Complete response is depicted in green, partial response in blue, and progressive disease in orange. No patients experienced stable disease.
Figure 2. Duration of response.
Months on treatment shown by individual patient. Complete response is depicted in green, partial response in blue, and progressive disease in orange. Patients 1 and 2 had Richter syndrome and suspected transformed follicular lymphoma, respectively.
After a median follow-up of 28 months (95% CI 24–30 months), 6 patients relapsed or progressed. At the most recent follow-up, median PFS was not reached, 2-year PFS rate was 69% (95% CI 51–93%) (Figure 3A). None of the 18 patients with iNHL experienced subsequent transformation into large B-cell lymphoma. Among the 6 patients who progressed, 2 died without additional treatment, and 4 were able to receive subsequent lines of therapy. Among these 4 patients, next line of therapy included: allogeneic anti-CD19 NK-CAR therapy (1), autologous anti-CD19 CAR T-cell therapy (2) and radiation (1). All patients have achieved CR and have not relapsed to date, including 1 patient died in CR of COVID-19 infection.
Figure 3. Progression-free and overall survival.
Kaplan-Meier of progression-free survival and overall survival since initiation of treatment. Solid lines represent median survival. Dotted lines represent 95% confidence interval. Median follow-up of 28 months (95% CI 24–30 months)
At most recent follow-up, 3 patients have died: 2 patients, primary refractory to ER2, died due to disease progression, and 1 due to COVID19 infection while in CR after subsequent cellular therapy. Median OS has not been reached and 2-year OS rate was 84% (95% CI 70–100%) (Figure 3B).
Correlative analyses
After quality control, tumoral tissue samples for scRNA-seq were available for 16 patients: 14 patients at baseline, 14 patients during cycle 1, and paired for 12 patients.
On clustering analysis, during treatment, a significant increase in T cells and macrophages was observed (Figure 4A–B). Among T-cells, CD4+ T cells, CD8+ T cells, and Tregs, along with proliferating fractions of T cells, significantly increased (Supplementary Figure 2A). Among myeloid cells, the total fraction of M1 and M2 macrophages remained unchanged, with a significant increase observed only for conventional dendritic cells (Supplementary Figure 2B).
Figure 4. Biological changes in the tumor immune microenvironment during treatment.
A. Fraction of immune cell subsets by single cell RNA sequencing and clustering at baseline (BL) and during cycle 1 (on treatment, OT) in 18 patients (samples matched in 12); in addition to B-cells, a significant change was observed for T cells and macrophages. B. Bar plot showing the fraction of each major cell type across BL and OT samples. A statistically significant increase was observed in T cells and macrophages at OT (**, p < 0.01), with no significant changes in other populations (ns). C. Gene Set Enrichment Analysis (GSEA) of macrophages in 12 paired samples, comparing OT to BL to identify dynamic changes in tissue-resident immune cells. The top 10 enriched pathways are shown. D. Differential gene expression analysis in macrophages from OT samples comparing 3 progressors to 11 non-progressors. The volcano plot highlights significantly upregulated and downregulated genes in progressive disease (adjusted p-value < 0.05, |log2 fold change| > 1), including key immune and inflammatory regulators.
On GSEA/pathway analysis, during treatment, macrophage pathways associated with cytokines, innate immune system, toll-like receptor signaling, endocytosis, phagocytosis, extracellular matrix interactions, and antigen presentation were upregulated (Figure 4C).
On DEGs and following GSEA/pathway analysis, when comparing 3 progressors to 11 non-progressors, macrophages showed upregulated genes of differentiation (CTSB), activation (B2M), polarization (SNX10) and pro-inflammatory responses (CALHM6, VIM, PSME1), and downregulated genes of anti-tumoral activity (EFHD2, ITGAX) and inter-cellular interaction (FOS, CST3, SAT1) during treatment (Figure 4D and Supplementary Table 2), but not at baseline. As elevated B2M can associate with NLRP3 inflammasome activation with IL-18 and other proinflammatory cytokines, we evaluated the NLRP3 inflammasome activation signature score and the expression of multiple cytokines, but the comparison between progressors and non-progressors within the macrophage compartment did not reveal statistically significant differences (Supplementary Figure 3A–B). Additional details about single cell analyses are provided in Supplementary Table 3 and Supplementary Figure 4A–D.
Discussion
Macrophages play a crucial role in the biology in iNHL. FL cell lines are dependent on the tumor myeloid microenvironment for propagation and survival in in-vitro models. (26,27) In addition, seminal gene expression profiling studies have shown that immune signatures enriched in macrophages are associated with poor prognosis in FL patients treated with chemotherapy.(28)
While macrophage phenotype can be extremely heterogeneous, the most adverse prognostic impact in FL is thought to be held by subsets of pro-tumoral macrophages with high expression of SIRPα.(29) In this regard, both in-vitro and in-vivo studies have demonstrated a promising therapeutic potential for the combination of CD47/SIRPα blockade and monoclonal antibodies, due to synergizing direct cytotoxicity and ADCP.(30) Despite this, the development of anti-CD47 blocking agents, including magrolimab and TTI-621, has been limited by hematological toxicity and short-lived responses.(31,32) In contrast, evorpacept possesses an inactive human immunoglobulin fragment crystallizable region (Fc), to minimize hematological toxicity, and a higher affinity CD47 blocking region, to increase efficacy. In the R2 combination with evorpacept, rituximab supplies the active Fc which enables lymphoma-specific targeting and ADCP.(7,8)
In our study, the combination of evorpacept to R2 showed to be safe and well tolerated. With all the limitations of interstudy comparison, rates of neutropenia and skin rash where comparable to those reported with R2 alone.(3) However, higher rates of G1–2 LFT elevation were observed with ER2, prompting a protocol amendment to allow continuation of lenalidomide with G1 increase in AST and/or ALT. Of interest, most elevations were transient, did not require dose modifications, and were mainly observed in responders. Importantly, the liver is one of the organs with the higher concentration of resident macrophages (Kupffer cells), and hence the low grade (G1–2) and transient LFT elevation observed in this study could be a surrogate marker of treatment activity.(33) No LFT elevation has been reported among the 22 patients with relapsed B-NHL treated in the phase 1b trial of magrolimab and rituximab nor among the 42 treated in the phase 1b trial of TTI-622.(31,32) While LFT elevation has been reported in 10% of patients with relapsed iNHL treated with lenalidomide and rituximab in the AUGMENT trial,(3) we hypothesize that its combination with evorpacept may further enhance macrophage activation and induce macrophage activation syndrome, with related liver function test abnormalities.(34)
Most importantly, ER2 showed a promising activity signal in iNHL patients. With the limitations related to sample size, phase I design and intrinsic differences across studies, CR rates were notably higher as compared to R2 alone, with a best CR rate 80% vs 34%; and more durable as compared to prior trials investigating CD47-blocking agents in B-NHL, with a median PFS not reached after a follow-up longer than 2 years. (3,31,32) Additionally, in our study, ER2 did not only increase anti-tumoral phagocytosis, but also induced a more anti-tumoral phenotype in tissue resident macrophages, predictive or response to therapy, and hence potentially explaining the observed increased activity.(35) It is important to notice that acquired resistance to this combination is suggested by our DEG analyses, with upregulation of alternative mechanisms relevant to macrophage biology, beyond mere phagocytosis, including polarization and proliferation, as shown in Supplementary Table 2. While a functional confirmation of our computational findings is needed, this suggests that targeting multiple macrophage checkpoints, beyond CD47, may result into higher efficacy for future combinations.(35)
The combination of R2 with other biological agents, including targeted agents (tafasitamab and polatuzumab), BTK inhibitors (acalabrutinib) and bispecific antibodies (epcoritamab) has shown equally promising efficacy.(33,36–38) While data from this and other studies mature, identifying an immunotherapy regimen that could show superior activity to CIT in the frontline setting for patients with FL remains a significant endpoint in the field.(39) Of note, in our study an immunosuppressive microenvironment associated with decreased response to ER2. As clinical studies increasingly demonstrate the negative impact of prior agents (such as bendamustine) on immune-based therapies, appropriate treatment selection and sequencing may improve outcomes in this patient population.(40) While anti-CD3/CD20 bispecific antibodies and autologous anti-CD19 CAR T-cell therapy yield a potential for cure in a fraction of patients with aggressive B-NHL, its curative potential remains to be proven in iNHL, CD19/CD20 negative relapses are increasingly observed, T-cell exhaustion is induced by prior lines of therapy, and the need for the clinical development of novel agents targeting macrophage biology is becoming much relevant
We acknowledge here the limitations of this study, including its single center nature, sample size, inclusion of only patients with non-bulky disease, and lack of viable progression samples to better characterize mechanisms of resistance to this combination.
In conclusion, the combination of evorpacept and R2 is well tolerated in patients with relapsed B-NHL, with similar toxicity and promising anti-tumoral activity as compared to historical data with R2 alone in iNHL. A phase 2 study investigating its efficacy in patients with previously untreated high tumor burden iNHL has completed enrollment. Due to lack of response improvement with treatment continuation beyond cycle 6, either with R2 alone in our prior experience or with the addition of evorpacept in this phase 1 study, completion of treatment after only 6 cycles in case of CR has been allowed in the phase 2 study.(4)
Supplementary Material
Statement of translational relevance.
In this phase I study, the addition of evorpacept (formerly known as ALX-148), a novel anti-CD47 monoclonal antibody, to standard immunotherapy with lenalidomide and rituximab was shown to be safe. In addition, it also demonstrated promising efficacy, with an increase in complete response rates in patients with indolent B-cell non-Hodgkin lymphoma from historical 34% to 83%, posing the basis for a phase 2 study in patients with previously untreated disease.
When performing single-cell RNA sequencing on pre-treatment and on-treatment tissue biopsies, not only phagocytosis, but also multiple other pathways of macrophage-mediated anti-tumoral activity were shown to be increased. In addition, expression of genes relevant to macrophage biology associated with response and resistance to this novel regimen, suggesting potential targets for future more effective combinations.
Acknowledgments
This research is supported in by ALX Oncology and by The University of Texas MD Anderson Cancer Center Support Grant from National Institutes of Health (P30 CA016672). The MD Anderson Lymphoma Tissue Bank was utilized in this study and is supported by KW Cares.
PS’s salary is supported by the Leukemia Lymphoma Society Scholar in Clinical Research Career Development Program, the Sabin Family Fellowship Award, and the Kite Gilead Scholar in Clinical Research Award.
Footnotes
Conflict of Interest Statement
PS is a consultant for Roche-Genentech, Abbvie-Genmab, Beigene, Ipsen, Kite/Gilead, AstraZeneca-Acerta, ADC Therapeutics, Sobi, and Incyte; he has received research funds from Sobi, AstraZeneca-Acerta, ALX Oncology, Kite Gilead and ADC Therapeutics.
SSN received research support from Kite/Gilead, BMS, Allogene, Precision Biosciences, Adicet Bio, and Sana Biotechnology; served as Advisory Board Member/Consultant for Kite/Gilead, Merck, Sellas Life Sciences, Athenex, Allogene, Incyte, Adicet Bio, BMS, Bluebird Bio, Fosun Kite, Sana Biotechnology, Caribou, Astellas Pharma, Morphosys, Janssen, Chimagen, ImmunoACT, Orna Therapeutics, Takeda, and Synthekine; has stock options from Longbow Immunotherapy, Inc; and has intellectual property related to cell therapy.
The remaining authors declare no significant conflicts of interest.
Scientific Category: Research Articles - Clinical Trials: Immunotherapy
References
- 1.Morschhauser F, Fowler NH, Feugier P, Bouabdallah R, Tilly H, Palomba ML, et al. Rituximab plus Lenalidomide in Advanced Untreated Follicular Lymphoma. N Engl J Med 2018;379(10):934–47 doi 10.1056/NEJMoa1805104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Morschhauser F, Nastoupil L, Feugier P, Schiano de Colella JM, Tilly H, Palomba ML, et al. Six-Year Results From RELEVANCE: Lenalidomide Plus Rituximab (R(2)) Versus Rituximab-Chemotherapy Followed by Rituximab Maintenance in Untreated Advanced Follicular Lymphoma. J Clin Oncol 2022;40(28):3239–45 doi 10.1200/JCO.22.00843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Leonard JP, Trneny M, Izutsu K, Fowler NH, Hong X, Zhu J, et al. AUGMENT: A Phase III Study of Lenalidomide Plus Rituximab Versus Placebo Plus Rituximab in Relapsed or Refractory Indolent Lymphoma. J Clin Oncol 2019;37(14):1188–99 doi 10.1200/JCO.19.00010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Strati P, Jain P, Johnson RJ, Forbes S, Feng L, Samaniego F, et al. Long-term follow-up of lenalidomide and rituximab as initial treatment of follicular lymphoma. Blood 2021;137(8):1124–9 doi 10.1182/blood.2020007994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Marques-Piubelli ML, Parra ER, Feng L, Soto LS, Gallardo M, Gouni S, et al. SIRPalpha+ macrophages are increased in patients with FL who progress or relapse after frontline lenalidomide and rituximab. Blood Adv 2022;6(11):3286–93 doi 10.1182/bloodadvances.2022007104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zheng Z, Wang JB, Sun R, Wang N, Weng XQ, Xu TY, et al. Dual targeting PD-L1 and 4–1BB to overcome dendritic cell-mediated lenalidomide resistance in follicular lymphoma. Signal Transduct Target Ther 2025;10(1):29 doi 10.1038/s41392-024-02105-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kauder SE, Kuo TC, Harrabi O, Chen A, Sangalang E, Doyle L, et al. ALX148 blocks CD47 and enhances innate and adaptive antitumor immunity with a favorable safety profile. PLoS One 2018;13(8):e0201832 doi 10.1371/journal.pone.0201832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kim TY, Yoon MS, Hustinx H, Sim J, Wan HI, Kim H. Assessing and mitigating the interference of ALX148, a novel CD47 blocking agent, in pretransfusion compatibility testing. Transfusion 2020;60(10):2399–407 doi 10.1111/trf.16009. [DOI] [PubMed] [Google Scholar]
- 9.Lakhani NJ, Chow LQM, Gainor JF, LoRusso P, Lee KW, Chung HC, et al. Evorpacept alone and in combination with pembrolizumab or trastuzumab in patients with advanced solid tumours (ASPEN-01): a first-in-human, open-label, multicentre, phase 1 dose-escalation and dose-expansion study. Lancet Oncol 2021;22(12):1740–51 doi 10.1016/S1470-2045(21)00584-2. [DOI] [PubMed] [Google Scholar]
- 10.Kim TM, Lakhani NJ, Soumerai J, Kamdar M, Gainor JF, Messersmith W, et al. Evorpacept plus rituximab for the treatment of relapsed or refractory non-Hodgkin lymphoma: results from the phase I ASPEN-01 study. Haematologica 2025. doi 10.3324/haematol.2024.286208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Brice P, Bastion Y, Lepage E, Brousse N, Haioun C, Moreau P, et al. Comparison in low-tumor-burden follicular lymphomas between an initial no-treatment policy, prednimustine, or interferon alfa: a randomized study from the Groupe d’Etude des Lymphomes Folliculaires. Groupe d’Etude des Lymphomes de l’Adulte. J Clin Oncol 1997;15(3):1110–7 doi 10.1200/JCO.1997.15.3.1110. [DOI] [PubMed] [Google Scholar]
- 12.Solal-Celigny P, Roy P, Colombat P, White J, Armitage JO, Arranz-Saez R, et al. Follicular lymphoma international prognostic index. Blood 2004;104(5):1258–65 doi 10.1182/blood-2003-12-4434. [DOI] [PubMed] [Google Scholar]
- 13.Bachy E, Maurer MJ, Habermann TM, Gelas-Dore B, Maucort-Boulch D, Estell JA, et al. A simplified scoring system in de novo follicular lymphoma treated initially with immunochemotherapy. Blood 2018;132(1):49–58 doi 10.1182/blood-2017-11-816405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Federico M, Bellei M, Marcheselli L, Luminari S, Lopez-Guillermo A, Vitolo U, et al. Follicular lymphoma international prognostic index 2: a new prognostic index for follicular lymphoma developed by the international follicular lymphoma prognostic factor project. J Clin Oncol 2009;27(27):4555–62 doi 10.1200/JCO.2008.21.3991. [DOI] [PubMed] [Google Scholar]
- 15.Cheson BD, Fisher RI, Barrington SF, Cavalli F, Schwartz LH, Zucca E, et al. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol 2014;32(27):3059–68 doi 10.1200/JCO.2013.54.8800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wolf FA, Angerer P, Theis FJ. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol 2018;19(1):15 doi 10.1186/s13059-017-1382-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wolock SL, Lopez R, Klein AM. Scrublet: Computational Identification of Cell Doublets in Single-Cell Transcriptomic Data. Cell Syst 2019;8(4):281–91 e9 doi 10.1016/j.cels.2018.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sturm G 2025. infercnvpy: Scanpy plugin to infer copy number variation (CNV) from single-cell transcriptomics data. <https://infercnvpy.readthedocs.io/en/latest>.
- 19.Nieto P, Elosua-Bayes M, Trincado JL, Marchese D, Massoni-Badosa R, Salvany M, et al. A single-cell tumor immune atlas for precision oncology. Genome Res 2021;31(10):1913–26 doi 10.1101/gr.273300.120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.2005 Human Protein Atlas. Version 24.0. <https://www.proteinatlas.org>.
- 21.Hu C, Li T, Xu Y, Zhang X, Li F, Bai J, et al. CellMarker 2.0: an updated database of manually curated cell markers in human/mouse and web tools based on scRNA-seq data. Nucleic Acids Res 2023;51(D1):D870–D6 doi 10.1093/nar/gkac947. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Szabo PA, Levitin HM, Miron M, Snyder ME, Senda T, Yuan J, et al. Single-cell transcriptomics of human T cells reveals tissue and activation signatures in health and disease. Nat Commun 2019;10(1):4706 doi 10.1038/s41467-019-12464-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Fang Z, Liu X, Peltz G. GSEApy: a comprehensive package for performing gene set enrichment analysis in Python. Bioinformatics 2023;39(1) doi 10.1093/bioinformatics/btac757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005;102(43):15545–50 doi 10.1073/pnas.0506580102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Liberzon A, Subramanian A, Pinchback R, Thorvaldsdottir H, Tamayo P, Mesirov JP. Molecular signatures database (MSigDB) 3.0. Bioinformatics 2011;27(12):1739–40 doi 10.1093/bioinformatics/btr260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Umetsu DT, Esserman L, Donlon TA, DeKruyff RH, Levy R. Induction of proliferation of human follicular (B type) lymphoma cells by cognate interaction with CD4+ T cell clones. J Immunol 1990;144(7):2550–7. [PubMed] [Google Scholar]
- 27.Johnson PW, Watt SM, Betts DR, Davies D, Jordan S, Norton AJ, Lister TA. Isolated follicular lymphoma cells are resistant to apoptosis and can be grown in vitro in the CD40/stromal cell system. Blood 1993;82(6):1848–57. [PubMed] [Google Scholar]
- 28.Dave SS, Wright G, Tan B, Rosenwald A, Gascoyne RD, Chan WC, et al. Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells. N Engl J Med 2004;351(21):2159–69 doi 10.1056/NEJMoa041869. [DOI] [PubMed] [Google Scholar]
- 29.Chen YP, Kim HJ, Wu H, Price-Troska T, Villasboas JC, Jalali S, et al. SIRPα expression delineates subsets of intratumoral monocyte/macrophages with different functional and prognostic impact in follicular lymphoma. Blood Cancer J 2019;9(10):84 doi 10.1038/s41408-019-0246-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Chao MP, Alizadeh AA, Tang C, Myklebust JH, Varghese B, Gill S, et al. Anti-CD47 antibody synergizes with rituximab to promote phagocytosis and eradicate non-Hodgkin lymphoma. Cell 2010;142(5):699–713 doi 10.1016/j.cell.2010.07.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Advani R, Flinn I, Popplewell L, Forero A, Bartlett NL, Ghosh N, et al. CD47 Blockade by Hu5F9-G4 and Rituximab in Non-Hodgkin’s Lymphoma. N Engl J Med 2018;379(18):1711–21 doi 10.1056/NEJMoa1807315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ansell SM, Maris MB, Lesokhin AM, Chen RW, Flinn IW, Sawas A, et al. Phase I Study of the CD47 Blocker TTI-621 in Patients with Relapsed or Refractory Hematologic Malignancies. Clin Cancer Res 2021;27(8):2190–9 doi 10.1158/1078-0432.CCR-20-3706. [DOI] [PubMed] [Google Scholar]
- 33.Dixon LJ, Barnes M, Tang H, Pritchard MT, Nagy LE. Kupffer cells in the liver. Compr Physiol 2013;3(2):785–97 doi 10.1002/cphy.c120026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Billiau AD, Roskams T, Van Damme-Lombaerts R, Matthys P, Wouters C. Macrophage activation syndrome: characteristic findings on liver biopsy illustrating the key role of activated, IFN-γ-producing lymphocytes and IL-6- and TNF-α-producing macrophages. Blood 2005;105(4):1648–51 doi 10.1182/blood-2004-08-2997. [DOI] [PubMed] [Google Scholar]
- 35.Mantovani A, Marchesi F, Malesci A, Laghi L, Allavena P. Tumour-associated macrophages as treatment targets in oncology. Nat Rev Clin Oncol 2017;14(7):399–416 doi 10.1038/nrclinonc.2016.217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Diefenbach C, Kahl BS, McMillan A, Briones J, Banerjee L, Cordoba R, et al. Polatuzumab vedotin plus obinutuzumab and lenalidomide in patients with relapsed or refractory follicular lymphoma: a cohort of a multicentre, single-arm, phase 1b/2 study. Lancet Haematol 2021;8(12):e891–e901 doi 10.1016/S2352-3026(21)00311-2. [DOI] [PubMed] [Google Scholar]
- 37.Strati P, Agajanian R, Lossos IS, Coleman M, Kridel R, Wood A, et al. Acalabrutinib in combination with rituximab and lenalidomide in patients with relapsed or refractory follicular lymphoma: Results of the phase 1b open-label study (ACE-LY-003). Br J Haematol 2025;206(3):887–98 doi 10.1111/bjh.19951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Falchi L, Sureda Balari A, Leppä S, Vermaat JSP, Nijland M, Christensen JH, et al. Fixed-Duration Epcoritamab + R2 Drives Deep and Durable Responses in Patients with Relapsed or Refractory Follicular Lymphoma: 2-Year Follow-up from Arm 2 of the Epcore NHL-2 Trial. Blood 2024;144(Supplement 1):342- doi 10.1182/blood-2024-200262.39023871 [DOI] [Google Scholar]
- 39.Zhuang TZ, Zhang C, Strati P. SOHO State of the Art Updates and Next Questions | Novel Immunotherapy Combinations for the Treatment of Indolent B-Cell Lymphoma. Clin Lymphoma Myeloma Leuk 2025. doi 10.1016/j.clml.2025.01.021. [DOI] [PubMed] [Google Scholar]
- 40.Neelapu SS, Chavez JC, Sehgal AR, Epperla N, Ulrickson M, Bachy E, et al. Three-year follow-up analysis of axicabtagene ciloleucel in relapsed/refractory indolent non-Hodgkin lymphoma (ZUMA-5). Blood 2024;143(6):496–506 doi 10.1182/blood.2023021243. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data reported in this article are not publicly available, as they are either confidential or proprietary. To the extent allowed, the authors will provide access to de-identified participant-level data underlying the data presented in this article to researchers who provide a methodologically sound proposal for academic purposes to interpret, verify and extend research in the article that does not violate privacy, data encumbrance, intellectual property or other legal, regulatory or contractual confidentiality obligations, immediately after article publication. Data provided will be subject to a data use agreement. Researchers should contact the corresponding author when applying for data access. Response to external data requests will be within a reasonable timeframe of a few weeks to months depending on the nature of the request. Use of data will be restricted to agreed purpose.