Role of CD38 in anti-tumor immunity of small cell lung cancer

Abstract

Introduction: Immune checkpoint blockade (ICB) with or without chemotherapy has a very modest benefit in patients with small cell lung cancer (SCLC). SCLC tumors are characterized by high tumor mutation burden (TMB) and low PD-L1 expression. Therefore, TMB and PD-L1 do not serve as biomarkers of ICB response in SCLC. CD38, a transmembrane glycoprotein, mediates immunosuppression in non-small cell lung cancer (NSCLC). In this brief report, we highlight the potential role of CD38 as a probable biomarker of immunotherapy response in SCLC.

Methods: We evaluated the role of CD38 as a determinant of tumor immune microenvironment in SCLC with bulk and single-cell transcriptomic analyses and protein assessments of clinical samples and preclinical models, including CD38 in vivo blockade.

Results: In SCLC clinical samples, CD38 levels were significantly correlated with the gene expression of the immunosuppressive markers FOXP3, PD-1 and CTLA-4. CD38 expression was significantly enhanced after chemotherapy and ICB treatment in SCLC preclinical models and clinical samples. A combination of cisplatin/etoposide, ICB, and CD38 blockade delayed tumor growth compared to cisplatin/etoposide.

Conclusion: Our study provides a preliminary but important direction toward exploring CD38 as a potential biomarker of ICB response and CD38 blockade as a combination strategy for chemo-immunotherapy in SCLC.

Keywords: CD38; immune checkpoint blockade; lung cancer; resistance; small cell lung cancer.

Figure 1

Correlation with CD38 and immune markers in SCLC datasets. (A). Correlation with CD38 and FOXP3, CTLA-4, PD-1. [upper; Rudin et al., dataset (5), bottom; George et al., dataset (6)]. P values are calculated using Spearman rank correlation test. (B) Frequency of CD38 [log2 transform with a pseudo-count of 1)] on B cell, CD4⁺ T cell, regulatory T cell, CD8⁺ T cell, and NK cell in a single cell atlas of human SCLC (Treatment naïve patients). p values were calculated by one-way ANOVA and adjusted P values were by Turkey’s test. * p <0.0332, ** p <0.0021, *** p <0.0002.

Figure 2

Expression of CD38 by treatment with anti-PD-L1 antibody and anti-tumor effect of anti-CD38 antibody with anti-PD-L1 antibody in vivo. (A) Boxplots from bulk RNAseq analysis corresponding to the CD38 expression (B) Quantitative mRNA expression of CD38 by real-time RT-PCR analysis (C) Western blot showing expression of CD38 and actin (loading control)- in RPP (flank and orthotopic tumors) and RPM flank tumors, which were treated with vehicle or anti-PD-L1 antibody (300µg/body, once weekly). (D) Tumor growth curves (mean ± standard error [SE]) of IgG control, anti-PD-L1 antibody alone (200 µg/body, once weekly), anti-CD38 antibody (300 µg/body, once weekly), and anti-CD38 antibody plus anti-PD-L1 antibody groups in B6129F1 mice injected with RPP cells (SCLC cells with conditional loss of Trp53, p130, and Rb1; n = 6 in each group). P values were calculated by student-T test.

Figure 3

Change of expression of CD38 by chemotherapy and anti-tumor effect of anti-CD38 antibody with anti-PD-L1 antibody and chemotherapy. (A) CD38 expression on CD45⁺ total immune cell, CD4⁺ T cell, and CD8⁺ T cell in clinical specimen from the patients with SCLC. (B, C) Flow cytometry analysis was performed on H82, DMS114, and H187 cell lines treated with 0.3 or 3 µM cisplatin (upper) and 0.03 or 0.3 µM etoposide for 72 hours. (D) Tumor growth curves (mean ± standard error [SE]) of IgG control and vehicle, anti-PD-L1 antibody alone (200 µg/body, once weekly) and anti-CD38 antibody (300 µg/body, once weekly), cisplatin (CDDP) (2mg/kg) and etoposide (ETP) (3mg/kg), and anti-CD38 antibody, anti-PD-L1 antibody, CDDP and ETP groups in B6129F1 mice injected with RPP cells (SCLC cells with conditional loss of Trp53, p130, and Rb1; n = = 6 in each group). P values were calculated by Mann-Whiteney for (A), * p<0.0332 and student-T test for (D).