STING agonism reprograms tumor-associated macrophages and overcomes resistance to PARP inhibition in BRCA1-deficient models of breast cancer

Abstract

PARP inhibitors (PARPi) have drastically changed the treatment landscape of advanced ovarian tumors with BRCA mutations. However, the impact of this class of inhibitors in patients with advanced BRCA-mutant breast cancer is relatively modest. Using a syngeneic genetically-engineered mouse model of breast tumor driven by Brca1 deficiency, we show that tumor-associated macrophages (TAMs) blunt PARPi efficacy both in vivo and in vitro. Mechanistically, BRCA1-deficient breast tumor cells induce pro-tumor polarization of TAMs, which in turn suppress PARPi-elicited DNA damage in tumor cells, leading to reduced production of dsDNA fragments and synthetic lethality, hence impairing STING-dependent anti-tumor immunity. STING agonists reprogram M2-like pro-tumor macrophages into an M1-like anti-tumor state in a macrophage STING-dependent manner. Systemic administration of a STING agonist breaches multiple layers of tumor cell-mediated suppression of immune cells, and synergizes with PARPi to suppress tumor growth. The therapeutic benefits of this combination require host STING and are mediated by a type I IFN response and CD8⁺ T cells, but do not rely on tumor cell-intrinsic STING. Our data illustrate the importance of targeting innate immune suppression to facilitate PARPi-mediated engagement of anti-tumor immunity in breast cancer.

Fig. 1. Brca1-deficient breast tumors have a modest response to olaparib in vivo with immune-suppressive TAMs.

a Generation of a syngeneic GEMM of Brca1-deficient breast tumors by intraductal injection of adenovirus expressing Cre recombinase (Ad-Cre) directly into the lumen of mammary glands. b Tumor-free survival of Brca1^L/L Trp53^L/L mice with or without intraductal injection of Ad-Cre (n = 6). c Tumor growth of Brca1^−/− Trp53^−/− (BP) allografts in FVB mice treated with olaparib or anti-PD-1 as monotherapy or in combination. Control, n = 8; anti-PD-1, n = 8; olaparib, n = 6; olaparib + anti-PD-1, n = 10. d Flow cytometry analysis of TCRβ⁺ T cells and tumor-associated macrophages (TAMs; CD45⁺ CD11b⁺ F4/80⁺) from BP tumors in FVB mice after 21 days of olaparib treatment. TCRβ⁺ (control), n = 6; TCRβ⁺ (olaparib), n = 8; TAMs (control), n = 6; TAMs (olaparib), n = 6. TAMs were further analyzed to identify M1-like (MHC-II^high CD206⁻) and M2-like (MHC-II^low CD206⁺) polarization phenotypes (n = 6 for each group). Each dot represents data from a single tumor. e Diagram of workflow for (f) and (g). TAMs (7AAD⁻ CD45⁺ CD11b⁺ F4/80⁺) were sorted from BP breast tumors and co-cultured with splenic CD8⁺ T cells isolated from naïve mice for 2 days. f Analysis of cytokine production by CD8⁺ T cells co-cultured with TAMs (CD8⁺ T cells, n = 4; CD8⁺ T cells+TAMs, n = 8). g Analysis of the proportion of effector cells (CD44^high CD62L^low) and surface expression of CD25 of CD8⁺ T cells co-cultured with TAMs (CD8⁺ T cells, n = 4; CD8⁺ T cells+TAMs, n = 8). Flow cytometry plot axes are displayed in logarithmic scale (f) and (g). Data are presented as mean ± SEM (c), (f), and (g), or median with quartiles (violin plots, d). Two-way analysis of variance (ANOVA) (c). Two-tailed unpaired t test or Mann–Whitney test (d), (f), and (g). ns not significant. Source data are provided as a Source Data file.

Fig. 2. BRCA1-deficient breast tumor cells induce M2-like macrophage polarization in vitro.

a Diagram of workflow. Top, mouse bone marrow-derived macrophages (BMDMs) co-cultured with tumor cells with or without olaparib treatment. Bottom, BMDMs were incubated with 50% conditioned media (CM) harvested from olaparib- or DMSO-treated tumor cells. TEMs, tumor cell-educated macrophages. b Flow cytometry analysis of BMDMs co-cultured with BP tumor cells with or without olaparib (5 μM) for 2 days. BMDMs (CD11b⁺) were plotted as CD206 versus MHC-II to identify M1-like (CD206^- MHC-II^high) and M2-like (CD206⁺ MHC-II^low) polarization phenotypes (BMDMs, n = 3; BP/BMDMs, n = 7). Flow cytometry plot axes are displayed in logarithmic scale. c Heat map of gene expression for anti-tumor and pro-tumor genes in BMDMs incubated with DMSO vehicle control, olaparib (OL, 5 μM), 50% BP-CM, or 50% OL-treated BP (BP/OL)-CM for 24 h (n = 2 for each group). d RT-qPCR analysis of mouse BMDMs incubated with control medium, 50% BP-CM, or 50% BP/OL-CM for 24 h (Il6, n = 5; Il1b, n = 3; Cxcl1, n = 4). e RT-qPCR analysis of THP-1 human macrophages incubated with control medium, 50% tumor cell-CM, or 50% CM of olaparib-treated tumor cells for 24 h (MDA-MB-436, n = 5; HCC1937, n = 4). Data are presented as mean ± SEM. One-way ANOVA (b). Two-tailed unpaired t test or Mann–Whitney test (d) and (e). ns, not significant. Source data are provided as a Source Data file.

Fig. 3. TEMs/TAMs suppress olaparib-induced DNA damage in BRCA1-deficient breast tumor cells.

a Diagram of workflow for (b)–(f). Conditioned media from control macrophages (Mφ) or tumor-educated macrophages (TEMs) were added to tumor cell cultures, followed by olaparib treatment. b BP cells were stained using DAPI and an anti-dsDNA antibody after two days of olaparib treatment. The intensity of dsDNA fragments in the cytosol was quantified. Scale bar, 50 μm (control, n = 21 fields; olaparib, n = 26 fields; control macrophages, n = 10 fields; control macrophages + olaparib, n = 10 fields; TEMs, n = 14 fields; TEMs + olaparib, n = 10 fields examined over two independent experiments). c BP cells were stained with anti-H2AX phospho-Ser139 antibody and analyzed by flow cytometry. MFI, median florescence intensity (n = 3). d BP cells were analyzed for apoptosis (Annexin V⁺ 7-AAD⁻) (n = 4). e MDA-MB-436 cells were stained with anti-H2AX phospho-Ser139 antibody and analyzed by flow cytometry (n = 4). f MDA-MB-436 cells were analyzed for apoptosis (Annexin V⁺ 7-AAD⁻) (n = 4). g Apoptotic analyses of BP cells co-cultured with or without TAMs sorted from BP tumors, followed by three days of olaparib treatment (n = 4). Flow cytometry plot axes are displayed in logarithmic scale. h Schematic representation of the experiments for (i) and (j). i and j Analysis of TAMs (i) and γ-H2AX in CD45-negative cells (j) in BP tumors after two IP injections of anti-CSF1R antibody and 7 days of treatment with olaparib (OL) as monotherapy or in combination. Control, n = 4; OL, n = 6; anti-CSF1R, n = 4; OL + anti-CSF1R, n = 6. Each dot represents data from a single tumor. Data are presented as mean ± SEM (b)–(g) or median with quartiles (violin plots, i and j). One-way ANOVA. ns, not significant. Source data are provided as a Source Data file.

Fig. 4. STING agonists can reprogram M2-like TEMs/TAMs into an M1-like state.

a and b Mouse BMDMs were subjected to transcriptome analysis after treatment for 24 h with DMSO vehicle control, olaparib (OL, 5 μM), DMXAA (0.05 mg/mL), 50% BP-CM with or without DMXAA, or 50% BP/OL-CM with or without DMXAA. a Heat map of anti-tumor and pro-tumorigenic gene expression in BMDMs (n = 2 for each group). b Left, volcano plot showing the significance and magnitude of changes in gene expression of BMDMs treated with BP-CM/DMXAA compared to BP-CM/DMSO (n = 2 for each group). Statistical significance was calculated using a two-sided Wald test and adjusted for multiple testing using the Benjamini–Hochberg procedure. Right, top-ranked up-regulated and down-regulated gene ontology (GO) terms in BMDMs treated with BP-CM/DMXAA. Significance of enriched terms were adjusted using the Benjamini–Hochberg procedure for multi-testing. c Analysis of control macrophages (Mφ) and BP TEMs treated with or without DMXAA (0.05 mg/mL) for 2 days. M1-like (CD11b⁺ CD206^- MHC-II^high) to M2-like (CD11b⁺ CD206⁺ MHC-II^low) ratio was analyzed by flow cytometry (n = 4). d Analysis of M1 to M2 ratio of TAMs (sorted from BP tumors) treated with or without DMXAA (0.05 mg/mL) for 2 days (n = 3). e Diagram of workflow for (f-h). BMDMs isolated from wildtype (WT) or STING knockout (STING^-/-) C57/BL6J mice were incubated with control medium or 50% BP-CM to generate control naïve BMDMs and TEMs, respectively. Cells were then treated with or without DMXAA (0.05 mg/mL) for 2 days. f and g M1 to M2 ratio (f) and surface levels of the co-stimulatory molecule CD86 (g) were analyzed by flow cytometry (n = 4). h BP cells were co-cultured with or without WT TEMs or STING^−/− TEMs pre-treated with or without DMXAA (0.05 mg/mL), followed by 2 days of olaparib (5 μM) treatment. BP cells were then analyzed for apoptosis (Annexin V⁺ 7-AAD⁻; BP cells, n = 6; BP + WT TEMs, n = 3; BP + STING^−/− TEMs, n = 3). i Expression of M2 (CD163) and M1 (CD86) markers in control THP-1 macrophages or MDA-MB-436 tumor cell-educated THP-1 macrophages (TEMs-436) treated with or without ADU-S100 (10 μM) for two days (n = 3). Data are presented as mean ± SEM. One-way ANOVA (c) and (f)–(i). Two-tailed unpaired t test (d). ns, not significant. Source data are provided as a Source Data file.

Fig. 5. STING agonists improve therapeutic response of orthotopic BP tumors to olaparib in syngeneic immunocompetent mice in vivo.

a BP tumor growth in FVB mice treated with olaparib (50 mg/kg, IP, QD) or intratumoral (IT) injections of DMXAA (10 mg/kg, one dose per week for 3 weeks [total of 3 doses]) as monotherapy or in combination. Control, n = 8; olaparib, n = 9; DMXAA, n = 6; olaparib + DMXAA, n = 7. b and c Analysis of BP tumors after 21 days of treatment for effector cytokine production by intratumoral CD8⁺ T cells and CD4⁺ T cells. Flow cytometry plot axes are displayed in logarithmic scale (b). Control, n = 6; olaparib, n = 6; DMXAA, n = 4; olaparib + DMXAA, n = 4. Each dot represents data from a single tumor (c). Data are presented as mean ± SEM (a), or median with quartiles (violin plots, c). Two-way ANOVA (a). One-way ANOVA (c). ns, not significant. Source data are provided as a Source Data file.

Fig. 6. Systemic delivery of a STING agonist sensitizes STING-null BP tumors to olaparib in vivo.

a Western blots for STING and VINCULIN in CRISPR/Cas9 control and STING knockout BP tumor cells (BP-sgControl and BP-sgSTING). Representative blots of two independent experiments are shown. b CellTiter-Glo analysis showing cell viability of BP-sgControl and BP-sgSTING cells after 3 days of treatment with serial dilution of olaparib (n = 3). c Flow cytometry analysis of mouse BMDMs treated with DMSO vehicle control, olaparib (OL, 5 μM), 50% BP-sgSTING-CM, or 50% BP-sgSTING/OL-CM for two days (n = 3). d ELISA analysis of IFNβ in media from BP-sgControl or BP-sgSTING cells with or without 2 days of olaparib treatment (n = 7). e and f RT-qPCR analysis of Ccl5 (e) and Cxcl10 (f) in BP-sgControl and BP-sgSTING cells treated with or without olaparib for 2 days (BP-sgControl, n = 4; BP-sgSTING, n = 3). g Tumor growth (left) and survival (right) of BP-sgControl tumor-bearing FVB mice treated with olaparib (50 mg/kg, IP, QD), DMXAA (10 mg/kg, IP) or olaparib + DMXAA. Median survivals are shown in parentheses. Left, Control, n = 13; Olaparib, n = 7; DMXAA, n = 9; Olaparib + DMXAA, n = 14. Right, Control, n = 8; olaparib, n = 5; DMXAA, n = 6; olaparib + DMXAA, n = 9. h Tumor growth (left) and survival (right) of BP-sgSTING tumor-bearing FVB mice treated with olaparib (50 mg/kg, IP, QD), DMXAA (10 mg/kg, IP) or olaparib + DMXAA. Median survivals are shown in parentheses. Left, Control, n = 24; olaparib, n = 11; DMXAA, n = 9; olaparib + DMXAA, n = 19. Right, Control, n = 13; olaparib, n = 7; DMXAA, n = 6; olaparib + DMXAA, n = 11. i and j BP-sgControl (i) and BP-sgSTING (j) tumor growth in FVB mice treated with olaparib + DMXAA with or without anti-CD8 or anti-IFNAR1 neutralizing antibodies (n = 6 per condition). Data are presented as mean ± SEM. One-way ANOVA (c, e, and f). Two-tailed Mann–Whitney test (d). Two-way ANOVA for tumor growth (g)–(j). Log-rank Mantel–Cox test for survival (g and h). ns, not significant. Source data are provided as a Source Data file.

Fig. 7. Harnessing anti-tumor immunity with STING agonists overcomes immune suppression and resistance to PARP inhibition in BRCA1-deficient breast cancer.

BRCA1-deficient breast tumors elicit pro-tumorigenic macrophage polarization. In turn, these tumor-educated macrophages not only exhibit suppressive activity against T cells, but also attenuate PARPi-mediated synthetic lethality and the production of double-stranded DNA (dsDNA) fragments, thus diminishing the activation of the DNA sensing adaptor STING and rendering BRCA1-deficient breast tumors resistant to PARPi therapy (blue shading). Exogenous agonists of the STING pathway reprogram the macrophages and trigger innate immune activation of both macrophages and DCs, potentiating PARPi therapy to induce tumor cell DNA damage and an adaptative immune response that re-sensitizes tumors to PARPi therapy (orange shading).