Metabolic adaptation towards glycolysis supports resistance to neoadjuvant chemotherapy in early triple negative breast cancers

Abstract

Background: Neoadjuvant chemotherapy (NAC) is the standard of care for patients with early-stage triple negative breast cancers (TNBC). However, more than half of TNBC patients do not achieve a pathological complete response (pCR) after NAC, and residual cancer burden (RCB) is associated with dismal long-term prognosis. Understanding the mechanisms underlying differential treatment outcomes is therefore critical to limit RCB and improve NAC efficiency.

Methods: Human TNBC cell lines and patient-derived organoids were used in combination with real-time metabolic assays to evaluate the effect of NAC (paclitaxel and epirubicin) on tumor cell metabolism, in particular glycolysis. Diagnostic biopsies (pre-NAC) from patients with early TNBC were analyzed by bulk RNA-sequencing to evaluate the predictive value of a glycolysis-related gene signature.

Results: Paclitaxel induced a consistent metabolic switch to glycolysis, correlated with a reduced mitochondrial oxidative metabolism, in TNBC cells. In pre-NAC diagnostic biopsies from TNBC patients, glycolysis was found to be upregulated in non-responders. Furthermore, glycolysis inhibition greatly improved response to NAC in TNBC organoid models.

Conclusions: Our study pinpoints a metabolic adaptation to glycolysis as a mechanism driving resistance to NAC in TNBC. Our data pave the way for the use of glycolysis-related genes as predictive biomarkers for NAC response, as well as the development of inhibitors to overcome this glycolysis-driven resistance to NAC in human TNBC patients.

Keywords: Early triple negative breast cancer; Glycolysis; Metabolism; Neoadjuvant chemotherapy; Organoids; Therapy resistance.

Fig. 1

Early TNBC cell lines exhibit different basal glycolytic activities. A, B Growth of HCC1143, HCC38 and HCC1937 TNBC cell lines after treatment with increasing concentrations of epirubicin (A) or paclitaxel (B) for 72 h. C Calculated IC₅₀ values and 95% confidence intervals for epirubicin and paclitaxel in TNBC cell lines. D–E Growth of HCC1143, HCC38 and HCC1937 TNBC cell lines after treatment with IC₅₀ of epirubicin (D) or paclitaxel (E) for 24 h and then for 48 h upon drug withdrawal (washout). F Extracellular acidification rates (ECAR) in HCC1143, HCC38 and HCC1937 cells upon sequential treatment with 10 mM glucose, 1 µM oligomycin and 50 mM 2-deoxyglucose (2-DG). G–H Glucose-dependent (G) and maximal ECAR (H) in HCC1143, HCC38 and HCC1937 cells. I–J Glucose consumption (I) and lactate secretion (J) in HCC1143, HCC38 and HCC1937 TNBC cell lines. K Representative immunoblotting for HK2, GAPDH and LDHA in HCC1143, HCC38 and HCC1937 cells. L Quantification of LDHA protein levels in HCC1143, HCC38 and HCC1937 cells. Data are plotted as the means ± SEM from n = 3–6 cultures, performed each time with ≥ 3 technical replicates (A, B, D–J and L). Significance was determined by one-way ANOVA with Tukey’s multiple comparison test (G–J and L). *p < 0.05; ***p < 0.001; ns, not significant

Fig.  2

Paclitaxel and epirubicin induce opposite effects on glycolytic activity in TNBC cells. A, B Glucose consumption (A) and lactate secretion (B) in HCC1143 cells upon treatment with 9 nM paclitaxel or 600 nM epirubicin for 72 h. C, D ECAR profile upon sequential treatment with 10 mM glucose, 1 µM oligomycin and 50 mM 2-DG (C) and glucose-dependent ECAR (D) in HCC1143 cells treated with 9 nM paclitaxel or 600 nM epirubicin for 24 h. E, F Glucose consumption (E) and lactate secretion (F) in HCC38 cells upon treatment with 4 nM paclitaxel or 25 nM epirubicin for 72 h. G, H ECAR profile upon sequential treatment with 10 mM glucose, 1 µM oligomycin and 50 mM 2-DG (G) and glucose-dependent ECAR (H) in HCC38 cells treated with 4 nM paclitaxel or 25 nM epirubicin for 24 h. I, J Representative immunoblotting (I) and quantification (J) for HK2, GAPDH and LDHA in HCC1143 and HCC38 cells upon treatment with 9 or 4 nM paclitaxel, respectively, and 600 or 25 nM epirubicin, respectively for 72 h. Data are plotted as the means ± SEM from n = 2–6 cultures, performed each time with ≥ 3 technical replicates (A–H, and J). Significance was determined by one-way ANOVA (A, B, D–F, and H) or two-way ANOVA (J) with Tukey’s multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant

Fig. 3

Mitochondrial oxidative metabolism is decreased in TNBC cells upon treatment with chemotherapy. A–C Oxygen consumption rates (OCR) upon sequential treatment with 1 µM oligomycin, 2 µM FCCP and 0.5 µM rotenone/antimycin A (A), and OCR at basal (B) and maximal levels (C) in HCC1143 cells treated with 9 nM paclitaxel or 600 nM epirubicin for 24 h. D–I OCR values in HCC38 (D–F) and HCC1937 (G–I) in the same experimental conditions than indicated for HCC1143 cells (except that 4 and 300 nM paclitaxel as well as 25 and 800 nM epirubicin were used for HCC38 and HCC1937 cells, respectively). Data are plotted as the means ± SEM from n = 6 cultures, performed each time with ≥ 3 technical replicates (A–I). Significance was determined by one-way ANOVA (B, C, E, F, and H, I) with Tukey’s multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant

Fig. 4

Basal glycolytic activity is heterogeneous in patient-derived TNBC organoid models. A Schematic representation of the experimental workflow for organoid initiation and culturing from treatment-naive TNBC patients. B, C Representative brightfield pictures B and molecular characteristics C of patient-derived TNBC organoid models BCO17 and IDC031. Scale bar: 50 µm. D Representative immunohistochemical staining for ER, PR, HER2 and Ki67 in BCO17 organoids and matching primary tumor tissue. Scale bars: 17.3 µm (organoids); 50 µm (primary tissue). E–F Glucose consumption (E) and lactate secretion (F) in BCO17 and IDC031 organoids. G, H ECAR profile upon sequential treatment with 20 mM glucose, 1 µM oligomycin and 50 mM 2-DG (G) and glucose-dependent ECAR (H) in BCO17 and IDC031 organoids. Data are plotted as the means ± SEM from n = 3–6 cultures, performed each time with ≥ 3 technical replicates (E–H). Significance was determined by Student’s t test (E, F, and G). *p < 0.05; ***p < 0.001

Fig. 5

Paclitaxel induces a metabolic shift from oxidative metabolism towards glycolysis in patient-derived TNBC organoids. A, B Growth of BCO17 and IDC031 TNBC organoids after treatment with increasing concentrations of paclitaxel A or epirubicin B for 7 days. C Calculated IC₅₀ values and 95% confidence intervals for epirubicin and paclitaxel in TNBC organoids. D, E Glucose consumption (D) and lactate secretion (E) in BCO17 and IDC031 TNBC organoids treated with 10 nM paclitaxel or 125 nM epirubicin for 7 days. F–G Glucose-dependent (F) and maximal ECAR (G) in BCO17 and IDC031 organoids treated with 10 nM paclitaxel or 125 nM epirubicin for 7 days. Data are plotted as the means ± SEM from n = 3–6 cultures, performed each time with ≥ 3 technical replicates (A, B, and D–G). Significance was determined by one-way ANOVA with Tukey’s multiple comparison test (D–G). *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant

Fig. 6

Increased glycolysis correlates with NAC resistance in patient-derived TNBC clinical specimens. A Schematic representation of the clinical workflow for collection of diagnostic biopsies from patients with early-stage TNBC treated with NAC, recruited in this study. B RNA-sequencing datasets, from early TNBC human patients, used for gene set enrichment analysis (GSEA) between responders (RCB 0/I) and non-responders (RCB II/III) to NAC. C–F Pathways significantly up- and down-regulated (C, E) and individual enrichment plots for glycolysis hallmark (D, F) from GSEA of RNA-seq data in NAC-responding and non-responding TNBC specimens (GSE240671 and GSE123845 datasets for C, D and E, F panels, respectively)

Fig. 7

Glycolysis inhibition improves response to NAC in 3D TNBC cell models. A Colony-forming capacity of HCC38 and HCC1143 cells treated with 12.5 mM 2-DG alone or in combination with 0.04 nM and 0.12 nM paclitaxel respectively, for 3 days. B, C Growth of 3D spheroids initiated from HCC38 and HCC1937 cells treated with 50 mM 2-DG alone or in combination with 22 and 37.8 µM paclitaxel respectively B or 624 nM and 617.5 nM epirubicin respectively C for 3 days. D, E Growth of BCO17 and IDC031 TNBC organoids after treatment with 50 mM 2-DG alone or in combination with 50 nM paclitaxel D or 100 nM epirubicin E for 7 days. Data are plotted as the means ± SEM from n = 2–3 cultures, performed each time with ≥ 2 technical replicates (A–E). Significance was determined by two-way ANOVA with Tukey’s multiple comparison test (A–E). *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant