NCAPH expression is associated with poor evolution and response to therapy in luminal A breast cancer patients
NCAPH is a non-SMC subunit of the condensin I complex that is involved in correct chromosome segregation(29) (Fig. 1a). Elevated intratumoral levels of NCAPH RNA have recently been associated with poor BC evolution according to freely available databases (22). We confirmed an increase in NCAPH in invasive ductal breast carcinomas relative to normal mammary tissues in a public database(63) (Fig. 1b). Indeed, NCAPH expression was higher in all molecular BC subtypes except luminal A tumors when evaluated with GenExMiner 4.5 (Fig. 1c)(64).
The intrinsic subtype classification of BC and its associated prognosis have been improved by the application of specific gene signatures, such as PAM50 (Prediction Analysis of Microarray 50) (9). After assessing the expression of NCAPH, we identified patients with subtypes of good and poor BC evolution defined by the PAM50 in the KM Plotter database(65). Although NCAPH expression was weakest in the luminal A subtype, we found that patients with high levels of NCAPH RNA were associated with poor evolution (p = 4.8 x 10− 7). Paradoxically, basal-like tumors were associated with good clinical evolution (p = 0.0066) (Fig. 1d).
All these results indicate the existence of a subpopulation of luminal A tumors that have poor evolution, which can be distinguished by their high intratumoral levels of NCAPH. We evaluated NCAPH levels by immunohistochemistry (IHC) in a cohort of patients with luminal A tumors (Additional File 2: Table S1a), and again, we confirmed the higher NCAPH protein levels in patients with a poor evolution (presence of liver metastases) than in those who evolved well in the 10-year follow-up (Fig. 1e, f). Moreover, NCAPH levels were not associated with other tumor characteristics, such as grade, stage, histological subtype, or Ki-67 staining (Additional File 2: Table S1b). Significantly, high levels of NCAPH in luminal A tumors were associated with poor response to endocrine therapy and chemotherapy (Fig. 1g). This contrasted with other intrinsic tumor subtypes that responded to therapy independently of NCAPH (Fig. 1h), except for basal tumors, which tended to respond well to chemotherapy when NCAPH levels were high (Fig. 1i). Hence, high NCAPH expression in luminal A breast tumors appears to be associated with poor evolution and response to therapy, implying that NCAPH expression may help identify luminal A tumors with a worse prognosis.
Overexpression of NCAPH increases the tumorigenicity of luminal breast cancer cells
To examine the role of NCAPH in the pathogenesis of BC in greater depth, we generated an MCF7 luminal A BC cell line in which NCAPH overexpression could be induced in response to doxycycline (Fig. 2a, b). The induction of NCAPH overexpression in MCF7 cells enhanced their viability and proliferation (Fig. 2c-e), which was associated with higher levels of Cyclin D1 expression (Fig. 2f) and the generation of significantly more colonies in soft agar (Fig. 2g). Moreover, when NCAPH was overexpressed, more cells migrated (Fig. 2h). Overexpression of NCAPH in the MCF7 cell line also increased the proportion of cells with chromosomal instability (CIN), in which micronuclei and chromosomal bridges could be detected (Fig. 2i, j). Genomic instability induces genomic stress(66, 67), as demonstrated by the increase in ɤH2AX and pCHEK1 levels after NCAPH overexpression and exposure to H2O2 (Fig. 2k-m). Significantly, in human tumors, there was a strong positive correlation between NCAPH expression and markers of proliferation (Ki67) or genomic instability (H2AX and CHEK1) (Fig. 2n).
We used this doxycycline-inducible system to determine whether elevated NCAPH expression is responsible for the resistance of luminal tumors to treatment. Interestingly, NCAPH upregulation resulted in partial resistance to therapy in MCF7 cells (Fig. 2o-q). This resistance to treatment was associated with an increase in total and phosphorylated AKT levels (Fig. 2r), which might explain this behavior(68–70). We did not observe any differences in cell viability in the basal BT-549 cell line after inducing NCAPH expression or doxorubicin treatment (Additional File 3: Fig. S1a-c). Thus, it would be of interest to assess the prognosis of patients with basal BC and elevated intratumoral levels of NCAPH (Fig. 1d, l). Since our in vitro results indicated that NCAPH is involved in the pathogenesis of luminal BC, we focused our attention on the role of NCAPH in the pathogenesis of these tumors in vivo.
NCAPH overexpression induces mammary gland hyperplasia and breast cancer in mice
NCAPH overexpression was associated with poor BC evolution and poor response to therapy in patients with luminal A BC (Fig. 1d-g). To determine whether NCAPH is a driver of BC, we generated transgenic mice overexpressing NCAPH under the control of the mouse mammary tumor virus (MMTV) promoter (Fig. 3a), inducing NCAPH overexpression in the mammary gland (Fig. 3b, c). MMTV-NCAPH nulliparous mice developed breast tumors during a long-term follow-up, with 30% of mice developing breast cancer after 120 weeks (Fig. 3d). Hence, the overexpression of NCAPH could drive BC development. Oncogenes driven by the MMTV promoter are overexpressed during pregnancy because of promoter induction by gravidity hormones, which results in more aggressive tumors(71). Indeed, the overexpression of NCAPH under the control of the MMTV promoter in this transgenic line led to breast tumor development with a shorter latency after pregnancy (Fig. 3e). Both MMTV-NCAPH#1 and #2 lines developed breast tumors after pregnancy, with no significant difference in tumor incidence (Fig. 3f).
Tumors generated by NCAPH overexpression had a range of histopathological features, including breast adenocarcinomas with papillary differentiation, squamous differentiation, or a mesenchymal pattern (Fig. 3g-j and Additional File 2: Table S2).
The distinct histopathological features observed could arise from high levels of NCAPH in the mammary gland, leading to exaggerated genomic instability (Fig. 2l-n) and the generation of secondary oncogenic events that target a variety of tumor differentiation pathways(72).
Interestingly, although most nulliparous MMTV-NCAPH mice did not develop breast tumors after two years, they did show a significant increase in their ductal epithelial components. Indeed, MMTV promoter-driven overexpression of NCAPH produced hypertrophic mammary glands with marked ductal hyperplasia. The ducts were formed by a normal, single row of epithelial cells with no atypia and a benign aspect (Fig. 3k); however, there was a substantial increase in the number of ducts, increasing the total parietal ductal area (Fig. 3l, m). Notably, two additional mice that were not included in the comparison exhibited massive ductal and stromal hyperplasia in the mammary gland and the absence of fatty tissue (Fig. 3n).
Together, these results indicate that NCAPH is oncogenic in luminal breast tumors.
High intratumoral levels of NCAPH are associated with poor evolution in luminal HER2+ tumors
NCAPH levels were not associated with changes in the evolution of HER2-enriched tumors as defined by PAM50 (Fig. 1d, h). HER2-enriched tumors do not include HER2+ luminal tumors, and PAM50 does not explicitly define this tumor subtype(73). Therefore, we defined these tumors based on their expression of ER and HER2 in the KM plotter database(65) to establish a possible association between NCAPH expression and evolution. High NCAPH levels were associated with poor clinical evolution of HER2+ luminal tumors (ER+ by IHC), confirming that NCAPH did not influence the evolution of the so-called HER2-enriched tumors (Fig. 4a, b).
To investigate the role of NCAPH in the pathogenesis of luminal-HER2+ tumors, we crossed MMTV-NCAPH mice with MMTV-ErbB2 transgenic mice, which developed luminal B ERBB2+ breast tumors(49) (Fig. 4c). An increase in epithelial proliferation was evident in the mammary glands of double-transgenic MMTV-NCAPHErBb2+ mice relative to their MMTV-ErbB2 counterparts when Ki-67+ cells were stained by IHC and counted (Fig. 4d-f). Interestingly, these MMTV-NCAPHErBb2+ mice developed significantly more tumors than their MMTV-ErbB2 counterparts (Fig. 4g, h), although no differences were observed in any other pathophenotype of this disease.
Notably, while NCAPH transgenic mice developed distinct histopathological types of BC (Fig. 3g-j), the MMTV-NCAPHErBb2+ double-transgenic mice developed only infiltrating ductal adenocarcinomas (Fig. 4i-k), suggesting that the ErbB2/Neu gene exerts a dominant effect on the tumor phenotype. Therefore, the ErbB2/Neu oncogene appears to reprogram tumor differentiation so that instead of the histopathologically distinct tumors induced by NCAPH overexpression, the characteristics of infiltrating ductal adenocarcinoma predominated. Moreover, the double-transgenic mice mainly developed infiltrating ductal adenocarcinomas with a solid histopathological pattern (Fig. 4k), enhanced vascularization and a higher mitotic index than their MMTV-ErbB2+ counterparts (Fig. 4l-n). Indeed, the tumors proliferated significantly more than those from MMTV-ErbB2 mice (Fig. 4o-q). Thus, overexpression of NCAPH in ERBB2+ tumors generates more aggressive histopathology, a solid tumor with a high mitotic index, and greater vascularization and cell proliferation.
The different histopathological behaviors of tumors from MMTV-NCAPHErbB2+ double-transgenic mice led us to study molecules from some of the main pathways downstream of ERBB2/NEU. Intratumoral signaling was evaluated by assessing pAKT/pmTOR and pERK in the western blots of seven tumors from each phenotype, and only pAKT levels were significantly higher in tumors from the double-transgenic mice than in those from the single-transgenic mice (Fig. 4r, s).
Taken together, these results suggest that high levels of NCAPH are associated with poor evolution of luminal-HER2+ BC.
NCAPH expression is associated with poor breast cancer evolution in a genetically heterogeneous cohort of mice
To analyze the contribution of NCAPH expression to the heterogeneous evolution of luminal ERBB2+ BC, we took advantage of a genetically heterogeneous cohort of MMTV-ErbB2 transgenic mice generated by backcrossing (BX-Neu+) (Fig. 5a). We crossed MMTV-ErbB2 transgenic mice, which are on a susceptible FVB genetic background, with nontransgenic mice on a C57BL/6 genetic background that is resistant to BC development. We generated F1C57BL6/FVB MMTV-ErbB2 mice (F1-Neu+ mice hereafter) that were backcrossed with FVB mice to generate BX-Neu+ mice, a cohort of mice with a more varied BC evolution than genetically homogenous mouse strains. This cohort better reflects the heterogeneity observed in the human population, facilitating the identification of genetic and transcriptomic determinants associated with disease evolution(50, 74, 75).
This BX-Neu+ cohort was then used to evaluate whether intratumoral levels of Ncaph were associated with heterogeneous BC evolution, demonstrating that high intratumoral Ncaph RNA levels were associated with shorter tumor latency and survival (Fig. 5b, c), faster tumor growth, and larger tumor volume (Fig. 5d, e). Thus, high levels of Ncaph were associated with poor evolution of luminal HER2+ BC in a cohort of BX-Neu+ mice.
Since elevated NCAPH levels mediate a poor response to chemotherapy in BC patients and cells (Fig. 1g, 2o-q), we evaluated the response of tumors generated in BX-Neu+ mice to chemotherapy. Following the laws of transplants in mice(76), BC tumors that developed in the backcross cohort were transplanted into F1 mice, and their responses to anthracycline and taxane chemotherapy were evaluated. This strategy allowed us to evaluate the response of BC to chemotherapy in an extracellular context as homogeneous as possible.
Thus, differences in treatment responses could be primarily attributed to differences at the cell-autonomous level (Fig. 5f). Tumors with high levels of Ncaph responded worse to docetaxel treatment, as reflected by a smaller reduction in tumor size (Fig. 5g). In addition, the growth rate of tumors with high levels of Ncaph was faster, and their evolution was worse after chemotherapy than those that expressed Ncaph more weakly (Fig. 5h). However, Ncaph levels did not appear to influence the local response to doxorubicin (Fig. 5i, j). After chemotherapy with either doxorubicin or docetaxel, lung metastases were most evident in mice with high intratumoral levels of Ncaph (Fig. 5k).
These findings suggest that high intratumoral levels of Ncaph are associated with resistance to BC chemotherapy in this model.
A gene signature based on NCAPH expression defines poor tumor evolution in mice and humans
We examined the transcriptomic context in which Ncaph expression was associated with poor BC evolution. The wide range of BC evolution in backcross mice(50, 77) and the diverse patterns of gene expression expected(78) (Fig. 5) make this cohort an excellent tool for identifying transcripts associated with high levels of Ncaph expression and poor BC evolution(50, 79). We identified 64 transcripts associated with high intratumoral Ncaph levels in breast tumors from the backcross cohort, of which 45 were shared with humans (Fig. 6a, Additional File 3: Fig. S2a, b, and Additional File 2: Table S3). The functions of this 45-gene signature were assessed through GO enrichment analysis.
Unsurprisingly, several genes associated with high Ncaph levels were involved in processes related to the correct condensation and segregation of chromosomes during mitosis (Fig. 6b, Additional File 3: Fig. S2c, and Additional File 2: Table S4), and some were also correlated with poor BC evolution in the BX-Neu+ cohort of mice (Fig. 6c). When we integrated several of these genes into a multivariate LASSO regression model to define the poor tumor evolution of the BX-Neu+ mouse cohort (Fig. 6d, e and Additional File 2: Table S5), four genes were identified that were related to poor survival in BX-Neu+ mice: Oip5, Higd1a, Shc4, and Scrg1 (Additional File 2: Fig. S2d-f).
Interestingly, part of the gene signature identified in the heterogeneous BX-Neu+ model was also related to poor clinical evolution (relapse-free survival [RFS]) of the intrinsic luminal A subtype of BC patients (65) when evaluated using the KM Plotter database (Fig. 6f, g and Additional File 2: Table S6). Thus, high NCAPH expression is associated with a gene signature related to poor evolution in both mice and humans.
Identification of a genetic model associated with poor evolution in patients with luminal A tumors
Despite having the best overall prognosis among intrinsic subtypes, luminal A tumors display significant prognosis variation. It is crucial to identify patients with a poor prognosis for improved survival via initial therapeutic enhancements. High levels of NCAPH were associated with poor evolution, especially in luminal A tumors (Fig. 1d). Our research discerned an array of genes that exhibited a notable correlation with elevated intratumoral levels of Ncaph, as depicted in Fig. 6. We also found that some of these genes were associated with poor RFS in humans (Fig. 6h); therefore, we used a penalized multivariate LASSO regression model to identify a gene signature that reflects the poor prognosis of luminal A tumors(48).
The LASSO regression model was generated from the GOBO database of 401 patients with luminal A-diagnosed BCs (56). This cohort was divided into a training set (70%) and a test set comprising the remaining 30% to generate a polygenic risk score. First, bivariant analyses of the training set using Cox regression identified genes associated with poor evolution in terms of RFS (Additional File 2: Table S7). Later, genes with a P value < 0.25 were used to generate a polygenic risk score using the restrictive LASSO regression model(48).
Thus, the LASSO model was developed, and ten genes were identified that define disease prognosis, which we referred to as the Gene Signature for Luminal A (GSLA10) (Fig. 7a, Additional File 3: Fig. S3a, b, and Additional File 2: Table S8). The prognoses in the training and testing sets and in the global model were evaluated using the C-index (Additional File 3: Fig. S3c, d).
GSLA10 discriminated between low (bottom 1/3 risk score), medium (middle 1/3 risk score), and high risk (top 1/3 risk score) RFS in patients (Fig. 7b). We validated our model in two independent patient cohorts, METABRIC and TCGA-BRCA, with 718 and 499 luminal A breast cancer cases, respectively. The C-index, AUC of the ROC curve, and P value log-rank of the KM analysis were assessed when applying GSLA10 to the METABRIC and TCGA-BRCA cohorts, confirming the prediction capability of GSLA10 (Fig. 7c, d and Additional File 3: Fig. S3d).
Additionally, we compared the ability of these signatures to define the prognosis of luminal A tumors in terms of RFS at different time points, evaluating the AUC for the first five years after diagnosis, between 5 and 10 years, and after ten years in METABRIC and TCGA-BRCA (Fig. 7e-g). The ROC curves indicate that GSLA10 can predict the risk of relapse in those patients at these different time points.
In addition, we constructed a risk model (oncotype) based on 16 genes in the Oncotype (excluding five housekeeping genes) and provided a comprehensive comparison between GSLA10 and the Oncotype in terms of patient risk stratification and prognostic power in both the training cohort (GOBO) (Fig. 7h-j) and the independent validation cohorts (METABRIC and TCGA-BRCA) (Fig. 7k-m and n-p and Additional File 3: Fig. S3e). The comparison further confirms the robustness and superior prognostic power of GSLA10 over the Oncotype in luminal A tumors.
In conclusion, the GSLA10 gene signature identified a poor prognosis in luminal A patients. This model could help assess the prognosis of luminal A tumors and thus favor more personalized follow-up and therapy for patients with BC.