Stem cell antigen Sca-1 is implicated in murine cancer stem cell biology and breast cancer models, but the role of its human homologs Ly6K and Ly6E in breast cancer are not established. Here we report increased expression of Ly6K/E in human breast cancer specimens correlates with poor overall survival, with an additional specific role for Ly6E in poor therapeutic outcomes. Increased expression of Ly6K/E also correlated with increased expression of the immune checkpoint molecules PDL1 and CTLA4, increased tumor-infiltrating T regulatory cells, and decreased natural killer (NK) cell activation. Mechanistically, Ly6K/E was required for TGFβ signaling and proliferation in breast cancer cells, where they contributed to phosphorylation of Smad1/5 and Smad2/3. Furthermore, Ly6K/E promoted cytokine-induced PDL1 expression and activation and binding of NK cells to cancer cells. Finally, we found that Ly6K/E promoted drug resistance and facilitated immune escape in this setting. Overall, our results establish a pivotal role for a Ly6K/E signaling axis involving TGFβ in breast cancer pathophysiology and drug response, and highlight this signaling axis as a compelling realm for therapeutic invention. Cancer Res; 76(11); 3376–86. ©2016 AACR.

In recent years, it has become increasingly clear that the process of breast cancer oncogenesis and therapeutic sensitivity is profoundly affected by pathways that also infringe on cancer stem cell biology (1, 2). One of the gene families at the interface of cancer stem cell biology and cancer biology is the human Ly-6 gene family, which is related to the murine stem cell antigen-1 gene (Sca-1) or mouse (m) Ly6A. It encodes a set of glycosylated cell surface proteins that have been recognized as stem cell markers (1, 3, 4). The mLy6A confers resistance to radiotherapy and promotes metastatic behavior of mammary tumors in animal models (5). Earlier work from our laboratory has demonstrated that mLy6A regulates TGFβ, PTEN, and ERK/AKT cell signaling pathways (6–8). Mechanistically, mLy6A binds to TGFβ receptor-1 (TβR1), disrupts the TβR1 ligand complex, and inhibits Smad2/3 signaling. In addition, increased mLy6A expression in tumor cells correlates with a reduced level of GDF10, a novel tumor suppressive cytokine (6). Recently, Ly6K and Ly6E, another members of the Ly6 gene family, have been shown to be involved in human malignancies and have been suggested as potential therapeutic targets for cancer immunotherapy (9–14). Here we set out to investigate their role in breast cancer progression and the underlying cellular signaling mechanisms and relate these findings to clinical cases of breast cancer using clinical informatics.

Bioinformatics analysis

Oncomine (www.oncomine.org) was used to visualize gene expression microarray dataset. ProgeneV2 online tool (http://www.abren.net/PrognoScan/) was used to study survival outcome. The Cancer Genome Atlas (TCGA) Breast Dataset is available from http://tcga-data.nci.nih.gov/tcga/.

Reagents and reporter plasmids

TGFβ1, IFNγ, and IL4 was obtained from R&D Systems. The Smad-responsive TGFβ reporter plasmids were used as described previously (6).

shRNAs and cell lines

Ly6K sh1 (cat. no. TRCN0000117952), Ly6K sh2 (cat. no. TRCN0000117953), Ly6E sh1 (cat. no. TRCN0000154460), and Ly6E sh2 (cat. no. TRCN0000155331) shRNAs cloned into pLKO.1 were obtained from Sigma Inc. Lentivirus was produced in 293T cells by cotransfection of the pMD2.g and VSVG vectors. At 24 hours after transfection, the medium was replaced and virus was collected. Cells were infected with lentivirus for 24 hours in the presence of 4 mg/mL of polybrene and selection carried out with 1 μg/mL of puromycin.

For overexpression, the open reading frame containing clones (Ly6K, Ly6E) in G418 selection markers were obtained by Origene MD. The cells were transfected using Fugene HD and selected in 1 mg/mL G41.

The KHYG-1 natural killer (NK) cell line stably expressing KIR2DS1*002 have been generated and used as described previously (37). Prior to each assay, cells were screened for the surface expression of KIR2DS1 by immunolabeling using 2DS1 MoAB–CD158a/h (clone EB6B or 11PB6) and flow assay. Zeocin was removed prior to NK-cell coculture assays.

The cells were obtained by ATCC, which provided cells authenticated by short random repeat DNA sequencing. Upon arrival, cells were propagated and stored in multiple vials as recommended.

Colony assay

Cells (5,000/dish) were seeded into 100-mm dishes in 10 mL of DMEM containing 5% FBS. After 7 to 10 days, colonies were fixed in 50% trichloroacetic acid, stained with sulforhodamine B, and solubilized in 10 mmol/L Tris-HCl (pH 10), and absorbance was measured at 560 nm (6).

Nude mice xenograft assay

500,000 cells were injected subcutaneously into opposite flanks in five-week-old athymic nude mice. For the mice receiving T47D cells, mice were transplanted with controlled release (60 days release time, 0.18 mg/pellet total dose) 17β-estradiol pellet (Innovative Research of America). Tumor volume was determined by caliper measurements at weekly intervals. All animal studies were performed under a protocol approved by Georgetown University's Animal Care and Use Committee.

Quantitative real-time PCR

Total RNA was extracted using the Qiagen RNAeasy Mini Kit and cDNA was prepared according to the manufacturer's protocol (Invitrogen). Quantitative real-time PCR (qRT-PCR) was performed in triplicate in an ABI 7900 instrument using SYBRGreen detection (Applied Biosystems) according to the manufacturer's protocol. All primer sequences are described in Supplementary Table S1. The expression of each target gene was normalized to the expression of GAPDH using SDS2.4 software (Applied Biosystems).

Reporter assays

Cells were seeded in triplicate into 24-well plates at a density of 4 × 104 cells/well and then transfected with 100 ng of Smad-responsive firefly luciferase reporter plasmid and 1 ng of pCMV Renilla luciferase (for normalization) using Fugene HD (Roche). Luciferase activity was measured at 48 hours after transfection using the Dual Luciferase Reporter Assay System (Promega) and a Leader 50 Luminometer (Gen-Probe).

TGFβ, IFNγ, and IL4 treatment

Cells were seeded at 60% cell density and serum starved in 0.5% charcoal-stripped serum overnight. Cells were treated with 100 pg/mL TGFβ1 for 30 minutes, 10 ng/mL IFNγ, or 50 ng/mL IL4 for overnight.

Western blot analysis

Cells were lysed in a buffer containing 60 mmol/L octylglucoside, 0.5% Nonidet P-40, 0.1% SDS, 0.25% sodium deoxycholate, 50 mmol/L Tris-HCl (pH 7.5), 125 mmol/L NaCl, 1 mmol/L EDTA, 1 mmol/L EGTA, 10% glycerol, phosphatase, and a protease inhibitor mixture (Roche Molecular Biochemicals) at 4°C. A total of 20 to 50 μg of lysate was separated in a 4% to 12% NuPAGE Bis-Tris gel (Invitrogen) and transferred to nitrocellulose membranes. Primary antibody was incubated for either 1.5 hours at room temperature or overnight at 4°C. Secondary antibody was incubated for 30 minutes at room temperature, and proteins were visualized with West Pico Stable (Pierce). For Western blot analysis, the primary antibodies anti-Ly6K (cat no. AF6648; R&D Systems), Ly6E (cat no. NBP1 68553; Novus Biologicals), pS423/425 Smad3 (cat no. 9520), pS463/465 Smad1/5 (cat no. 9516), Smad2/3 (cat no. 9523), and Smad5 (cat no. 12534) were purchased from Cell Signaling Technology.

Semiquantitative IHC

IHC was performed using tissue staining HRP-DAB System kits (R&D Systems). Antigen retrieval was performed using citrate buffer (10 mmol/L citric acid, 0.05% Tween 20, pH 6.0). The primary antibodies were used for 1 hour at room temperature using anti-Ly6K antibody (AF6648; R&D Systems) at 1:100 and anti-Ly6E antibody (NBP1 68553; Novus Biologicals) at 1:400 dilution on clinical samples of breast cancer (tissue microarrays; USBiomax, Histopathology and Tissue Shared Resources at Georgetown University). Slides were semiquantitatively scored in a blind fashion using both intensity and percent positive cells. Intensity was scored on a scale of 0 to 3 as follows: 0 = negative, 1 = weak, 2 = moderate, and 3 = intense. Percent positive labeling was scored on a scale of 0 to 3 as follows: 0 = negative, 1 = <10%, 2 = 11%–50%, and 3 = >50%. Both these values were added for final numerical score. A score of 2 and below was considered negative and a score of 3 and above was considered positive.

NK-cell activation assay

NK-cell activation was measured by the release of granzyme B and the proinflammatory cytokine MIP1α from the KIR2DS1*002 NK (KHYG-1) cell line (37,49). Briefly, 2DS1 NK cells were cocultured with indicated cells at a ratio of 1:2 for 48 hours at 37°C in a 5% CO2 environment. Each assay condition consisted of 1 × 105 effectors (2DS1 NK cells) and 0.5 × 105 targets in 200-μL culture media in one well of a 48-well plate. After 48 hours, the cell culture supernatant was collected and stored at −80°C. Using the Luminex 100 platform (Luminex), secreted levels of granzyme B and MIP1α were assayed via the Procarta Immunoassay Kit (eBiosciences, Affymetrix) according to the manufacturer's instructions.

Flow cytometry

Single-cell suspension primary cultures or cell lines were prepared by incubation with 0.05% trypsin for 1 to 2 minutes, filtering through a 40-μm cell strainer, two washes with 1 × PBS, and blocking for 15 minutes on ice with 1 × PBS supplemented with 3% FBS. Cells were washed twice with 1 × PBS, fixed with 1% paraformaldehyde in PBS, and stored in the dark at 4°C until being sorted on a FACSAria flow cytometer using FACS Express De Novo software (BD Biosciences).

PDL1 surface expression assay

To assay the surface expression of PDL1, cells at a density of 5 × 106 cells/mL were incubated for 1 hour at room temperature with 1:200 dilutions of a BV421 anti-human CD274 (PDL1), antibody (Biolegend), or IgG (of the same isotype was used as a negative control) to perform flow cytometry.

NK-cell conjugate assay

2DS1 NK cells were labeled with Vybrant DiO Cell-Labeling Solution and the MDA-MB-231 control, Ly6K, and Ly6E knockdown cells were labeled with Vybrant DiI Cell-Labeling Solution according to the manufacturer's protocol (Life Technologies) and cocultured for 48 hours. The cells were analyzed by flow cytometry. We calculated the percent conjugate by obtaining the ratio of MDA-MB-231 bound to NK cells versus total live MDA-MB-231.

Statistical analysis

Statistical significance (P < 0.05) was determined from the mean ± SEM using two-tailed Student t tests, and between groups of two or more variables using two-tailed Fisher exact tests.

Increased expression of Ly6K and Ly6E in breast cancer

Analysis of the normal tissue samples showed that Ly6K was highly expressed in testis and Ly6E in liver tissue among the array of normal tissue samples (Supplementary Fig. S1A). Analysis of DNA copy number in invasive ductal breast carcinoma in TCGA dataset showed a significant genomic amplification of Ly6K and Ly6E in cancer tissue (Supplementary Fig. S1B). Statistical analysis across seven comparisons (cancer vs. normal) in three independent studies TCGA (unpublished), Richardson and colleagues (16), and Curtis and colleagues (17) showed increased Ly6K and Ly6E in breast cancer (Supplementary Fig. S1C; Supplementary Table S2).

Increased expression of Ly6K and Ly6E and overall patient survival

The status of gene expression and survival outcome was assessed by using PROGgeneV2 online tool, which showed that increased Ly6K (Fig. 1A) and increased Ly6E status (Fig. 1B) correlate well with poor 5-year overall survival in a breast cancer study (18) and in the Netherland Cancer Institute (6, 9–14, 19–23), respectively.

Figure 1.

Ly6K and Ly6E are markers for poor prognosis in human breast cancer. A and B, increased Ly6K (A) and increased Ly6E (B) expression in breast cancer samples is significantly associated with poor overall survival (endpoint—death) in the clinical cases of breast cancer. C and D, increased Ly6E expression in breast cancer samples is significantly associated with poor metastasis-free survival in chemotherapy nonresponding group (n = 104; C) and chemotherapy responding group (n = 110; D). E and F, increased Ly6E expression is significantly associated with poor metastasis-free survival in hormonal therapy nonresponders (n = 254; E) but not in hormonal therapy responders (n = 40; F). All graphs were generated using the PROGgeneV2 online tool. n, number of patients.

Figure 1.

Ly6K and Ly6E are markers for poor prognosis in human breast cancer. A and B, increased Ly6K (A) and increased Ly6E (B) expression in breast cancer samples is significantly associated with poor overall survival (endpoint—death) in the clinical cases of breast cancer. C and D, increased Ly6E expression in breast cancer samples is significantly associated with poor metastasis-free survival in chemotherapy nonresponding group (n = 104; C) and chemotherapy responding group (n = 110; D). E and F, increased Ly6E expression is significantly associated with poor metastasis-free survival in hormonal therapy nonresponders (n = 254; E) but not in hormonal therapy responders (n = 40; F). All graphs were generated using the PROGgeneV2 online tool. n, number of patients.

Close modal

We used comprehensive NKI dataset to study metastasis-free survival. The NKI study did not detect Ly6K expression data, but provided us with a wide range of information regarding the status of Ly6E expression and therapeutic outcome. Increased Ly6E expression is correlated with a poor 5-year metastasis-free chemotherapy outcome (Fig. 1C and D) and hormone therapy (Fig. 1E and F).

Ly6K and Ly6E expression in a breast cancer cell lines and breast cancer tissue

We next screened a panel of breast cancer cell lines for the RNA expression and found that triple-negative breast cancer (TNBC) cell lines expressed a higher levels of Ly6K (Fig. 2A), whereas Ly6E expression was detected in all breast cancer cell lines with higher levels in ER-positive breast cancer cell lines (Fig. 2B). Ly6K and Ly6E protein expression was tested in breast cancer clinical specimens. An IHC study containing 52 clinical samples (n = 29 ER-positive, n = 23 TNBC) revealed that 87% of TNBC tumors express detectable levels of Ly6K, whereas 24% TNBC tumors express Ly6E. Seventy-six percent of ER-positive tumors express Ly6E and 33% of ER-positive tumors express Ly6K (Fig. 2C).

Figure 2.

Ly6K and Ly6E are differentially expressed in breast cancer cell lines and cancer tissues. A, qRT-PCR analysis shows that Ly6K is highly expressed in TNBC cell lines. B, qRT-PCR analysis shows that Ly6E expression is expressed in all breast cancer cell lines with a 7-fold higher expression in ER-positive cell lines than in TNBC cell lines. C, example of a positive immunolabeling of Ly6K (II, ductal carcinoma in situ ER-positive; III, invasive ductal ER-positive; IV, TNBC) and a negative control (I); example of a positive immunolabeling of Ly6E protein (VI, ductal carcinoma in situ, ER-positive; VII, invasive/lobular TNBC; VIII, invasive and ductal carcinoma in situ, ER-positive), and negative control (V). Quantification chart (bottom) shows percentage of high numerical grading of Ly6K and Ly6E in tested ER-positive and TNBC cases.

Figure 2.

Ly6K and Ly6E are differentially expressed in breast cancer cell lines and cancer tissues. A, qRT-PCR analysis shows that Ly6K is highly expressed in TNBC cell lines. B, qRT-PCR analysis shows that Ly6E expression is expressed in all breast cancer cell lines with a 7-fold higher expression in ER-positive cell lines than in TNBC cell lines. C, example of a positive immunolabeling of Ly6K (II, ductal carcinoma in situ ER-positive; III, invasive ductal ER-positive; IV, TNBC) and a negative control (I); example of a positive immunolabeling of Ly6E protein (VI, ductal carcinoma in situ, ER-positive; VII, invasive/lobular TNBC; VIII, invasive and ductal carcinoma in situ, ER-positive), and negative control (V). Quantification chart (bottom) shows percentage of high numerical grading of Ly6K and Ly6E in tested ER-positive and TNBC cases.

Close modal

Effect of Ly6K and Ly6E on breast cancer growth

To investigate the functional role of Ly6K in breast cancer cells, we chose to stably knockdown Ly6K in MDA-MB-231 TNBC cells (Fig. 3A and Supplementary Fig. S2A). Ly6K knockdown cells gave rise to fewer colonies (Fig. 3B and Supplementary Fig. S3A). Ly6K knockdown cells gave rise to fewer and smaller xenografts (Fig. 3C). MDA-MB-231 cells exhibit a strong metastatic phenotype (24, 25), so we examined the effect of Ly6K knockdown in these cells on the invasion and distant colonization of tumor cells in well-characterized zebrafish model (26–28). The Ly6K knockdown cells showed diminished cell migration and colonization (Supplementary Fig. S4). To investigate the functional role of Ly6E in breast cancer cells, we chose to stably knockdown Ly6E in T47D ER-positive cells (Fig. 3D and Supplementary Fig. S2B). In colony assays, Ly6E knockdown cells gave rise to fewer colonies (Fig. 3E and Supplementary Fig. S3B). T47D cells have been demonstrated to establish xenograft tumors when supplemented with estrogen (29). Ly6E knockdown led to reduced xenograft tumors (Fig. 3F). These results suggest that Ly6K or Ly6E are required for tumorigenic growth.

Figure 3.

Ly6K and Ly6E are required for tumor cell growth. A, the qRT-PCR and Western blot analysis shows Ly6K expression in the indicated cells. B, indicated cells were seeded in low dilution for colony assay. Ly6K knockdown cells have significantly reduced colony formation. C, indicated cells were transplanted into opposite flanks of nude mice. Control cells gave rise to xenograft tumors in 2 weeks, whereas Ly6K knockdown cells gave rise to significantly fewer tumors. D, the qRT-PCR and Western blot analysis shows Ly6E expression. E, control and Ly6E knockdown T47D cells were seeded in low dilution for colony assays. Ly6E knockdown cells have significantly reduced colony formation. F, indicated cells were transplanted into opposite flanks of nude mice previously transplanted with control-release estrogen pellets. Control cells gave rise to xenograft tumors in 3 weeks, whereas Ly6E knockdown cells gave rise to significantly fewer tumors. The P values and fold change are calculated compared with control cells.

Figure 3.

Ly6K and Ly6E are required for tumor cell growth. A, the qRT-PCR and Western blot analysis shows Ly6K expression in the indicated cells. B, indicated cells were seeded in low dilution for colony assay. Ly6K knockdown cells have significantly reduced colony formation. C, indicated cells were transplanted into opposite flanks of nude mice. Control cells gave rise to xenograft tumors in 2 weeks, whereas Ly6K knockdown cells gave rise to significantly fewer tumors. D, the qRT-PCR and Western blot analysis shows Ly6E expression. E, control and Ly6E knockdown T47D cells were seeded in low dilution for colony assays. Ly6E knockdown cells have significantly reduced colony formation. F, indicated cells were transplanted into opposite flanks of nude mice previously transplanted with control-release estrogen pellets. Control cells gave rise to xenograft tumors in 3 weeks, whereas Ly6E knockdown cells gave rise to significantly fewer tumors. The P values and fold change are calculated compared with control cells.

Close modal

Ly6K and Ly6E promote gene signatures associated with drug resistance and epithelial-to-mesenchymal transition in breast cancer

We next examined whether Ly6K and Ly6E contribute to tumor growth by inducing characteristic gene signatures that lead to drug resistance or epithelial-to-mesenchymal transition (EMT). Ly6K knockdown in MDA-MB-231 and Ly6E knockdown in T47D cells significantly reduced RNA levels of the ABCC3 (Fig. 4A), an inducer of chemotherapeutic drug resistance (30), ABCG2 (Fig. 4B), and FGF-7 (Fig. 4C), inducers of hormonal therapy resistance in breast cancer (31, 32). Moreover, Ly6K and Ly6E knockdown also reduced the RNA expression of NANOG (Fig. 4D), CD34 (Fig. 4E), and PSCA (Fig. 4F), well-known stem cell genes that have been recently described to induce drug resistance (33–35). Next, we tested for the effect of Ly6K and Ly6E on regulators of EMT pathway, which is closely related to drug resistance and tumor metastasis. MDA-MB-231 cells express higher levels of Zeb1, an EMT inducer than T47D cells (Fig. 4G, compare first bar in main and inset graphs). The knockdown of Ly6K or Ly6E reduced expression of ZEB1 (Fig. 4G–J). Knockdown of either Ly6K or Ly6E decreased E-cadherin and increased N-cadherin (Fig. 4K), reversing the hallmark of EMT switch.

Figure 4.

Ly6K and Ly6E modulate drug resistance, stem cell genes, and EMT pathways. A–F, knockdown of Ly6K and Ly6E led to significantly reduced RNA levels ABCC3 (A), ABCG2 (B), FGF-7 (C), NANOG (D), CD34 (E), and PSCA (F). G, the qRT-PCR revealed that Zeb1 is considerably higher in MDA-MB-231 than T47D; compare first bar in the main and inset graph. Ly6K knockdown led to reduced ZEB1, an inducer of EMT. H and I, qRT-PCR of Ly6E (H) and Ly6K in the indicated cells (I). J and K, qRT-PCR of ZEB1 (J) and qRT-PCR of E-cadherin and N-cadherin (K). In some cases, part of a graph is presented in the inset on a magnified scale to show the relative expressions. **, P < 0.05; ***, P < 0.005; ****, P < 0.0005. The fold change is indicated numerically. The P values and fold change are calculated compared with control cells.

Figure 4.

Ly6K and Ly6E modulate drug resistance, stem cell genes, and EMT pathways. A–F, knockdown of Ly6K and Ly6E led to significantly reduced RNA levels ABCC3 (A), ABCG2 (B), FGF-7 (C), NANOG (D), CD34 (E), and PSCA (F). G, the qRT-PCR revealed that Zeb1 is considerably higher in MDA-MB-231 than T47D; compare first bar in the main and inset graph. Ly6K knockdown led to reduced ZEB1, an inducer of EMT. H and I, qRT-PCR of Ly6E (H) and Ly6K in the indicated cells (I). J and K, qRT-PCR of ZEB1 (J) and qRT-PCR of E-cadherin and N-cadherin (K). In some cases, part of a graph is presented in the inset on a magnified scale to show the relative expressions. **, P < 0.05; ***, P < 0.005; ****, P < 0.0005. The fold change is indicated numerically. The P values and fold change are calculated compared with control cells.

Close modal

Ly6K and Ly6E are required for TGFβ signaling in cancer cells

We have previously shown that the mouse Ly6A/E (Sca-1) gene, a potential functional homolog of human Ly6E and Ly6K, regulate TGFβ/Smad2–3 signaling (6). We found that Ly6K and Ly6E are required for TGFβ/Smad signaling as seen by Smad2/3–specific reporter activity (Fig. 5A). Depletion of Ly6K and Ly6E reduced the endogenous phosphorylation status of Smad3 (TGFβ signaling) and Smad1/5 (BMP signaling), suggesting that Ly6K and Ly6E are required for the increased TGFβ/Smad signaling in cancer cells (Fig. 5B). Loss of Ly6K and Ly6E also reduced levels of the TGFβ-responsive genes PAI1 (Fig. 5C) and cTGF (Fig. 5D). Next, we investigated whether Ly6K and Ly6E are required for ligand-induced TGFβ/Smad signaling in cancer cells. Knockdown of Ly6K or Ly6E reduced phosphorylation levels of Smad1/5 and Smad3 (Fig. 5E). These findings suggest that Ly6K and Ly6E are required for endogenous and ligand-induced TGFβ and BMP signaling in cancer cells.

Figure 5.

Ly6K and Ly6E are required for TGFβ signaling in cancer cells. A, a luciferase reporter with Smad-binding elements was transfected in the indicated cells and showed that Ly6K and Ly6E knockdown cells significantly reduced constitutively active TGFβ signaling. B, Western blotting revealed that knockdown of Ly6K and Ly6E led to reduced phosphorylation of Smad3. C–D, qRT-PCR analysis showed that knockdown of Ly6K and Ly6E led to reduced levels of well-known markers of TGFβ signaling PAI1 (C) and CTGF (D). E, Western blotting revealed that ligand-induced phosphorylation of Smad proteins require Ly6K and Ly6E. **, P < 0.05; ***, P < 0.005; ****, P < 0.0005. The fold change is indicated numerically. The P values and fold change are calculated compared with control cells.

Figure 5.

Ly6K and Ly6E are required for TGFβ signaling in cancer cells. A, a luciferase reporter with Smad-binding elements was transfected in the indicated cells and showed that Ly6K and Ly6E knockdown cells significantly reduced constitutively active TGFβ signaling. B, Western blotting revealed that knockdown of Ly6K and Ly6E led to reduced phosphorylation of Smad3. C–D, qRT-PCR analysis showed that knockdown of Ly6K and Ly6E led to reduced levels of well-known markers of TGFβ signaling PAI1 (C) and CTGF (D). E, Western blotting revealed that ligand-induced phosphorylation of Smad proteins require Ly6K and Ly6E. **, P < 0.05; ***, P < 0.005; ****, P < 0.0005. The fold change is indicated numerically. The P values and fold change are calculated compared with control cells.

Close modal

Ly6K and Ly6E facilitate tumor cell escape from immune surveillance

TGFβ signaling plays an important role in the tumor microenvironment including the innate immune response generated by NK cells and the adoptive immune response carried out by CTLs (36). A statistical analysis of gene expression in across six comparisons (TNBC vs. non-TNBC, normal vs. cancer) in three independent datasets showed Ly6K and Ly6E correlate significantly with breast cancer subset with increased PDL1, CTLA4 (tumor immune checkpoint inhibitors), and increased infiltration of suppressive T regulatory cells (marker CD25+) in Curtis and colleagues (17), Gluck and colleagues (50), and TCGA (Fig. 6A and Supplementary Fig. S5).

Figure 6.

Ly6K and Ly6E are required for cancer cell escape from immune surveillance. A, significant coexpression of Ly6K and Ly6E with CD25, CTLA4, and PDL1 was observed in clinical cases of breast cancer by statistical analysis across six comparisons in three independent datasets [1, TNBC n = 211 vs. non-TNBC n = 1340 (17); 2, grade 1 n = 89 vs. grade 3 n = 857 (17); 3, grade 1 n = 19 vs. grade 3 n = 69 (50); 4, TNBC n = 50 vs. non-TNBC n = 101; TCGA #5, normal n = 61 vs. invasive ductal n = 389; 6, TNBC n = 46 vs. non-TNBC n = 250] visualized by Oncomine (see Supplementary Fig. S5 for detailed information on each comparison). B, Luminex assay using conditioned medium showed that NK cells released significantly higher levels of granzyme B (I) and MIP1α (II) when cocultured with either Ly6K and Ly6E knockdown cancer cells. 2DS1 NK cells and indicated MDA-MB-231 cells were labeled with DiO and DiI cell-labeling dyes, respectively, and cocultured for 48 hours before flow cytometry analysis. The bar graph shows the percentage ratio of cancer cells bound to NK cells. NK-cell binding was increased with Ly6K knockdown and Ly6E knockdown cells (III). C, indicated cells were treated with IFNγ or IL4 and PDL1 surface expression was analyzed by flow cytometry. The knockdown of Ly6K or Ly6E led to reduced PDL1 expression induced by IFNγ (I) or IL4 (II). D, working model suggesting that Ly6K and Ly6E affect multiple aspects of cancer progression involving TGFβ signaling, drug response, and tumor immune escape. **, P < 0.05; ***, P < 0.005; ****, P < 0.0005. The fold change is indicated numerically. The P values and fold change are calculated compared with control cells.

Figure 6.

Ly6K and Ly6E are required for cancer cell escape from immune surveillance. A, significant coexpression of Ly6K and Ly6E with CD25, CTLA4, and PDL1 was observed in clinical cases of breast cancer by statistical analysis across six comparisons in three independent datasets [1, TNBC n = 211 vs. non-TNBC n = 1340 (17); 2, grade 1 n = 89 vs. grade 3 n = 857 (17); 3, grade 1 n = 19 vs. grade 3 n = 69 (50); 4, TNBC n = 50 vs. non-TNBC n = 101; TCGA #5, normal n = 61 vs. invasive ductal n = 389; 6, TNBC n = 46 vs. non-TNBC n = 250] visualized by Oncomine (see Supplementary Fig. S5 for detailed information on each comparison). B, Luminex assay using conditioned medium showed that NK cells released significantly higher levels of granzyme B (I) and MIP1α (II) when cocultured with either Ly6K and Ly6E knockdown cancer cells. 2DS1 NK cells and indicated MDA-MB-231 cells were labeled with DiO and DiI cell-labeling dyes, respectively, and cocultured for 48 hours before flow cytometry analysis. The bar graph shows the percentage ratio of cancer cells bound to NK cells. NK-cell binding was increased with Ly6K knockdown and Ly6E knockdown cells (III). C, indicated cells were treated with IFNγ or IL4 and PDL1 surface expression was analyzed by flow cytometry. The knockdown of Ly6K or Ly6E led to reduced PDL1 expression induced by IFNγ (I) or IL4 (II). D, working model suggesting that Ly6K and Ly6E affect multiple aspects of cancer progression involving TGFβ signaling, drug response, and tumor immune escape. **, P < 0.05; ***, P < 0.005; ****, P < 0.0005. The fold change is indicated numerically. The P values and fold change are calculated compared with control cells.

Close modal

To evaluate the effect of Ly6K and Ly6E knockdown on NK-cell activation, we used NK cells expressing KIR2DS1 receptor (37) and MDA-MB-231 cells, which have HLA-C*02:02, C*17:01 (unpublished data from Drs. Hurley and Upadhyay, Georgetown University), the classical ligand for the stimulatory KIR2DS1 receptor found on a subset of NK cells. We observed that 2DS1 NK cells released significantly higher levels of granzyme B and MIP1α when cocultured with Ly6K and Ly6E knockdown cells (Fig. 6B-I and II), suggesting that high expression levels of Ly6K and Ly6E in cancer cells can reduce NK-cell activation. Binding of NK cells to cancer cells precedes the NK-cell activation and subsequent killer cytokine release such as granzyme B. We observed that knockdown of either Ly6K or Ly6E led to significantly increased binding of cancer cells to NK cells (Fig. 6B-III and Supplementary Fig. S6).

Several cytokines including IFNγ and IL4-induced PDL1 expression on the cell surface of cancer cells is a major mechanism of tumor cell immune escape (38, 39). We found that knockdown of Ly6K or Ly6E led to reduced stimulation PDL1 expression followed by IFNγ and IL4 treatment (Fig. 6C-I and II and Supplementary Fig. S7), suggesting that Ly6K or Ly6E is required for cytokine-induced PDL1 expression by cancer cells. We found that rescue by forced expression of Ly6K or Ly6E increased colony growth, reduced granzyme B release, and increased PDL1 expression (Supplementary Fig. S8).

Stem cell antigen-1, also known as mouse Ly6E/Ly6A, is an established marker of cancer stem cells (40–42). Sca-1 can regulate tumor suppressor cytokine GDF10 production by cancer cells and disrupt TGFβ signaling by directly sequestering TGFβ receptor-1 (6). In spite of these recent findings, the role of homo sapiens Ly6 genes in the biology of human cancer progression remains poorly understood. In this study, we have revealed several novel aspects of the human Ly6 genes Ly6K and Ly6E in the pathobiology of breast cancer progression and tumor immune escape. Our study reveal the following major findings: first, both Ly6K and Ly6E are show elevated expression levels at the RNA and protein level in human breast cancer; second, Ly6K or Ly6E are prognosis markers for poor overall patient survival; third, Ly6K or Ly6E are required for the growth of human breast cancer cells; fourth, Ly6K and Ly6E are required for endogenous and ligand-induced phosphorylation of Smad2/3 and Smad1/5 in cancer cells; and fifth, Ly6K and Ly6E are required for tumor immune escape. Interestingly, knockdown of Ly6K or Ly6E was sufficient to induce phenotypic changes.

We found that high Ly6K and Ly6E expression and increased growth of cancer cells are associated with their ability to maintain TGFβ and BMP signaling in cancer cells. TGFβ signaling plays a central role in growth, metastasis, EMT, and invasiveness of cancer cells (43–46). The functional role of Ly6 family members, however, may be more complex than a linear regulation of a downstream pathway as highlighted by two findings: First, Ly6K and Ly6E expression is segregated in ER-positive and TNBC. Second, Ly6K and Ly6E are required for high expression levels of several markers of chemotherapeutic and hormonal drug resistance. The role of Ly6E in drug resistance was further supported by clinical survival data, that the expression of Ly6E may also be a marker of therapeutic response. Mechanistically, the differential expression of Ly6K and Ly6E and their effect in tumorigenic processes may be a function of an altered genetic network specific to certain tumor types, which can converge on TGFβ signaling. TGFβ signaling regulates both chemotherapy and hormonal therapy drug resistance (47, 48).

The underlying mechanism of the noted phenotypic effects of Ly6K and Ly6E on tumor immune escape relevant to NK-cell binding, and cytokine release and PDL1 expression involves critical essential roles of Ly6K and Ly6E for supporting constitutive TGFβ signaling and cancer cells cross-talk with tumor microenvironment especially in cancer immune surveillance (Fig. 6D). These finding suggest that the Ly6 family is one of the nodal centers involved in the regulation of TGFβ signaling. Our findings suggest that Ly6K and Ly6E are likely to emerge as attractive upstream targets to specifically modify TGFβ signaling in cancer cells. We hope this approach will leave the TGFβ signaling intact in normal cells and simultaneously decrease the intrinsic ability of cancer cells to grow and increase the effectiveness of immune surveillance, leading to an improved chemotherapy or hormonal therapy.

S. Madhavan is a consultant at Perthera. No potential conflicts of interest were disclosed by the other authors.

Conception and design: W.R. Frazier, S. Madhavan, R. Kumar, G. Upadhyay

Development of methodology: W.R. Frazier, N. Steiner, Y. Gusev, E. Glasgow, S. Madhavan, R. Kumar, G. Upadhyay

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M. AlHossiny, B. Kallakury, E. Glasgow, K. Creswell, G. Upadhyay

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. AlHossiny, L. Luo, W.R. Frazier, N. Steiner, Y. Gusev, B. Kallakury, K. Creswell, G. Upadhyay

Writing, review, and/or revision of the manuscript: W.R. Frazier, Y. Gusev, B. Kallakury, S. Madhavan, R. Kumar, G. Upadhyay

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): N. Steiner, G. Upadhyay

Study supervision: G. Upadhyay

The authors thank Dr. Carolyn Hurly, Dr. Louis Weiner, and Dr. Michael Atkins (Georgetown University, Washington, DC) for studies of tumor immune escape. The authors thank Dr. Rik Derynck (University of California San Francisco, San Francisco, CA), Dr. Lars J. von Buchholtz (National Institute of Dental and Craniofacial Research, Bethesda, MD), and Dr. Robert I. Glazer (Georgetown University) for their helpful comments.

This work was supported by the National Cancer Institute (NCI) grant 1R21CA175862 and the sub American Cancer Society Institutional Research Grant (IRG-92-152-20, M. Atkins) to G. Upadhyay; and RO1CA090970 to R. Kumar). This investigation was conducted using the Shared Resources Grant C06 RR14567, Cancer Center Support Grant 1P30-CA-51008, and Shared Instrumentation Grant 1 S10 RR019291-01A2 from the NIH.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Atena
M
,
Reza
AM
,
Mehran
G
. 
A review on the biology of cancer stem cells
.
Stem Cell Discov
2014
;
4
:
83
.
2.
Oskarsson
T
,
Batlle
E
,
Massagué
J
. 
Metastatic stem cells: sources, niches, and vital pathways
.
Cell Stem Cell
2014
;
14
:
306
21
.
3.
Holmes
C
,
Stanford
WL
. 
Concise review: stem cell antigen-1: expression, function, and enigma
.
Stem Cells
2007
;
25
:
1339
47
.
4.
Gumley
T
,
McKenzie
I
,
Sandrin
M
. 
Tissue expression, structure and function of the murine Ly-6 family of molecules
.
Immunol Cell
1995
;
73
:
277
96
.
5.
Woodward
WA
,
Chen
MS
,
Behbod
F
,
Alfaro
MP
,
Buchholz
TA
,
Rosen
JM
. 
WNT/beta-catenin mediates radiation resistance of mouse mammary progenitor cells
.
Proc Natl Acad Sci U S A
2007
;
104
:
618
23
.
6.
Upadhyay
G
,
Yin
Y
,
Yuan
H
,
Li
X
,
Derynck
R
,
Glazer
RI
. 
Stem cell antigen-1 enhances tumorigenicity by disruption of growth differentiation factor-10 (GDF10)-dependent TGF-{beta} signaling
.
Proc Natl Acad Sci U S A
2010
;
108
:
7820
5
.
7.
Upadhyay
G
,
Pollock
C
,
Yin
Y
,
Yuan
H
,
Kopelovich
L
,
Glazer
RI
. 
Stem cell antigen-1 (Sca-1) modulates PI3K/PTEN signaling and mammary tumor cell survival [abstract]
. In:
Proceedings of the 101st Annual Meeting of the American Association for Cancer Research
; 
2010
Apr 17–21;
Washington, DC. Philadelphia (PA)
:
AACR
.
Abstract nr 4214
.
8.
Reya
T
,
Morrison
SJ
,
Clarke
MF
,
Weissman
IL
. 
Stem cells, cancer, and cancer stem cells
.
Nature
2001
;
414
:
105
11
.
9.
Ishikawa
N
,
Takano
A
,
Yasui
W
,
Inai
K
,
Nishimura
H
,
Ito
H
, et al
Cancer-testis antigen lymphocyte antigen 6 complex locus K is a serologic biomarker and a therapeutic target for lung and esophageal carcinomas
.
Cancer Res
2007
;
67
:
11601
11
.
10.
Choi
SH
,
Kong
HK
,
Park
SY
,
Park
JH
. 
Metastatic effect of LY-6K gene in breast cancer cells
.
Int J Oncol
2009
;
35
:
601
7
.
11.
Ishikawa
H
,
Imano
M
,
Shiraishi
O
,
Yasuda
A
,
Peng
Y-F
,
Shinkai
M
, et al
Phase I clinical trial of vaccination with LY6K-derived peptide in patients with advanced gastric cancer
.
Gastric Cancer
2014
;
17
:
173
80
.
12.
Yoshitake
Y
,
Fukuma
D
,
Yuno
A
,
Hirayama
M
,
Nakayama
H
,
Tanaka
T
, et al
Phase II clinical trial of multiple peptide vaccination for advanced head and neck cancer patients revealed induction of immune responses and improved OS
.
Clin Cancer Res
2015
;
21
:
312
21
.
13.
Kono
K
,
Mizukami
Y
,
Daigo
Y
,
Takano
A
,
Masuda
K
,
Yoshida
K
, et al
Vaccination with multiple peptides derived from novel cancer-testis antigens can induce specific T-cell responses and clinical responses in advanced esophageal cancer
.
Cancer Sci
2009
;
100
:
1502
9
.
14.
Asundi
J
,
Crocker
L
,
Tremayne
J
,
Chang
P
,
Sakanaka
C
,
Tanguay
J
, et al
An antibody-drug conjugate directed against lymphocyte antigen 6 complex, locus E (LY6E) provides robust tumor killing in a wide range of solid tumor malignancies
.
Clin Cancer Res
2015
;
21
:
3252
62
.
15.
Shyamsundar
R
,
Kim
YH
,
Higgins
JP
,
Montgomery
K
,
Jorden
M
,
Sethuraman
A
, et al
A DNA microarray survey of gene expression in normal human tissues
.
Genome Biol
2005
;
6
:
R22
.
16.
Richardson
AL
,
Wang
ZC
,
De Nicolo
A
,
Lu
X
,
Brown
M
,
Miron
A
, et al
X chromosomal abnormalities in basal-like human breast cancer
.
Cancer Cell
2006
;
9
:
121
32
.
17.
Curtis
C
,
Shah
SP
,
Chin
SF
,
Turashvili
G
,
Rueda
OM
,
Dunning
MJ
, et al
The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups
.
Nature
2012
;
486
:
346
52
.
18.
Clarke
C
,
Madden
SF
,
Doolan
P
,
Aherne
ST
,
Joyce
H
,
O'Driscoll
L
, et al
Correlating transcriptional networks to breast cancer survival: a large-scale coexpression analysis
.
Carcinogenesis
2013
;
34
:
2300
8
.
19.
Hughes
TR
,
Mao
M
,
Jones
AR
,
Burchard
J
,
Marton
MJ
,
Shannon
KW
, et al
Expression profiling using microarrays fabricated by an ink-jet oligonucleotide synthesizer
.
Nat Biotechnol
2001
;
19
:
342
7
.
20.
Straver
ME
,
Glas
AM
,
Hannemann
J
,
Wesseling
J
,
van de Vijver
MJ
,
Rutgers
EJT
, et al
The 70-gene signature as a response predictor for neoadjuvant chemotherapy in breast cancer
.
Breast Cancer Res Treat
2010
;
119
:
551
8
.
21.
Liu
R
,
Wang
X
,
Chen
GY
,
Dalerba
P
,
Gurney
A
,
Hoey
T
, et al
The prognostic role of a gene signature from tumorigenic breast-cancer cells
.
N Engl J Med
2007
;
356
:
217
26
.
22.
van de Vijver
MJ
,
He
YD
,
van't Veer
LJ
,
Dai
H
,
Hart
A
,
Voskuil
DW
, et al
A gene-expression signature as a predictor of survival in breast cancer
.
N Engl J Med
2002
;
347
:
1999
2009
.
23.
van't Veer
LJ
,
Dai
HY
,
van de Vijver
MJ
,
He
Y
,
Hart
A
,
Mao
M
, et al
Gene expression profiling predicts clinical outcome of breast cancer
.
Nature
2002
;
415
:
530
6
.
24.
Vincent
KM
,
Findlay
SD
,
Postovit
LM
. 
Assessing breast cancer cell lines as tumour models by comparison of mRNA expression profiles
.
Breast Cancer Res
2015
;
17
:
114
.
25.
Kao
J
,
Salari
K
,
Bocanegra
M
,
Choi
Y-L
,
Girard
L
,
Gandhi
J
, et al
Molecular profiling of breast cancer cell lines defines relevant tumor models and provides a resource for cancer gene discovery
.
PLoS One
2009
;
4
:
e6146
.
26.
Teng
Y
,
Xie
X
,
Walker
S
,
White
DT
,
Mumm
JS
,
Cowell
JK
. 
Evaluating human cancer cell metastasis in zebrafish
.
BMC Cancer
2013
;
13
:
453
.
27.
Lee
SLC
,
Rouhi
P
,
Jensen
LD
,
Zhang
D
,
Ji
H
,
Hauptmann
G
, et al
Hypoxia-induced pathological angiogenesis mediates tumor cell dissemination, invasion, and metastasis in a zebrafish tumor model
.
Proc Natl Acad Sci USA
2009
;
106
:
19485
90
.
28.
Drabsch
Y
,
He
S
,
Zhang
L
,
Snaar-Jagalska
BE
,
Dijke ten
P
. 
Transforming growth factor-β signalling controls human breast cancer metastasis in a zebrafish xenograft model
.
Breast Cancer Res
2013
;
15
:
R106
.
29.
Gottardis
MM
,
Robinson
SP
,
Jordan
VC
. 
Estradiol-stimulated growth of MCF-7 tumors implanted in athymic mice: a model to study the tumoristatic action of tamoxifen
.
J Steroid Biochem
1988
;
30
:
311
4
.
30.
van der Schoor
LWE
,
Verkade
HJ
,
Kuipers
F
,
Jonker
JW
. 
New insights in the biology of ABC transporters ABCC2 and ABCC3: impact on drug disposition
.
Expert Opin Drug Metab Toxicol
2015
;
11
:
273
93
.
31.
Selever
J
,
Gu
G
,
Lewis
MT
,
Beyer
A
,
Herynk
MH
,
Covington
KR
, et al
Dicer-mediated upregulation of BCRP confers tamoxifen resistance in human breast cancer cells
.
Clin Cancer Res
2011
;
17
:
6510
21
.
32.
Chang
H-L
,
Sugimoto
Y
,
Liu
S
,
Ye
W
,
Wang
L-S
,
Huang
Y-W
, et al
Keratinocyte growth factor (KGF) induces tamoxifen (Tam) resistance in human breast cancer MCF-7 cells
.
Anticancer Res
2006
;
26
:
1773
84
.
33.
Jeter
CR
,
Liu
B
,
Liu
X
,
Chen
X
,
Liu
C
,
Calhoun-Davis
T
, et al
NANOG promotes cancer stem cell characteristics and prostate cancer resistance to androgen deprivation
.
Oncogene
2011
;
30
:
3833
45
.
34.
Geller
RB
,
Zahurak
M
,
Hurwitz
CA
,
Burke
PJ
,
Karp
JE
,
Piantadosi
S
, et al
Prognostic importance of immunophenotyping in adults with acute myelocytic leukaemia: the significance of the stem-cell glycoprotein CD34 (My10)
.
Br J Haematol
1990
;
76
:
340
7
.
35.
Morris
MJ
,
Eisenberger
MA
,
Pili
R
,
Denmeade
SR
,
Rathkopf
D
,
Slovin
SF
, et al
A phase I/IIA study of AGS-PSCA for castration-resistant prostate cancer
.
Ann Oncol
2012
;
23
:
2714
9
.
36.
Yang
L
,
Pang
Y
,
Moses
HL
. 
TGF-beta and immune cells: an important regulatory axis in the tumor microenvironment and progression
.
Trends Immunol
2010
;
31
:
220
7
.
37.
Frazier
WR
,
Steiner
N
,
Hou
L
,
Dakshanamurthy
S
,
Hurley
CK
. 
Allelic variation in KIR2DL3 generates a KIR2DL2-like receptor with increased binding to its HLA-C ligand
.
J Immunol
2013
;
190
:
6198
208
.
38.
Schoenborn
JR
,
Wilson
CB
. 
Regulation of interferon‐-γ during innate and adaptive immune responses
.
Adv Immunol
2007
;
190
:
6198
208
.
39.
Thompson
RH
,
Gillett
MD
,
Cheville
JC
,
Lohse
CM
,
Dong
H
,
Webster
WS
, et al
Costimulatory B7-H1 in renal cell carcinoma patients: indicator of tumor aggressiveness and potential therapeutic target
.
Proc Natl Acad Sci U S A
2004
;
101
:
17174
9
.
40.
Burger
PE
,
Xiong
X
,
Coetzee
S
,
Salm
SN
,
Moscatelli
D
,
Goto
K
, et al
Sca-1 expression identifies stem cells in the proximal region of prostatic ducts with high capacity to reconstitute prostatic tissue
.
Proc Natl Acad Sci U S A
2005
;
102
:
7180
5
.
41.
Ito
CY
,
Li
CY
,
Bernstein
A
,
Dick
JE
,
Stanford
WL
. 
Hematopoietic stem cell and progenitor defects in Sca-1/Ly-6A-null mice
.
Blood
2003
;
101
:
517
23
.
42.
Grange
C
,
Lanzardo
S
,
Cavallo
F
,
Camussi
G
,
Bussolati
B
. 
Sca-1 identifies the tumor-initiating cells in mammary tumors of BALB-neuT transgenic mice
.
Neoplasia
2008
;
10
:
1433
43
.
43.
Derynck
R
,
Akhurst
RJ
,
Balmain
A
. 
TGF-beta signaling in tumor suppression and cancer progression
.
Nat Genet
2001
;
29
:
117
29
.
44.
Lamouille
S
,
Xu
J
,
Derynck
R
. 
Molecular mechanisms of epithelial-mesenchymal transition
.
Nat Rev Mol Cell Biol
2014
;
15
:
178
96
.
45.
Massagué
J
. 
TGFβ signalling in context
.
Nat Rev Mol Cell Biol
2012
;
13
:
616
30
.
46.
Derynck
R
,
Muthusamy
BP
,
Saeteurn
KY
. 
Signaling pathway cooperation in TGF-β-induced epithelial-mesenchymal transition
.
Curr Opin Cell Biol
2014
;
31
:
56
66
.
47.
Brunen
D
,
Willems
SM
,
Kellner
U
,
Midgley
R
,
Simon
I
,
Bernards
R
. 
TGF-β: an emerging player in drug resistance
.
Cell Cycle
2013
;
12
:
2960
8
.
48.
Yoo
YA
,
Kim
YH
,
Kim
JS
,
Seo
JH
. 
The functional implications of Akt activity and TGF-beta signaling in tamoxifen-resistant breast cancer
.
Biochim Biophys Acta
2008
;
1783
:
438
47
.
49.
Steiner
NK
,
Dakshanamurthy
S
,
VandenBussche
CJ
,
Hurley
CK
. 
Extracellular domain alterations impact surface expression of stimulatory natural killer cell receptor KIR2DS5
.
Immunogenetics
2008
;
60
:
655
67
.
50.
Gluck
S
,
Ross
JS
,
Royce
M
,
McKenna
EFJ
,
Perou
CM
,
Avisar
E
, et al
TP53 genomics predict higher clinical and pathologic tumor response in operable early-stage breast cancer treated with docetaxel-capecitabine +/- trastuzumab
.
Breast Cancer Res Treat
2012
;
132
:
781
91
.