A Methyl-Deviator Epigenotype of Estrogen Receptor–Positive Breast Carcinoma Is Associated with Malignant Biology
2011; Elsevier BV; Volume: 179; Issue: 1 Linguagem: Inglês
10.1016/j.ajpath.2011.03.022
ISSN1525-2191
AutoresJonathan Keith Killian, Sven Bilke, Sean Davis, Robert L. Walker, Erich Jaeger, M. Scott Killian, Joshua J. Waterfall, Marina Bibikova, Jian‐Bing Fan, William I. Smith, Paul S. Meltzer,
Tópico(s)Breast Cancer Treatment Studies
ResumoWe broadly profiled DNA methylation in breast cancers (n = 351) and benign parenchyma (n = 47) for correspondence with disease phenotype, using FFPE diagnostic surgical pathology specimens. Exploratory analysis revealed a distinctive primary invasive carcinoma subclass featuring extreme global methylation deviation. Subsequently, we tested the correlation between methylation remodeling pervasiveness and malignant biological features. A methyl deviation index (MDI) was calculated for each lesion relative to terminal ductal-lobular unit baseline, and group comparisons revealed that high-grade and short-survival estrogen receptor–positive (ER+) cancers manifest a significantly higher MDI than low-grade and long-survival ER+ cancers. In contrast, ER− cancers display a significantly lower MDI, revealing a striking epigenomic distinction between cancer hormone receptor subtypes. Kaplan-Meier survival curves of MDI-based risk classes showed significant divergence between low- and high-risk groups. MDI showed superior prognostic performance to crude methylation levels, and MDI retained prognostic significance (P < 0.01) in Cox multivariate analysis, including clinical stage and pathological grade. Most MDI targets individually are significant markers of ER+ cancer survival. Lymphoid and mesenchymal indexes were not substantially different between ER+ and ER− groups and do not explain MDI dichotomy. However, the mesenchymal index was associated with ER+ cancer survival, and a high lymphoid index was associated with medullary carcinoma. Finally, a comparison between metastases and primary tumors suggests methylation patterns are established early and maintained through disease progression for both ER+ and ER− tumors. We broadly profiled DNA methylation in breast cancers (n = 351) and benign parenchyma (n = 47) for correspondence with disease phenotype, using FFPE diagnostic surgical pathology specimens. Exploratory analysis revealed a distinctive primary invasive carcinoma subclass featuring extreme global methylation deviation. Subsequently, we tested the correlation between methylation remodeling pervasiveness and malignant biological features. A methyl deviation index (MDI) was calculated for each lesion relative to terminal ductal-lobular unit baseline, and group comparisons revealed that high-grade and short-survival estrogen receptor–positive (ER+) cancers manifest a significantly higher MDI than low-grade and long-survival ER+ cancers. In contrast, ER− cancers display a significantly lower MDI, revealing a striking epigenomic distinction between cancer hormone receptor subtypes. Kaplan-Meier survival curves of MDI-based risk classes showed significant divergence between low- and high-risk groups. MDI showed superior prognostic performance to crude methylation levels, and MDI retained prognostic significance (P < 0.01) in Cox multivariate analysis, including clinical stage and pathological grade. Most MDI targets individually are significant markers of ER+ cancer survival. Lymphoid and mesenchymal indexes were not substantially different between ER+ and ER− groups and do not explain MDI dichotomy. However, the mesenchymal index was associated with ER+ cancer survival, and a high lymphoid index was associated with medullary carcinoma. Finally, a comparison between metastases and primary tumors suggests methylation patterns are established early and maintained through disease progression for both ER+ and ER− tumors. Breast cancer is a heterogeneous disease, manifesting variation at the clinical, biological, histopathological, and molecular levels. Profiling studies1Sorlie T. Perou C.M. Tibshirani R. Aas T. Geisler S. Johnsen H. Hastie T. Eisen M.B. van de Rijn M. Jeffrey S.S. Thorsen T. Quist H. Matese J.C. Brown P.O. Botstein D. Eystein Lonning P. Borresen-Dale A.L. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications.Proc Natl Acad Sci U S A. 2001; 98: 10869-10874Crossref PubMed Scopus (8330) Google Scholar, 2Chin S.F. Teschendorff A.E. Marioni J.C. Wang Y. Barbosa-Morais N.L. Thorne N.P. Costa J.L. Pinder S.E. van de Wiel M.A. Green A.R. Ellis I.O. Porter P.L. Tavaré S. Brenton J.D. Ylstra B. Caldas C. High-resolution aCGH and expression profiling identifies a novel genomic subtype of ER negative breast cancer.Genome Biol. 2007; 8: R215Crossref PubMed Scopus (242) Google Scholar of gene expression and DNA copy number have identified molecular markers that can be used to distinguish clinically relevant tumor subtypes. DNA methylation analysis is emerging as a promising avenue for cancer classification; several studies3Hartmann O. Spyratos F. Harbeck N. Dietrich D. Fassbender A. Schmitt M. Eppenberger-Castori S. Vuaroqueaux V. Lerebours F. Welzel K. Maier S. Plum A. Niemann S. Foekens J.A. Lesche R. Martens J.W. DNA methylation markers predict outcome in node-positive, estrogen receptor-positive breast cancer with adjuvant anthracycline-based chemotherapy.Clin Cancer Res. 2009; 15: 315-323Crossref PubMed Scopus (77) Google Scholar, 4Holm K. Hegardt C. Staaf J. Vallon-Christersson J. Jönsson G. Olsson H. Borg A. Ringnér M. Molecular subtypes of breast cancer are associated with characteristic DNA methylation patterns.Breast Cancer Res. 2010; 12: R36Crossref PubMed Scopus (215) Google Scholar, 5Bloushtain-Qimron N. Yao J. Snyder E.L. Shipitsin M. Campbell L.L. Mani S.A. Hu M. Chen H. Ustyansky V. Antosiewicz J.E. Argani P. Halushka M.K. Thomson J.A. Pharoah P. Porgador A. Sukumar S. Parsons R. Richardson A.L. Stampfer M.R. Gelman R.S. Nikolskaya T. Nikolsky Y. Polyak K. Cell type-specific DNA methylation patterns in the human breast.Proc Natl Acad Sci U S A. 2008; 105: 14076-14081Crossref PubMed Scopus (192) Google Scholar, 6Martens J.W. Margossian A.L. Schmitt M. Foekens J. Harbeck N. DNA methylation as a biomarker in breast cancer.Future Oncol. 2009; 5: 1245-1256Crossref PubMed Scopus (43) Google Scholar, 7Widschwendter M. Siegmund K.D. Muller H.M. Fiegl H. Marth C. Muller-Holzner E. Jones P.A. Laird P.W. Association of breast cancer DNA methylation profiles with hormone receptor status and response to tamoxifen.Cancer Res. 2004; 64: 3807-3813Crossref PubMed Scopus (275) Google Scholar point toward the potential for DNA methylation markers to identify distinct breast cancer phenotypes using candidate gene measurements and microarray analyses. As a robust biomarker conserved in routinely processed clinical specimens, DNA methylation is amenable to high-throughput microarray-based discovery,8Killian J.K. Bilke S. Davis S. Walker R.L. Killian M.S. Jaeger E.B. Chen Y. Hipp J. Pittaluga S. Raffeld M. Cornelison R. Smith Jr, W.I. Bibikova M. Fan J.B. Emmert-Buck M.R. Jaffe E.S. Meltzer P.S. Large-scale profiling of archival lymph nodes reveals pervasive remodeling of the follicular lymphoma methylome.Cancer Res. 2009; 69: 758-764Crossref PubMed Scopus (45) Google Scholar, 9Killian J.K. Walker R.L. Suuriniemi M. Jones L. Scurci S. Singh P. Cornelison R. Harmon S. Boisvert N. Zhu J. Wang Y. Bilke S. Davis S. Giaccone G. Smith Jr, W.I. Meltzer P.S. Archival fine-needle aspiration cytopathology (FNAC) samples: untapped resource for clinical molecular profiling.J Mol Diagn. 2010; 12: 739-745Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar providing a justification for translational epigenotype-phenotype correlation in routine breast cancer pathological samples. In the current study, we present a large-scale DNA methylation analysis of primary invasive breast cancers for deviation from the epigenetic state of the normal mammary terminal ductal-lobular unit (TDLU). The TDLU is the structural and functional unit of the mammary gland and is generally considered the origin of breast carcinomas.10Wellings S.R. A hypothesis of the origin of human breast cancer from the terminal ductal lobular unit.Pathol Res Pract. 1980; 166: 515-535Crossref PubMed Scopus (78) Google Scholar, 11Wellings S.R. Jensen H.M. On the origin and progression of ductal carcinoma in the human breast.J Natl Cancer Inst. 1973; 50: 1111-1118Crossref PubMed Scopus (248) Google Scholar, 12Gusterson B.A. Ross D.T. Heath V.J. Stein T. Basal cytokeratins and their relationship to the cellular origin and functional classification of breast cancer.Breast Cancer Res. 2005; 7: 143-148Crossref PubMed Scopus (205) Google Scholar, 13Mills S.E. Histology for Pathologists.in: Lippincott Williams & Wilkins, Philadelphia2007: xiGoogle Scholar In addition to providing a normal tissue epigenetic baseline, the TDLU profile defined by DNA methylation targets invariant among numerous unrelated patients permits filtration of array signals potentially arising from neutral genetic and epigenetic polymorphisms.14Byun H.M. Siegmund K.D. Pan F. Weisenberger D.J. Kanel G. Laird P.W. Yang A.S. Epigenetic profiling of somatic tissues from human autopsy specimens identifies tissue- and individual-specific DNA methylation patterns.Hum Mol Genet. 2009; 18: 4808-4817Crossref PubMed Scopus (198) Google Scholar The use of archival diagnostic formalin-fixed, paraffin embedded (FFPE) pathological samples for biomarker discovery provides multiple unique benefits, including an indication of potential biomarker applicability to clinical practice.15Paik S. Shak S. Tang G. Kim C. Baker J. Cronin M. Baehner F.L. Walker M.G. Watson D. Park T. Hiller W. Fisher E.R. Wickerham D.L. Bryant J. Wolmark N. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer.N Engl J Med. 2004; 351: 2817-2826Crossref PubMed Scopus (4689) Google Scholar Contributions to measurements of cancer versus normal tissue epigenetic deviation may arise from both within and outside the cancer cell nucleus. Intrinsic to the cancer cell, de novo methyltransferase activity may generate divergent epialleles; not to be overlooked, faithful maintenance methylation of conserved cell lineage–specific marks,5Bloushtain-Qimron N. Yao J. Snyder E.L. Shipitsin M. Campbell L.L. Mani S.A. Hu M. Chen H. Ustyansky V. Antosiewicz J.E. Argani P. Halushka M.K. Thomson J.A. Pharoah P. Porgador A. Sukumar S. Parsons R. Richardson A.L. Stampfer M.R. Gelman R.S. Nikolskaya T. Nikolsky Y. Polyak K. Cell type-specific DNA methylation patterns in the human breast.Proc Natl Acad Sci U S A. 2008; 105: 14076-14081Crossref PubMed Scopus (192) Google Scholar coupled with malignant cell population enrichment, could also manifest as differential methylation between benign and cancer tissues. Meanwhile, cancer-lesion epigenetic distinctions may be extrinsic to the cancer cell, arising from characteristic microanatomical embedding of benign elements among cancer epithelial cells that often determine histopathological classification.16Tavassoli F.A. Devilee P. International Agency for Research on Cancer, World Health OrganizationPathology and Genetics of Tumours of the Breast and Female Genital Organs.in: IAPS Press, Lyon2003: 432Google Scholar, 17Fisher E.R. Sass R. Fisher B. Pathologic findings from the National Surgical Adjuvant Project for Breast Cancers (protocol no. 4), X: discriminants for tenth year treatment failure.Cancer. 1984; 53: 712-723PubMed Google Scholar For example, the microarchitecture of breast medullary carcinoma displays syncytial cords of high-grade malignant epithelial cells interwoven with channels of benign lymphoid cells18Ridolfi R.L. Rosen P.P. Port A. Kinne D. Mike V. Medullary carcinoma of the breast: a clinicopathologic study with 10 year follow-up.Cancer. 1977; 40: 1365-1385Crossref PubMed Scopus (338) Google Scholar, 19Rakha E.A. Aleskandarany M. El-Sayed M.E. Blamey R.W. Elston C.W. Ellis I.O. Lee A.H. The prognostic significance of inflammation and medullary histological type in invasive carcinoma of the breast.Eur J Cancer. 2009; 45: 1780-1787Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar; and the subtype-specific molecular signature of the lesion will derive from both compartments. By contrast, nonspecific heterogeneity across biological subclasses may arise from benign glandular and inflammatory elements and possibly iatrogenic effects (ie, needle-core-biopsy–related changes). Therefore, microscopy-based histological control and molecular quantification of constituent benign lymphoid and mesenchymal epialleles may be beneficial for understanding cancer tissue differential methylation signatures. Subsequent to primary invasive tumorigenesis, the fidelity of maintenance and de novo DNA methylation during disease progression is incompletely understood. Archival pathological specimens provide an opportunity to compare primary tumors with longitudinal recurrences to probe the status of these processes. Thus, finally, in our study, we compare primary tumors with matched longitudinal recurrences to obtain a global snapshot of the methylome at different tumor stages and to investigate the stability of DNA methylation patterns during disease evolution. FFPE breast cancer (n = 351), benign breast TDLU (n = 32), reactive lymph node (n = 9), and benign mesenchyme (fibromuscular tissue, n = 5) samples were retrieved from the pathology department archives of Suburban Hospital, Bethesda, MD (Table 1 and Figure 1). To reduce case selection bias, we included all available archival breast cancers from a consecutive 2-year period in the analysis. Available clinical registry data included cancer stage, follow-up interval, and time to distant recurrence. Survival analyses were based on the end point of distant recurrence. Specimens and corresponding clinical data were deidentified according to the NIH Office of Human Subjects Research policy.Table 1Patient and Sample CharacteristicsCharacteristics by type of tissueNo. affectedBreast carcinoma351 Primary invasive breast carcinoma312 Age at primary diagnosis (median, 60 years) (years) 7 years' follow up with no distant metastasis. (median, 8 years)118 NA92 ER− Failure†Failure indicates subsequent distant breast cancer metastasis. (median, 1.8 years)19 Censor‡Censor indicates >7 years' follow up with no distant metastasis. (median, 8 years)11 NA19 Molecular subtype comparisons⁎Tested and informative samples in each category. Basallike status for ER− cancers Basallike16 Not basallike8 Ki-67 low vs high ER+ cancers High32 Low47 Her-2 status for ER+ cancers Amplified23 Not amplified104 Her-2 status for ER− cancers Amplified10 Not amplified21 Metastatic breast carcinoma30 Invasive breast carcinoma NOS9Benign tissues46 Mammary TDLU32 Muscle tissue (female)5 Benign lymph node (female)9There were 397 total lesions and tissues.ERS, ER status; NA, not annotated; NHG, Nottingham histological grade; NOS, not otherwise specified. Tested and informative samples in each category.† Failure indicates subsequent distant breast cancer metastasis.‡ Censor indicates >7 years' follow up with no distant metastasis. Open table in a new tab There were 397 total lesions and tissues. ERS, ER status; NA, not annotated; NHG, Nottingham histological grade; NOS, not otherwise specified. Histological sections were reviewed by a pathologist (J.K.K.) for characteristic pathological features and scored for cancer grade according to the Nottingham system.20Elston C.W. Ellis I.O. Pathological prognostic factors in breast cancer, I: the value of histological grade in breast cancer: experience from a large study with long-term follow-up.Histopathology. 1991; 19: 403-410Crossref PubMed Scopus (4680) Google Scholar The region of characteristic tumor histological features with maximal tumor-cell fraction was marked on the slide section, and the target region was then manually dissected from the homologous region of the corresponding FFPE tissue block using a 1- to 2-mm needle micropunch (J.K.K.). Similarly, benign TDLUs, lymph nodes, and mesenchymal muscle and fibrous elements were needle dissected from paraffin blocks under histological guidance. Tissue cores were lysed by incubation at 65°C for 2 to 3 days in 200 μL of FFPE tissue lysis solution (160 μL of Qiagen ATL + 20 μL of Qiagen proteinase K + 20 μL of Dako target retrieval solution), and lysates were processed to yield 1 to 2 μg of bisulfite-modified DNA using the EZ DNA methylation kit (Zymo Research, Irvine, CA). The yield of bisulfite-converted DNA was measured by Nanodrop (ThermoScientific, Wilmington, DE). From available paraffin blocks with residual tumor, adjacent 2-mm cores to those used for methylation profiling were taken to construct TMAs for immunophenotyping. TMA slide sections were immunostained for estrogen receptor (ER), progesterone receptor, Her-2, CK5/6, pan-CK, Ki-67, and epidermal growth factor receptor in a diagnostic pathology laboratory using a Ventana autostainer (Ventana Medical Systems, Inc., Tucson, AZ) with antibody clones SP1, 1E2, 4B5, D5/16B4, AE1AE3, 30-9, and 2-18C9, respectively. The cutoff for Ki-67 low versus high proliferative index was positive staining of 10% cancer cell nuclei.21Gaglia P. Bernardi A. Venesio T. Caldarola B. Lauro D. Cappa A.P. Calderini P. Liscia D.S. Cell proliferation of breast cancer evaluated by anti-BrdU and anti-Ki-67 antibodies: its prognostic value on short-term recurrences.Eur J Cancer. 1993; 29A: 1509-1513Abstract Full Text PDF PubMed Scopus (53) Google Scholar, 22Pierga J.Y. Leroyer A. Viehl P. Mosseri V. Chevillard S. Magdelenat H. Long term prognostic value of growth fraction determination by Ki-67 immunostaining in primary operable breast cancer.Breast Cancer Res Treat. 1996; 37: 57-64Crossref PubMed Scopus (50) Google Scholar The basallike immunophenotype was determined by a five-marker panel.23Nielsen T.O. Hsu F.D. Jensen K. Cheang M. Karaca G. Hu Z. Hernandez-Boussard T. Livasy C. Cowan D. Dressler L. Akslen L.A. Ragaz J. Gown A.M. Gilks C.B. van de Rijn M. Perou C.M. Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma.Clin Cancer Res. 2004; 10: 5367-5374Crossref PubMed Scopus (2078) Google Scholar Bisulfite-converted DNA, 250 ng, was assayed using the GoldenGate Cancer Panel I methylation assay (Illumina, Inc., San Diego, CA), as previously described.24Bibikova M. Fan J.B. GoldenGate assay for DNA methylation profiling.Methods Mol Biol. 2009; 507: 149-163Crossref PubMed Scopus (124) Google Scholar, 25Bibikova M. Lin Z. Zhou L. Chudin E. Garcia E.W. Wu B. Doucet D. Thomas N.J. Wang Y. Vollmer E. Goldmann T. Seifart C. Jiang W. Barker D.L. Chee M.S. Floros J. Fan J.B. High-throughput DNA methylation profiling using universal bead arrays.Genome Res. 2006; 16: 383-393Crossref PubMed Scopus (519) Google Scholar Briefly, this assay measures DNA methylation at 1505 distinct CpG targets distributed among 807 genes. Sample target methylation β values that approximate percentage methylation within the sample homogenate were extracted in BeadStudio (Illumina, Inc.) from raw Cy3 and Cy5 signal intensities. Samples that did not pass array internal controls were excluded. The lesion β is the average β of any samples that were technical replicates (DNA or needle cores) derived from a single patient lesion. Methylation β data are provided in Supplemental Table S1 (available at http://ajp.amjpathol.org). Methylation data may also be retrieved from Gene Expression Omnibus. Dynamic data exploration and discovery analyses were performed using Qlucore Omics Explorer version 2.1 (Qlucore AB, Lund, Sweden), as follows. The 1505 array target methylation β values from 32 TDLU and 312 primary carcinoma lesions were extracted from BeadStudio and imported to QOE. Data normalization was set as follows: mean = 0 and variance = 1; the hierarchical clustering module was set to maximum linkage, and the variance filter was dynamically tuned while observing sample and variable clustering. The variance was set to 0.5 to yield the set of 242 target variables shown in Figure 2A. Target methyl deviation was calculated as the methylation β difference between sample and TDLU baseline. The baseline target β is the TDLU group average β from 32 different individuals.Target Methyl Deviation=abs(βlesion−βbaseline), whereβbaseline=avgβTDLU.Target methyl deviation values were summed to compute the methyl deviation index (MDI) of each lesion:MDI=∑ abs(βlesion−βbaseline).Of 1505 array targets, 237 had a baseline variance >0.1, and these targets were excluded from calculations of MDI that implement a baseline uniformity filter. Target MDI rank from highest to lowest is the SD of the target within the group. For example, the top 100 MDI targets in the ER+ cancer group are the 100 targets with the greatest SD in that group. Alternative to MDI, the methylation β index was calculated as the sum of all target methylation levels within the lesion without reference to a baseline:Methylation β Index=∑ βlesion.The performances of multiple different arbitrary cancer variance cutoffs for MDI-based survival prognostication were compared using receiver operating characteristic (ROC) area under the curve (AUC) analysis, as was the performance of MDI versus methylation β index. The statistical significance (P values and false-discovery-rate–corrected Q scores) of MDI target measures between long- versus short-survival ER+ breast cancers was calculated by two-group comparison of array CpG target variables using analysis of variance in QOE. The MDI_72sig refers to the intersection of statistically significant survival targets between long- and short-survival ER+ cancers (P < 0.05), with the top 100 MDI targets in the ER+ cancer group. To calculate the lymphoid index (LI), statistically significant lymphoid-specific methyl markers relative to TDLU (analysis of variance, P < 0.05) were identified in QOE. The input target variables were the 1268 conforming TDLU targets (variance <0.1, as previously indicated), and the samples were the 32 TDLU and the 9 female lymphoid tissues. Next, the LI_59 was calculated for each primary cancer lesion after setting the variance filter to 0.5 (to enrich for lymphoid-specific markers of highest contrast) and dividing by 59, the number of targets in the resulting cassette:LI=∑ (1−[abs(βlesion−βlymphoid)]/59).The same concept was used to calculate the lesion mesenchymal index (MI) by summing mesenchyme-specific methylation markers relative to TDLU:MI=∑ (1−[abs(βlesion−βmesenchymal)]/44).MDI, LI, and MI were treated as continuous variables and were not stratified or discretized for ROC and pairwise analyses. For the Kaplan-Meier survival analysis and Cox regression analyses, patients with ER+ unilateral primary invasive carcinomas and ≥7 years of follow-up (n = 157) were assigned to low-, middle-, and high-risk groups based on MDI_109 rank (bottom, 30%; middle, 40%; and top, 30%; respectively) and low- and high-risk groups based on MI_44 and LI_59 rank (bottom, 50%; and top, 50%; respectively). Box plot graphs, ROC calculations, and survival analyses were performed using SigmaPlot11.2 (Systat Software, Inc., Chicago, IL) and R. Heat plots were generated in QOE. Target significance for ER+ cancer survival (P values and Q scores) was measured in QOE using analysis of variance. The methylation array profiles of 351 individual breast cancers and 46 noncancer tissues were included in the analysis (Table 1 and Figure 1; see also Supplemental Table S1 at http://ajp.amjpathol.org). Dynamic data exploration of 312 primary invasive carcinomas and 32 TDLUs yielded 242 CpG target variables when the variance filter was tuned to 0.5. Hierarchical clustering revealed an out-group comprising roughly 25% of cancers and manifesting maximal deviation from baseline (Figure 2A). Target deflections from baseline TDLU included both hypomethylations and hypermethylations, and the out-group was subsequently referred to as the methyl-deviator group (Figure 2A). Annotation of the clustered samples for ER status and Nottingham histological grade further suggested that the deviator out-group is substantially enriched for high-grade ER+ cancers (Figure 2A). The least methyl-deviant cancers form a neighboring branch to TDLU and appear to be enriched for ER− cancers (Figure 2A). Subsequently, we calculated an MDI for each sample to use as a metric in group comparisons and survival analyses. The MDI is calculated as the global sum of target methyl deviations in a cancer relative to TDLU baseline, for all targets that meet generic TDLU homogeneity and cancer heterogeneity variance thresholds. By summing the absolute values of target methylation difference between a cancer sample and the baseline, both positive and negative deflections from baseline positively contribute to the MDI score. The MDI captures both the amplitude and frequency of methyl deviation across the cancer genome, while suppressing signals from neutral epigenetic polymorphisms. In our initial comparative analysis of MDI across various sample groups (Figure 2B), the baseline TDLU variance filter was set to 0.7, yielding an overlap set of 109 CpG targets (MDI_109) distributed among 85 discrete genes. Summary statistics of MDI_109 values in clinically relevant cancer subclasses are shown in Figure 2B. This analysis confirmed the impression from the hierarchical clustering that ER+ and ER− cancer groups manifest significant differences in global methylation reprogramming. Notably, ER+ cancers have a greater MDI (P < 0.001), whereas the ER− cancers are the most normal, as in this parameter. Thus, the data exploration revealed significant contrast in global deviation between ER− and ER+ tumors. Coupled with clinical and biological insight that typically regards ER+ and ER− cancers as distinct entities, ER+ and ER− groups were subsequently treated separately for further clinicopathological correlation of methyl deviation. The analysis focused on ER+ cancers revealed a significantly higher MDI among tumors with high-grade histological features and a poor prognosis (Figure 2B and Table 2). A high tumor proliferative index based on Ki-67 staining was correlated with a higher MDI, with borderline significance (P = 0.06). We did not observe a correlation between MDI and Her-2 amplification status of ER+ cancers (P = 0.9).Table 2Cox Multivariate Regression Analysis of Prognostic Factors for Breast CarcinomaPrognostic variableUnivariateMultivariate⁎Variables included in the multivariate analysis were significant by univariate analysis and had data available for 10% of samples.HR95% CIP value†By Wald's test.HR95% CIP value†By Wald's test.LI_590.9141— Low1—— High0.9660.5155–1.81——MI_440.0022820.45764 Low11 High0.3370.1679–0.67810.740.33445–1.6377MDI_109<0.0010.0198 Low11 Intermediate6.6431.527–28.93.9550.8687–18.01 High13.23.091–56.327.5031.6172–34.812HER20.0165— −1—— +5.3131.356–20.82——PR0.358— −1—— +0.490.1066–2.248——Ki-670.0639— Low1—— High4.7150.9141–24.32——Histologic grade0.0022890.188 111 21.9290.8388–4.4381.1160.4388–2.84 34.581.893–11.0792.1150.7793–5.741Age at diagnosis (years)0.7759— <501—— 50–690.7570.3493–1.64—— ≥700.8640.3671–2.036——Stage<0.001<0.001 I and II11 III and IV7.4473.685–15.054.3632.0429–9.316CI, confidence interval; HR, hazard ratio; PR, progesterone receptor; —, did not meet the criteria for inclusion in multivariate analysis. Variables included in the multivariate analysis were significant by univariate analysis and had data available for 10% of samples.† By Wald's test. Open table in a new tab CI, confidence interval; HR, hazard ratio; PR, progesterone receptor; —, did not meet the criteria for inclusion in multivariate analysis. Tuning of the cancer and baseline variance cutoffs was performed to include between 3.5% (MDI_53) and 85% (MDI_1268) of array targets in the MDI calculation (Figure 3A). These adjustments to variance cutoffs had little effect on the performance of MDI as a prognostic metric. For example, the ROC AUC for MDI-based prognosis is approximately 0.78 (Figure 3A), whether the cancer variance is titrated to be more target inclusive (MDI_1268: variance = 0.0, AUC = 0.78) or target restrictive (MDI_53: variance = 0.8, AUC = 0.78). Moreover, all MDI target sets were significantly prognostic for ER+ cancer survival (P < 0.001). In contrast to this relative insensitivity to adjusting the variance filters, prognostic performance is substantially undermined when the TDLU baseline reference is removed and crude methylation levels are summed, as the AUC decreases to 0.60 (Figure 3A, MBI_1505). Even more important, failure to evaluate survival separately for ER+ and ER− groups shifts the AUC to 0.49, totally effacing the prognostic performance of MDI. Thus, we find the following: counting cancer hypomethylations as positive contributors to methyl-deviance computation has a substantial positive impact for methylation-based prognostication; and it is essential to perform MDI-based prognosis separately for ER+ and ER− cancers. Kaplan-Meier survival analysis further showed a significant difference in time to distant recurrence between MDI-low and MDI-high ER+ cancers (Figure 3B). Because MDI target summation captures methyl deviation in cancer as a global process, it does not determine the statistical significance of any given target for association with aggressive cancer biological features. Therefore, MDI targets were individually tested by analysis of variance P values and FDR-based Q scores for significant differences between short- and long-survival ER+ p
Referência(s)