Dataset Information

Development of a ten-signature classifier using a support vector machine integrated approach to subdivide the M1 stage into M1a and M1b stages of nasopharyngeal carcinoma with synchronous metastases to better predict patients' survival.

ABSTRACT: The aim of this study was to develop a prognostic classifier and subdivided the M1 stage for nasopharyngeal carcinoma patients with synchronous metastases (mNPC). A retrospective cohort of 347 mNPC patients was recruited between January 2000 and December 2010. Thirty hematological markers and 11 clinical characteristics were collected, and the association of these factors with overall survival (OS) was evaluated. Advanced machine learning schemes of a support vector machine (SVM) were used to select a subset of highly informative factors and to construct a prognostic model (mNPC-SVM). The mNPC-SVM classifier identified ten informative variables, including three clinical indexes and seven hematological markers. The median survival time for low-risk patients (M1a) as identified by the mNPC-SVM classifier was 38.0 months, and survival time was dramatically reduced to 13.8 months for high-risk patients (M1b) (P < 0.001). Multivariate adjustment using prognostic factors revealed that the mNPC-SVM classifier remained a powerful predictor of OS (M1a vs. M1b, hazard ratio, 3.45; 95% CI, 2.59 to 4.60, P < 0.001). Moreover, combination treatment of systemic chemotherapy and loco-regional radiotherapy was associated with significantly better survival outcomes than chemotherapy alone (the 5-year OS, 47.0% vs. 10.0%, P < 0.001) in the M1a subgroup but not in the M1b subgroup (12.0% vs. 3.0%, P = 0.101). These findings were validated by a separate cohort. In conclusion, the newly developed mNPC-SVM classifier led to more precise risk definitions that offer a promising subdivision of the M1 stage and individualized selection for future therapeutic regimens in mNPC patients.

SUBMITTER: Jiang R

PROVIDER: S-EPMC4823134 | biostudies-literature | 2016 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development of a ten-signature classifier using a support vector machine integrated approach to subdivide the M1 stage into M1a and M1b stages of nasopharyngeal carcinoma with synchronous metastases to better predict patients' survival.

Jiang Rou R You Rui R Pei Xiao-Qing XQ Zou Xiong X Zhang Meng-Xia MX Wang Tong-Min TM Sun Rui R Luo Dong-Hua DH Huang Pei-Yu PY Chen Qiu-Yan QY Hua Yi-Jun YJ Tang Lin-Quan LQ Guo Ling L Mo Hao-Yuan HY Qian Chao-Nan CN Mai Hai-Qiang HQ Hong Ming-Huang MH Cai Hong-Min HM Chen Ming-Yuan MY

Oncotarget 20160101 3

The aim of this study was to develop a prognostic classifier and subdivided the M1 stage for nasopharyngeal carcinoma patients with synchronous metastases (mNPC). A retrospective cohort of 347 mNPC patients was recruited between January 2000 and December 2010. Thirty hematological markers and 11 clinical characteristics were collected, and the association of these factors with overall survival (OS) was evaluated. Advanced machine learning schemes of a support vector machine (SVM) were used to se ...[more]

PMID: 26636646

Similar Datasets

Project description:BackgroundIn 2009, the American Joint Committee on Cancer version 7 staging system introduced the M1 subclassifications M1a (single metastatic site) and M1b (peritoneal or multiple metastatic sites). The study objectives were to evaluate the prognostic effect of site of metastasis and M1a/b category among patients with newly diagnosed colorectal cancer and synchronous metastasis.Patients and methodsPatients with newly diagnosed pathologic or clinical category M1 colorectal cancer referred to the British Columbia Cancer Agency between 1999 and 2007 were included. Demographic, tumor, treatment, and outcome data were prospectively collected, and prognostic factors were identified. Univariate Cox models were used to assess the prognostic impact of individual sites of metastasis and to determine the effect of M1a/b category on overall survival (OS).ResultsAmong 2,049 eligible patients, 70% had M1a and 30% M1b category disease. The most common sites of common single sites of metastasis included liver (56%), lung (5.3%), and peritoneum (3.6%). Metastasis to a single organ or site, including peritoneum, was associated with improved OS compared with multiple sites of metastasis. In multivariate analysis, M1b category conferred inferior survival and hazard ratio (HR) 1.38 (95% confidence interval [CI]: 1.22, 1.55), along with age >70 and Eastern Cooperative Oncology Group performance status of 3-4. Resection of primary tumor was associated with improved survival, HR 0.46 (95% CI: 0.41, 0.52). Results were similar in subgroup analysis of patients undergoing resection of their primary tumor when histology, tumor, and node category were included.ConclusionThe results lend support to the introduction of M1a/b colorectal cancer categories. Consideration may be given to classifying patients with solitary peritoneal metastasis only as M1a rather than M1b category. Further refinement of category M1a to reflect resectability of metastasis at initial diagnosis may improve prognostication.

Project description:BACKGROUND:Breast cancer is a very heterogeneous disease and some patients are cured by the surgical removal of the primary tumour whilst other patients suffer from metastasis and spreading of the disease, despite adjuvant therapy. A number of prognostic and treatment predictive factors have been identified such as tumour size, oestrogen (ER) and progesterone (PgR) receptor status, human epidermal growth factor receptor type 2 (HER2) status, histological grade, Ki67 and age. Lymph node involvement is also assessed during surgery to determine if the tumour has spread which requires dissection of the axilla and adjuvant treatment. The prognostic and treatment predictive factors assessing the nature of the tumour are all routinely based on the status of the primary tumour. RESULTS:We have analysed a unique tumour set of fourteen primary breast cancer tumours with matched synchronous axillary lymph node metastases and a set of nine primary tumours with, later developed, matched distant metastases from different sites in the body. We used a pairwise tumour analysis (from the same individual) since the difference between the same tumour-type in different patients was greater. Glycopeptide capture was used in this study to selectively isolate and quantify N-linked glycopeptides from tumours mixtures and the captured glycopeptides were subjected to label-free quantitative tandem mass spectrometry analysis. Differentially expressed proteins between primary tumours and matched lymph node metastasis and distant metastasis were identified. Two of the top hits, ATPIF1 and tubulin ?-chain were validated by immunohistochemistry to be differentially regulated. CONCLUSIONS:We show that the expression of a large number of glycosylated proteins change between primary tumours and matched lymph node metastases and distant metastases, confirming that cancer cells undergo a molecular transformation during the spread to a secondary site. The proteins are part of important pathways such as cell adhesion, migration pathways and immune response giving insight into molecular changes needed for the tumour to spread. The large difference between primary tumours and lymph node and distant metastases also suggest that treatment should be based on the phenotype of the lymph node and distant metastases.

Dataset Information

Development of a ten-signature classifier using a support vector machine integrated approach to subdivide the M1 stage into M1a and M1b stages of nasopharyngeal carcinoma with synchronous metastases to better predict patients' survival.

Publications

Development of a ten-signature classifier using a support vector machine integrated approach to subdivide the M1 stage into M1a and M1b stages of nasopharyngeal carcinoma with synchronous metastases to better predict patients' survival.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure