Limits...
Cancer metastasis networks and the prediction of progression patterns.

Chen LL, Blumm N, Christakis NA, Barabási AL, Deisboeck TS - Br. J. Cancer (2009)

Bottom Line: Network modelling may allow the incorporation of the temporal dimension in the analysis of these patterns.These cancer metastasis networks capture both temporal and subtle relational information, the dynamics of which differ between cancer types.Using these networks as entities on which the metastatic disease of individual patients may evolve, we show that they may be used, for certain cancer types, to make retrograde predictions of a primary cancer type given a sequence of metastases, as well as anterograde predictions of future sites of metastasis.

View Article: PubMed Central - PubMed

Affiliation: Complex Biosystems Modeling Laboratory, Harvard-MIT (HST) Athinoula A Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA 02129, USA.

ABSTRACT

Background: Metastasis patterns in cancer vary both spatially and temporally. Network modelling may allow the incorporation of the temporal dimension in the analysis of these patterns.

Methods: We used Medicare claims of 2,265,167 elderly patients aged > or = 65 years to study the large-scale clinical pattern of metastases. We introduce the concept of a cancer metastasis network, in which nodes represent the primary cancer site and the sites of subsequent metastases, connected by links that measure the strength of co-occurrence.

Results: These cancer metastasis networks capture both temporal and subtle relational information, the dynamics of which differ between cancer types. Using these networks as entities on which the metastatic disease of individual patients may evolve, we show that they may be used, for certain cancer types, to make retrograde predictions of a primary cancer type given a sequence of metastases, as well as anterograde predictions of future sites of metastasis.

Conclusion: Improvements over traditional techniques show that such a network-based modelling approach may be suitable for studying metastasis patterns.

Show MeSH

Related in: MedlinePlus

Prediction of the primary cancer site from a sequence of metastases. The primary cancer types for which the true positive rates exceed 25% from each model are shown. The multinomial logistic regression (MLR) algorithm takes into account the number of patients in the respective categories, and therefore, a relatively rare cancer type will be classified as a common cancer type with similar metastasis patterns. The MLR algorithm and the network algorithm perform in different ways: the MLR classifies everything into a few common cancer types, whereas the network algorithm is able to differentiate between rarer cancer types.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2736851&req=5

fig4: Prediction of the primary cancer site from a sequence of metastases. The primary cancer types for which the true positive rates exceed 25% from each model are shown. The multinomial logistic regression (MLR) algorithm takes into account the number of patients in the respective categories, and therefore, a relatively rare cancer type will be classified as a common cancer type with similar metastasis patterns. The MLR algorithm and the network algorithm perform in different ways: the MLR classifies everything into a few common cancer types, whereas the network algorithm is able to differentiate between rarer cancer types.

Mentions: Rather than classifying patients into one of the six major cancer types, the network model for predicting the primary cancer site classifies patients into many more categories. Eleven cancer types achieved a true positive rate of >25%, most of which are less common cancers (Supplementary Table S4). For example, although almost all patients with colon cancer were classified into other categories, ovary had a true positive rate of 81%, hypopharynx 75%, male breast 70%, pleura 46%, pancreas 40%, small intestine 39%, female breast 37%, male genital 33%, lung and bronchus 28%, cervix 27%, and female genital 26%. Even though the overall accuracy may be less than that of the MLR algorithm, the network model has the advantage of broader specificity and sensitivity toward cancers of less common sites (Supplementary Table S5). The true positive rates for those sites exhibiting true positive rates >25% with either method are shown in Figure 4.


Cancer metastasis networks and the prediction of progression patterns.

Chen LL, Blumm N, Christakis NA, Barabási AL, Deisboeck TS - Br. J. Cancer (2009)

Prediction of the primary cancer site from a sequence of metastases. The primary cancer types for which the true positive rates exceed 25% from each model are shown. The multinomial logistic regression (MLR) algorithm takes into account the number of patients in the respective categories, and therefore, a relatively rare cancer type will be classified as a common cancer type with similar metastasis patterns. The MLR algorithm and the network algorithm perform in different ways: the MLR classifies everything into a few common cancer types, whereas the network algorithm is able to differentiate between rarer cancer types.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2736851&req=5

fig4: Prediction of the primary cancer site from a sequence of metastases. The primary cancer types for which the true positive rates exceed 25% from each model are shown. The multinomial logistic regression (MLR) algorithm takes into account the number of patients in the respective categories, and therefore, a relatively rare cancer type will be classified as a common cancer type with similar metastasis patterns. The MLR algorithm and the network algorithm perform in different ways: the MLR classifies everything into a few common cancer types, whereas the network algorithm is able to differentiate between rarer cancer types.
Mentions: Rather than classifying patients into one of the six major cancer types, the network model for predicting the primary cancer site classifies patients into many more categories. Eleven cancer types achieved a true positive rate of >25%, most of which are less common cancers (Supplementary Table S4). For example, although almost all patients with colon cancer were classified into other categories, ovary had a true positive rate of 81%, hypopharynx 75%, male breast 70%, pleura 46%, pancreas 40%, small intestine 39%, female breast 37%, male genital 33%, lung and bronchus 28%, cervix 27%, and female genital 26%. Even though the overall accuracy may be less than that of the MLR algorithm, the network model has the advantage of broader specificity and sensitivity toward cancers of less common sites (Supplementary Table S5). The true positive rates for those sites exhibiting true positive rates >25% with either method are shown in Figure 4.

Bottom Line: Network modelling may allow the incorporation of the temporal dimension in the analysis of these patterns.These cancer metastasis networks capture both temporal and subtle relational information, the dynamics of which differ between cancer types.Using these networks as entities on which the metastatic disease of individual patients may evolve, we show that they may be used, for certain cancer types, to make retrograde predictions of a primary cancer type given a sequence of metastases, as well as anterograde predictions of future sites of metastasis.

View Article: PubMed Central - PubMed

Affiliation: Complex Biosystems Modeling Laboratory, Harvard-MIT (HST) Athinoula A Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA 02129, USA.

ABSTRACT

Background: Metastasis patterns in cancer vary both spatially and temporally. Network modelling may allow the incorporation of the temporal dimension in the analysis of these patterns.

Methods: We used Medicare claims of 2,265,167 elderly patients aged > or = 65 years to study the large-scale clinical pattern of metastases. We introduce the concept of a cancer metastasis network, in which nodes represent the primary cancer site and the sites of subsequent metastases, connected by links that measure the strength of co-occurrence.

Results: These cancer metastasis networks capture both temporal and subtle relational information, the dynamics of which differ between cancer types. Using these networks as entities on which the metastatic disease of individual patients may evolve, we show that they may be used, for certain cancer types, to make retrograde predictions of a primary cancer type given a sequence of metastases, as well as anterograde predictions of future sites of metastasis.

Conclusion: Improvements over traditional techniques show that such a network-based modelling approach may be suitable for studying metastasis patterns.

Show MeSH
Related in: MedlinePlus