What is the KM plotter?

The Kaplan Meier plotter is capable to assess the effect of 54,675 genes on survival using 10,461 cancer samples. These include 5,143 breast, 1,816 ovarian, 2,437 lung and 1,065 gastric cancer patients with a mean follow-up of 69 / 40 / 49 / 33 months. Primary purpose of the tool is a meta-analysis based biomarker assessment.




For a general citation of the KM-plotter, please use: Szasz AM, Lanczky A, Nagy A, Forster S, Hark K, Green JE, Boussioutas A, Busuttil R, Szabo A, Gyorffy B. Cross-validation of survival associated biomarkers in gastric cancer using transcriptomic data of 1,065 patients. Oncotarget. 2016. DOI: 10.18632/oncotarget.10337 >>> More publications


We thank for the almost 300 users who provided feedback

During a one-month survey in 2016, we asked users about future development of the KM-plotter. We are working on making all the requested issues available. Progress so far:

> Remember filtering settings ... now available as a quick link

> Computation of median survival ... available as a new feature

> Displaying probe ID on the plot ... now included as default setting

> Backward reference ... now available in the settings menu

> miRNA analysis ... now online and also in development

> New cancer type ... coming summer 2017

We will run the next survey in September 2017. --------------------------------------------------------------------------------------------------------------------------

How does it work?

The background database is manually curated. Gene expression data and relapse free and overall survival information are downloaded from GEO (Affymetrix microarrays only), EGA and TCGA. The database is handled by a PostgreSQL server, which integrates gene expression and clinical data simultaneously. To analyze the prognostic value of a particular gene, the patient samples are split into two groups according to various quantile expressions of the proposed biomarker. The two patient cohorts are compared by a Kaplan-Meier survival plot, and the hazard ratio with 95% confidence intervals and logrank P value are calculated. Each database is updated biannually.

Which gene ID can I use?

KM-plot recognizes 54,675 Affymetrix probe set IDs and 70,632 gene symbols (including HUGO Gene Nomenclature Committee approved official gene symbols, previous symbols and aliases - all these are listed in the results page). As the different names can overlap, we recommend to cross-check the identity of the selected gene.

Can I use multiple genes?

Yes. Click on the button "Use multigene classifier" and enter multiple genes. You can run the analysis on all these biomarkers simultaneously (default setting), or using the mean expression of the genes. For this, tick the "Use mean expression of the selected probes" radio button. Maximum 65 genes are allowed.

Can I have a better image?

There are four options: 1) utilize the scalable PDF provided at the results; 2) adjust "Settings" to generate a hi-res TIFF file; 3) use our powerpoint template to change font and the text size; 4) still not satisfied?! -> adjust "Settings" to export plot data as text and format it in any other software.

I have several candidates. Which biomarker is the most significant?

To correct for multiple testing, please use our multiple testing calculator.

Can I use mutation or copy number alterations?

To utilize mutation or CNV data, try our online platform Genotype 2 Outcome, available at the G-2-O website.

Want to predict survival for a single patient?

Try Recurrence Online, a tool capable to predict response to hormonal treatment, to targeted therapy and survival (recurrence score) for breast cancer patients using gene expression data obtained by Affymetrix gene chips.

Are Illumina and Agilent microarrays or RNA-seq datasets included?

No. Affymetrix HG-U133A, HG-U133 Plus 2.0 and HG-U133A 2.0 are included only, because these are regularly used, FDA-cleared and because these arrays have 22,277 probe sets in common. Utilization of the same probe sets enables to measure the same gene with the same sensitivity, specificity and dynamic range.

I have selected gene XXX but the results are for gene YYY - why is this?!

The problem is that the genes symbols are not unambiguous and the HUGO database we use for the selection of the probe sets also includes overlapping gene symbols. Please read following example to understand the phenomena: let's assume we want to measure EPHA3. Once we start typing EPHA3, the system suggests a probe set: 206070_s_at. Now, in case the all probe sets per gene is enabled, the system looks up all gene symbols for 206070_s_at. These are EPHA3, ETK1, HEK4, TYRO4, ETK, and HEK. As all probe sets should be included, the system looks up all 21 probe sets linked to any of these symbols. At the end, again, these will have an updated gene symbol.

The KM-plotter has been utilized among others in studies published in:

KMplot.com Copyright © 2009-2017 | Contact