Extent, impact, and mitigation of batch effects in tumor biomarker studies using tissue microarrays.

Konrad H Stopsack,Massimo Loda,J Bailey Vaselkiv,Lorelei A Mucci,Stephen P Finn,Kathryn L Penney,Tamara L Lotan,Philip W Kantoff,Svitlana Tyekucheva,Giovanni Parmigiani,Michelangelo Fiorentino,Travis A Gerke,Molin Wang

doi:10.7554/elife.71265

Abstract

Tissue microarrays (TMAs) have been used in thousands of cancer biomarker studies. To what extent batch effects, measurement error in biomarker levels between slides, affects TMA-based studies has not been assessed systematically. We evaluated 20 protein biomarkers on 14 TMAs with prospectively collected tumor tissue from 1448 primary prostate cancers. In half of the biomarkers, more than 10% of biomarker variance was attributable to between-TMA differences (range, 1-48%). We implemented different methods to mitigate batch effects (R package batchtma), tested in plasmode simulation. Biomarker levels were more similar between mitigation approaches compared to uncorrected values. For some biomarkers, associations with clinical features changed substantially after addressing batch effects. Batch effects and resulting bias are not an error of an individual study but an inherent feature of TMA-based protein biomarker studies. They always need to be considered during study design and addressed analytically in studies using more than one TMA.

Highlights

Tissue microarrays (TMAs) were first developed in the 1990s as an efficient way to examine tissue-based biomarkers [1]
To evaluate the presence of batch effects in studies using TMAs, we studied tumor tissue from 1,448 men with primary prostate cancer on 14 TMAs, each including multiple tumor cores from 47 to 158 patients per TMA (Figure 1)
We estimated that across the 20 biomarkers, between-TMA variation explained between 1% and 48% of overall variation in biomarker levels, with half of the biomarkers having ICCs greater than 10%

Summary

Introduction

Tissue microarrays (TMAs) were first developed in the 1990s as an efficient way to examine tissue-based biomarkers [1]. TMAs have been used in thousands of studies to evaluate histologic and molecular biomarkers, mostly in cancer tissue. Even when biomarker assays are well standardized and run conditions are diligently kept fixed, some TMA slides (batches) may have measurements systematically too low or too high, and some batches may have wider spread around the true values of the biomarker than others. Such batch effects can have a profound impact on the validity of biomarker studies, such those using RNA microarrays [3, 4]. Whether such measurement error induces upward or downward bias in results is not guaranteed to follow simple heuristics [5]

Methods

Results

Conclusion