Rapid point-of-care breath test for biomarkers of breast cancer and abnormal mammograms

Abstract

Background: Previous studies have reported volatile organic compounds (VOCs) in breath as biomarkers of breast cancer and abnormal mammograms, apparently resulting from increased oxidative stress and cytochrome p450 induction. We evaluated a six-minute point-of-care breath test for VOC biomarkers in women screened for breast cancer at centers in the USA and the Netherlands.

Methods: 244 women had a screening mammogram (93/37 normal/abnormal) or a breast biopsy (cancer/no cancer 35/79). A mobile point-of-care system collected and concentrated breath and air VOCs for analysis with gas chromatography and surface acoustic wave detection. Chromatograms were segmented into a time series of alveolar gradients (breath minus room air). Segmental alveolar gradients were ranked as candidate biomarkers by C-statistic value (area under curve [AUC] of receiver operating characteristic [ROC] curve). Multivariate predictive algorithms were constructed employing significant biomarkers identified with multiple Monte Carlo simulations and cross validated with a leave-one-out (LOO) procedure.

Results: Performance of breath biomarker algorithms was determined in three groups: breast cancer on biopsy versus normal screening mammograms (81.8% sensitivity, 70.0% specificity, accuracy 79% (73% on LOO) [C-statistic value], negative predictive value 99.9%); normal versus abnormal screening mammograms (86.5% sensitivity, 66.7% specificity, accuracy 83%, 62% on LOO); and cancer versus no cancer on breast biopsy (75.8% sensitivity, 74.0% specificity, accuracy 78%, 67% on LOO).

Conclusions: A pilot study of a six-minute point-of-care breath test for volatile biomarkers accurately identified women with breast cancer and with abnormal mammograms. Breath testing could potentially reduce the number of needless mammograms without loss of diagnostic sensitivity.

Conflict of interest statement

Competing Interests: Michael Phillips is President and CEO of Menssana Research, Inc., which develops and patents breath tests for biomarkers of diseases. Renee N Cataneo, Peter D Kaplan, Mayur Mundada and Urvish Patel are current or former employees of Menssana Research, Inc. Schmitt & Associates, Newark, NJ, analyzed the data. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Figures

Figure 1. Breath test outcome in healthy women (normal screening mammogram) versus women with breast biopsy positive for cancer.

Identification of breath biomarkers (upper panel): A list of candidate breath biomarkers of disease was obtained by segmenting chromatograms into a time series of alveolar gradients, where the alveolar gradient comprised detector response in breath minus corresponding detector response in room air. The diagnostic accuracy of each candidate biomarker was quantified as the area under curve (AUC) of its associated receiver operating characteristic (ROC) curve. This figure displays the number of candidate biomarkers (y-axis) as a function of their diagnostic accuracy (x-axis). The “correct” curve employed the correct assignment of diagnosis (normal mammogram or cancer on biopsy). The “random” curve employed multiple Monte Carlo simulations comprising 40 random assignments of diagnosis in order to determine the random behavior of each candidate biomarker. The horizontal separation between the “correct” and “random” curves varies with the amount of diagnostic information in the breath signal. Where the number in the “random” curve declines to <1, its vertical distance from the “correct” curve identifies the excess number of candidate biomarkers that identified the disease group with greater than random accuracy. The number of apparent biomarkers with greater than random accuracy exceeded 30, but several segments were closely adjacent in the time series, consistent with approximately 10 biomarker peaks in the chromatogram. Similar analyses were also performed in normal versus abnormal screening mammograms and cancer versus no cancer on breast biopsy in order to develop separate algorithms. Diagnostic accuracy of the breath test (lower panel): The ROC curve displays the breath test's accuracy in distinguishing healthy women with a normal screening mammogram from women whose breast biopsy was positive for cancer. The breath test employed a multivariate predictive algorithm derived from the biomarkers with greater than random accuracy that were identified with the Monte Carlo simulations in the left panel. Sensitivity and specificity values were determined from the point on the ROC curve where their sum was maximal. Cross-validation of predicted outcomes is shown in red ROC curve. A repeated leave-one-out bootstrap method was employed to estimate the prediction error (method described in text).

Figure 2. Breath test outcomes in screening mammography and in breast biopsy.

Identification of breath biomarkers, determination of diagnostic accuracy of the breath test, and LOO cross-validation of predicted outcomes were performed in the same fashion as described in Figure 1. Sensitivity and specificity values were determined from the point on the ROC curve where their sum was maximal. The left panel displays comparison of women with normal and abnormal screening mammograms, and the right panel displays comparison of women with cancer and no cancer on breast biopsy.