Impact of centralization on aCGH-based genomic profiles for precision medicine in oncology

Abstract

Background: Comparative genomic hybridization (CGH) arrays are increasingly used in personalized medicine programs to identify gene copy number aberrations (CNAs) that may be used to guide clinical decisions made during molecular tumor boards. However, analytical processes such as the centralization step may profoundly affect CGH array results and therefore may adversely affect outcomes in the precision medicine context.

Patients and methods: The effect of three different centralization methods: median, maximum peak, alternative peak, were evaluated on three datasets: (i) the NCI60 cell lines panel, (ii) the Cancer Cell Line Encyclopedia (CCLE) panel, and (iii) the patients enrolled in prospective molecular screening trials (SAFIR-01 n = 283, MOSCATO-01 n = 309), and compared with karyotyping, drug sensitivity, and patient-drug matching, respectively.

Results: Using the NCI60 cell lines panel, the profiles generated by the alternative peak method were significantly closer to the cell karyotypes than those generated by the other centralization strategies (P < 0.05). Using the CCLE dataset, selected genes (ERBB2, EGFR) were better or equally correlated to the IC50 of their companion drug (lapatinib, erlotinib), when applying the alternative centralization. Finally, focusing on 24 actionable genes, we observed as many as 7.1% (SAFIR-01) and 6.8% (MOSCATO-01) of patients originally not oriented to a specific treatment, but who could have been proposed a treatment based on the alternative peak centralization method.

Conclusion: The centralization method substantially affects the call detection of CGH profiles and may thus impact precision medicine approaches. Among the three methods described, the alternative peak method addresses limitations associated with existing approaches.

Keywords: aCGH; comparative genomic hybridization; precision medicine; targeted therapy.

Figure 1.

Distance from karyotypes. (A–C) In order to estimate the effect of the centralization methods, squared distances between genomic profiles (colored line) and karyotypes (green line) have been calculated. Distances are symbolized by the colored areas. (D) An alternative peak was available for 18 of the 57 analyzed cell lines (bold black points). Choosing the alternative peak for adjusting the genomic profiles significantly reduce the discrepancies with the corresponding karyotypes, compared with the other methods. P = 1.13e−4 and 6.35e−5, compared with the maximum peak and the LogR median, respectively. Vertical colored curves represent the densities, and horizontal gray segments are the Q25, Q50, and Q75 quantiles of each distribution.

Figure 2.

Correlation between copy number variation and sensitivity to related inhibitors. The Spearman correlation between ERBB2 and lapatinib increased significantly when applying the maximum peak or the alternative peak centralizations, compared with the median value adjustment (ρ = 0.122, 0.156, and 0.012, respectively. P = 0.043 and 0.012, respectively). The alternative centralization even improved the correlation, but not significantly (P = 0.294). The same trend was observed for EGFR and erlotinib, even though none of the improvements appeared significant.

Figure 3.

Effect of different genomic profiles centralization methods on possible therapeutic orientations. Top panel: Centralizing on an alternative peak leaded to identify actionable amplifications (new calls) in 20 more samples (7.1%) in the SAFIR01 data, and in 21 supplementary samples (6.8%) in MOSCATO, for whom no amplification was found by using the maximum peak for centralizing the genomic profiles. In 22 (7.8%) and 6 (1.9%) cases in SAFIR01 and MOSCATO, respectively, supplementary amplifications (sup. calls) were also identified, leading to supplementary options for a therapeutic decision making. For 42% and 68%, in Safir01 and Moscato, respectively, no therapeutic option appeared, and in 43.1% and 23.3%, the same actionable genes were identified with both methods. Bottom panel: FGF4, CCND1, and RPTOR were the most frequently newly detected amplified genes in Safir01, while EGFR and FGFR1, principally, were impacted by the centralization strategy in Moscato data (frequencies are summarized in the table, on right).