Concordance between cancer gene alterations in tumor and circulating tumor DNA correlates with poor survival in a real-world precision-medicine population

Abstract

Genomic analysis, performed on tumoral tissue DNA and on circulating tumor DNA (ctDNA) from blood, is the cornerstone of precision cancer medicine. Herein, we characterized the clinical prognostic implications of the concordance of alterations in major cancer genes between tissue- and blood-derived DNA in a pan-cancer cohort. The molecular profiles of both liquid (Guardant Health) and tissue (Foundation Medicine) biopsies from 433 patients were analyzed. Mutations and amplifications of cancer genes scored by these two tests were assessed. In 184 (42.5%) patients, there was at least one mutual gene alteration. The mean number of mutual gene-level alterations in the samples was 0.67 per patient (range: 0-5). A higher mutual gene-level alteration number correlated with shorter overall survival (OS). As confirmed in multivariable analysis, patients with ≥2 mutual gene-level alterations in blood and tissue had a hazard ratio (HR) of death of 1.49 (95% confidence interval [CI]=1-2.2; P=0.047), whereas patients with ≥3 mutual gene-level alterations had an HR of death 2.38 (95% CI=1.47-3.87; P=0.0005). Together, our results show that gene-level concordance between tissue DNA and ctDNA analysis is prevalent and is an independent factor predicting significantly shorter patient survival.

Keywords: cancer; circulating DNA; genomics; survival; tissue DNA.

Fig. 1

The frequency of mutation‐level concordance given gene‐level concordance. Only genes with gene concordance in more than five samples are shown. Gene‐level concordance implies that the same gene (but not necessarily the same locus) is aberrant in both ctDNA and tissue; mutation‐level concordance implies that the same locus in the same gene is aberrant in both ctDNA and tissue.

Fig. 2

Survival analysis in relation to the number of concordant genes (gene‐level concordance). (A) Kaplan–Meier, univariate analysis. The plots show that the greater the number of gene‐level alteration concordance in tissue and blood ctDNA, the shorter the survival. Number of samples in each mutual gene number category are provided in the right legend. (B) Multivariable analysis (Cox regression). The figure shows that greater gene‐level alteration concordance in tissue and blood (denoted as “mutual gene number”) was associated with significantly shorter survival, independent of other factors. All parameters aside from sex and cancer type were assessed as continuous variables. Younger age, longer time between tests, and lower ctDNA percent was associated with longer survival times, whereas brain and hepatopancreatobiliary cancers were associated with shorter survival times. (Note that the box overlaps the “1” line at times, because of a graphical limitation, although the P value [and 95% CI] are significant.) CI, confidence interval; ctDNA, circulating tumor DNA; HR, hazard ratio; TMB, tumor mutational burden. P values: *: 0.01–0.05, ***: 0–0.001.

Fig. 3

Multivariable survival analysis (Cox regression) with different cutoffs for gene‐level concordance. Regardless, if the cutoff was ≥2 vs <2 or ≥3 vs <3, greater gene‐level concordance is independently associated with shorter survival. (A) Patients with two or more concordant altered genes vs one or zero concordant genes. (B) Patients with three or more concordant genes vs zero to two concordant genes. CI, confidence interval; ctDNA, circulating tumor DNA; HR, hazard ratio; TMB, tumor mutational burden. P values: *: 0.01–0.05, **: 0.001–0.01, ***: 0–0.001.

Fig. 4

Survival analysis of mutual gene‐level and mutual mutation‐level alteration concordance status between blood and tissue. N_N, no mutual gene‐level nor mutual mutation‐level alteration concordance; Y_N, at least one mutual gene‐level but no mutual mutation‐level alteration concordance; Y_Y, at least one mutual gene‐level and at least one mutual mutation‐level alteration concordance between blood and tissue. (A) Kaplan–Meier, univariate analysis. The plot shows that there is no difference in survival between patients with no mutual gene‐level nor mutual mutation‐level alteration concordance (N_N) and at least one mutual gene‐level but no mutual mutation‐level alteration concordance (Y_N) (P=0.6); however, patients with at least one mutual gene‐level and at least one mutual mutation‐level alteration concordance between blood and tissue (Y_Y) have significantly shorter survival than the first two groups (Y_Y vs N_N, P=2.64e‐7; Y_Y vs Y_N, P=0.0015, Cox univariate test). Examining all three curves confirms this observation (P = 4.35e‐7), as shown in the graphic. These results suggest that it is mutation‐level alteration concordance that is important for predicting outcome in univariate analysis. The number of samples in each mutual gene‐mutation status category are provided in the right legend. (B) Multivariable analysis (Cox regression). In multivariable analysis, the difference between the concordance groups was not significant (the lack of impact of mutation level concordance on outcome in multivariable analysis could be attributed to the fact that only 22 tumors had mutual genes but not mutual gene mutation locus concordance). CI, confidence interval; ctDNA, circulating tumor DNA; HR, hazard ratio; TMB, tumor mutational burden. P values: *: 0.01–0.05, **: 0.001–0.01, ***: 0–0.001.

Fig. 5

(A) Correlation plot of the independent variables tested. Balloon plot that represents the residuals of the Chi‐square test. Blue implies positive correlation between factors and red implies negative correlation. Intensity of the color reflects the magnitude of the correlation (greater intensity = greater degree of correlation). Mutual status is defined in three levels: N_N, – no mutual gene‐level nor mutual mutation‐level alteration concordance; Y_N, at least one mutual gene‐level but no mutual mutation‐level alteration concordance; Y_Y, at least one mutual gene‐level and at least one mutual mutation‐level alteration concordance between blood and tissue. (B) ctDNA gene number compared to the mutual status of gene‐level and mutation‐level alteration concordance between blood and tissue. Central line represents median value and the box represents the range of 1^st to 3^rd quartiles. The graphic shows that a higher ctDNA gene alteration number is associated with a greater number of mutual gene‐level alteration concordance (Y‐N) between blood and tissue and an even greater number of mutation‐level alteration concordance between tissue and blood (Y‐Y). Number of samples in each mutual gene‐mutation status category are provided in the right legend. ctDNA, circulating tumor DNA; TMB, tumor mutational burden.