Urinary-cell mRNA profile and acute cellular rejection in kidney allografts

Abstract

Background: The standard test for the diagnosis of acute rejection in kidney transplants is the renal biopsy. Noninvasive tests would be preferable.

Methods: We prospectively collected 4300 urine specimens from 485 kidney-graft recipients from day 3 through month 12 after transplantation. Messenger RNA (mRNA) levels were measured in urinary cells and correlated with allograft-rejection status with the use of logistic regression.

Results: A three-gene signature of 18S ribosomal (rRNA)-normalized measures of CD3ε mRNA and interferon-inducible protein 10 (IP-10) mRNA, and 18S rRNA discriminated between biopsy specimens showing acute cellular rejection and those not showing rejection (area under the curve [AUC], 0.85; 95% confidence interval [CI], 0.78 to 0.91; P<0.001 by receiver-operating-characteristic curve analysis). The cross-validation estimate of the AUC was 0.83 by bootstrap resampling, and the Hosmer-Lemeshow test indicated good fit (P=0.77). In an external-validation data set, the AUC was 0.74 (95% CI, 0.61 to 0.86; P<0.001) and did not differ significantly from the AUC in our primary data set (P=0.13). The signature distinguished acute cellular rejection from acute antibody-mediated rejection and borderline rejection (AUC, 0.78; 95% CI, 0.68 to 0.89; P<0.001). It also distinguished patients who received anti-interleukin-2 receptor antibodies from those who received T-cell-depleting antibodies (P<0.001) and was diagnostic of acute cellular rejection in both groups. Urinary tract infection did not affect the signature (P=0.69). The average trajectory of the signature in repeated urine samples remained below the diagnostic threshold for acute cellular rejection in the group of patients with no rejection, but in the group with rejection, there was a sharp rise during the weeks before the biopsy showing rejection (P<0.001).

Conclusions: A molecular signature of CD3ε mRNA, IP-10 mRNA, and 18S rRNA levels in urinary cells appears to be diagnostic and prognostic of acute cellular rejection in kidney allografts. (Funded by the National Institutes of Health and others.).

Figure 1. Patients, Biopsy Results, and Urine Samples

A total of 4300 urine samples were collected from 485 patients for urinary-cell messenger RNA (mRNA) profiling after transplantation on days 3, 7, 15, and 30; in months 2, 3, 4, 5, 6, 9, and 12; and at the time of kidney-allograft biopsy and 2 weeks thereafter. Of the 4300 urine specimens, 3559 were classified as passing quality control (QC) and 741 were classified as not passing. A total of 220 patients underwent 410 kidney-allograft biopsies, and 265 did not undergo biopsy. The numbers of patients with biopsy-matched urine samples (urine samples that were collected from 3 days before to 1 day after biopsy and that passed QC) are shown for patients with acute cellular rejection (defined as Banff grade IA or higher), for those without any rejection features in the biopsy sample, for those with acute antibody-mediated rejection, for those with borderline changes, and for those with other biopsy findings. The number of patients listed under different diagnostic categories exceeds the 220 patients who underwent biopsy because several patients had multiple specimens with different diagnoses. Among the 265 patients who did not undergo biopsy, 202 met the criteria for stable graft function, of whom 201 had urine samples that passed QC. Patient-enrollment information and recipient and donor characteristics are provided in the Supplementary Appendix.

Figure 2. (facing page). Levels of mRNA in Urinary Cells

Box-and-whisker plots show the log₁₀-transformed ratios of mRNA copies per microgram of total RNA to 18S ribosomal RNA (rRNA) copies (×10^-6) per microgram of total RNA for CD3ε, perforin, granzyme B, interferon-inducible protein 10 (IP-10), CXCR3, CD103, transforming growth factor β1 (TGF-β1), and proteinase inhibitor 9 in 43 urine samples matched to 43 biopsy specimens (from 34 patients) showing acute cellular rejection, 163 urine samples matched to 163 biopsy specimens (from 126 patients) showing no rejection, and 1540 longitudinally collected urine samples from 201 patients with stable graft function who did not undergo biopsy. The horizontal line within each box represents the median, the bottom and top of each box represent the 25th and 75th percentile values, and the I bars represent the 10th and 90th percentile values; the diamond indicates the mean, and circles indicate outliers. The mRNA levels of CD3ε, perforin, granzyme B, and IP-10 differed significantly among the three groups (P<0.001 for all comparisons), but not the levels of CXCR3 (P = 0.06), CD103 (P = 0.13), TGF-β1 (P = 0.11), and proteinase inhibitor 9 (P = 0.38). P values are based on the Kruskal–Wallis test, with the log₁₀-transformed, 18S-normalized mRNA levels treated as the dependent variable. Pairwise group comparisons by means of the Mann–Whitney test showed that the mRNA levels for CD3ε, perforin, granzyme B, and IP-10 in patients with acute cellular rejection were significantly higher than the levels in those with specimens showing no rejection (P<0.001 for each mRNA) and in those with stable graft function (P<0.001 for each mRNA).

Figure 3. Receiver-Operating-Characteristic Curves and Calibration Curve for the Diagnostic Signature

The fraction of true positive results (sensitivity) and the fraction of false positive results (1 - specificity) for the diagnostic signature (calculated from log₁₀-transformed values for 18S-normalized CD3ε mRNA, 18S-normalized IP-10 mRNA, and 18S rRNA) as a biomarker of acute cellular rejection are shown in Panels A and B. Panel C shows the calibration plot based on bootstrap validation; vertical lines at the top of the plot indicate individual observations in the data set. In a comparison of the group of patients who had biopsy specimens showing acute cellular rejection with the group of patients who had biopsy specimens showing no rejection, the area under the curve (AUC) was 0.85 (95% CI, 0.78 to 0.91) (Panel A). In a comparison of the group of patients who had biopsy specimens showing acute cellular rejection with the group of patients who had stable graft function, the AUC was 0.81 (95% CI, 0.75 to 0.87) (Panel B). An analysis that included only specimens matched to biopsies performed because of clinical signs of rejection (38 specimens showing acute cellular rejection vs. 107 showing no rejection) showed that the three-gene signature was diagnostic of acute cellular rejection with 80% specificity (95% CI, 73 to 88) and 79% sensitivity (95% CI, 66 to 92) (AUC, 0.85; 95% CI, 0.79 to 0.92; P<0.001). The combination of perforin and IP-10 mRNAs or perforin and CD3ε mRNAs predicted acute cellular rejection almost as well as the combination of 18S-normalized CD3ε mRNA, 18S-normalized IP-10 mRNA, and 18S rRNA. The addition of perforin to the combination of 18S-normalized CD3ε mRNA, 18S-normalized IP-10 mRNA, and 18S rRNA did not improve the prediction. For the combination of 18S-normalized perforin mRNA, 18S-normalized IP-10 mRNA, and 18S rRNA, the AUC was 0.84 (95% CI, 0.78 to 0.90; P<0.001), and for the combination of 18S-normalized CD3ε mRNA, 18S-normalized perforin mRNA, and 18S rRNA, the AUC was 0.84 (95% CI, 0.76 to 0.91; P<0.001). Bootstrap validation confirmed that the best model consisted of 18S-normalized CD3ε mRNA, 18S-normalized IP-10 mRNA, and 18S rRNA as predictors. Cross-validated estimates of the AUC and calibration-curve intercept and slope were 0.83, −0.06, and 0.92, respectively. The loess-smoothed estimates of the cross-validated and unadjusted calibration curves are overlaid on a diagonal reference line representing perfect model calibration (Panel C). The three-gene model of 18S-normalized CD3ε mRNA, 18S-normalized IP-10 mRNA, and 18S rRNA, with all values log₁₀-transformed, was superior to any of the single-gene models considered. The ROC curve of the three-gene signature discriminating between specimens showing acute cellular rejection and those showing no rejection in the external-validation data set had an AUC of 0.74 (95% CI, 0.61 to 0.86; P<0.001) (Panel D), which was not significantly lower than the AUC of 0.85 in the primary data set (P = 0.13).

Figure 4. Retrospective Trajectory of Diagnostic Signature

The average within-person retrospective trajectory of the diagnostic signature (i.e., the trajectory as a function of the time before biopsy) in urine samples obtained at or before biopsy that passed quality control are shown for the group of 38 patients with first biopsy specimens showing acute cellular rejection (201 urine samples) (Panel A) and the group of 113 patients with specimens showing no rejection (833 urine samples) (Panel B). Only specimens obtained during the first 400 days after transplantation were included. The diagnostic signature remained relatively flat and well below the −1.213 threshold that was diagnostic of acute cellular rejection during the 270 days before biopsy in the group of patients with findings showing no rejection (Panel C). There was a significant difference in the trajectories between the two groups, with a marked increase in the diagnostic signature during the 20-day period before the first specimen showing acute cellular rejection (P<0.001) (Panel D). The y-axis values are diagnostic-signature scores without intrinsic units of measurement; they were calculated from the logistic-regression equation (−6.1487 + 0.8534 log₁₀[CD3ε/18S] + 0.6376 log₁₀[IP-10/18S] + 1.6464 log₁₀[18S]) as follows. Absolute levels of CD3ε mRNA, IP-10 mRNA, and 18S rRNA in the cells from each urine sample were measured by polymerase-chain-reaction assay, with the units of measurement being copies per microgram of total RNA for each mRNA measure and copies (×10⁻⁶) per microgram of total RNA for 18S rRNA. The mRNA copy numbers were 18S-normalized by dividing the mRNA copy number by the 18S rRNA copy number in the same sample, and the ratio was log₁₀-transformed. In all the panels, the black lines indicate the trajectory, the colored bands the 95% confidence interval, and the red lines the diagnostic threshold.