Abstract: SA-PO463
Molecular Diagnosis of Rejection in Formalin Fixed Paraffin Embedded Kidney Transplant Biopsies Is Feasible but Extensively Compromises the RNA
Session Information
- Transplantation: Balancing Rejection and Infection
November 04, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Transplantation
- 1702 Transplantation: Clinical and Translational
Authors
- Famulski, Konrad S., University of Alberta, Edmonton, AB, Alberta, Canada
- Reeve, Jeff, University of Alberta, Edmonton, AB, Alberta, Canada
- Halloran, Philip F., University of Alberta, Edmonton, AB, Alberta, Canada
Background
While molecular assessment of transplant biopsies is usually performed on biopsies immediately frozen or stabilized in RNAlater™, there would be advantages to analyze the FFPE biopsies already prepared for routine histology. We undertook a systematic characterization of microarray gene expression performance with degraded RNA from FFPE compared to intact RNA from RNAlater™ specimens from the same kidney transplant biopsies.
Methods
We compared selected paired 70 FFPE and RNAlater™ samples from 67 patients representing T cell-mediated rejection (n=11), antibody-mediated rejection (n=16) and no rejection in the Molecular Microscope system. The RNA performance was evaluated by quality matrices, rejection-associated transcripts (RATs) obtained from each method and diagnostic classifiers were developed.
Results
The FFPE samples yielded less RNA, and showed massive deterioration in RNA quality compared to RNAlater™, causing false positive and false negative hybridization with the microarray probe sets (Figure 1A, B). Most RNAlater™-RATs were not detectable in FFPE specimens and vice versa, and their association strength (p-value) with rejection was stronger than FFPE-RATs. RNAlater™-RATs and FFPE-RATs distributed rejection phenotypes in Principal Component Analysis similarly, only when tested in their originating material (Figure 1C, D). FFPE classifier of rejection performed in FFPE samples with lower accuracy (71%) compared to RNAlater™ classifier in RNAlater samples (80%).
Conclusion
Molecular system for assessing FFPE samples is possible when RNAlater™ material is not available, but with disadvantages of smaller RNA yields (potential sampling error) and increase in diagnostic errors. Rejection associated transcript sets or classifiers developed with RNAlater™ samples cannot be used for FFPE samples: they must be developed in FFPE samples.
Funding
- Government Support - Non-U.S.