Predicting liver transplant rejection
The survival rate of liver transplant patients one year after treatment has improved from about 30% in the 1970s to more than 80%, with acute cellular rejection (ACR) the most common complication. It occurs in about 30% of cases and is generally arrested by drug treatment. However, if ACR occurs more than one year after the transplant, survival rates plummet.
The diagnosis of ACR requires a tissue biopsy, which is risky and uncomfortable for the patient. Even then, the interpretation of samples is difficult because the three clinical predictors are not always present.
These problems have prompted scientists in the US to look for a non-invasive alternative diagnosis for ACR and they turned to proteomics for the solution. Michael Charlton and colleagues from the Mayo Clinic and Foundation, Rochester, MN, and the University of Alabama at Birmingham decided to look at the serum proteome to see if any indicators of ACR were detectable.
Human serum contains many high-abundant proteins but, if they are removed before protein analysis, it is possible to detect low-abundant proteins which are transiently present in serum.
So, proteins secreted by cells or produced during cell destruction become visible. These include hormones and cytokines which are transported in serum to their destinations within the human body.
Serum proteome of liver transplant patients
Serum was collected from eight liver transplant patients with a hepatitis C virus infection who appeared to be undergoing ACR at day seven. A non-ACR group comprising eight more liver transplant patients with HCV infection showing no signs of rejection were studied as controls. The patients were age-matched and gender-matched.
The sera were treated with a multiple affinity removal column which eliminated the six most abundant proteins present in serum: albumin, immunoglobulins G and A, antitrypsin, transferrin and haptoglobin.
The proteins remaining in the depleted serum were treated by trypsin digestion in combination with stable isotope-labelling using the iTRAQ reagents for the relative quantitation of peptides in the digests by liquid chromatography-tandem mass spectrometry.
A total of 2801 proteins were detected across all of the samples but only 41 were differentially abundant in the samples of the ACR patients, 28 being more abundant and 13 less abundant, compared with the control set. They were identified by database searching.
The spread of protein function across the set of 41 implied a complex response to ACR but the researchers identified an increase in immune activation as evidenced by several proteins. Raised levels of heat shock proteins 60 and 70, several complement components and the glycoprotein CD33 all supported this conclusion.
These differentially expressed proteins could be used to establish a mass spectrometric procedure for detecting ACR but Charlton was concerned about the associated technical difficulties, as well as the speed and cost. Instead he chose to investigate the ELISA route, because this type of assay is simpler, cheaper and quicker.
ELISA passes the test
A search of the commercial product catalogues revealed that ELISA kits were available for 7 of the 41 differentially expressed proteins, so they were all tested for their ACR predictive ability. The sera from two new groups of liver transplant patients were tested, one showing ACR after 7 days, the other showing no evidence of rejection.
The 7 proteins were serum amyloid A, C4, C1q, C3, heat shock proteins 60 and 70 and fibrinogen. C4 was the most predictive, distinguishing between ACR and non-ACR patients with a sensitivity of 97%, a specificity of 62%, and positive and negative predictive values of 74 and 94%, respectively.
Levels of alanine aminotransferase (ALT), which were measured clinically, were greater in ACR than non-ACR patients at means levels of 198 and 153 U/L, respectively. When these were considered as well as C4, the specificity and positive predictive value were markedly improved to 81 and 86%, respectively, with the other two values about the same.
So, it appears that serum C4 and ALT can be used to predict ACR in liver transplant patients. The researchers were quick to note that the sample size used was small and a far greater number of patients should be tested in a follow-up study.
If the results back up this original work, then a quick, inexpensive, accurate and non-invasive ELISA test for ACR looks to be on the cards.
The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.