Abstract
ABSTRACT. Molecular biologic techniques are currently considered as new diagnostic and prognostic parameters with a sensitivity and specificity exceeding those of histologic and functional data currently used in clinical practice. The results in various clinical settings have been of limited value up to now. This study is an investigation of the use of tissue levels of RNA determined in routine clinical kidney biopsies as prognostic tools. The focus was on RNA encoding for molecules known to be involved in the pathogenesis of renal disorders. Fresh kidney biopsy tissue was obtained from 52 patients with various renal diseases. The GFR was followed for 12 mo. The extent of glomerulosclerosis and interstitial fibrosis in the biopsies was determined with quantitative digital image analysis. Glomerular and tubulointerstitial compartments from each biopsy specimen were separated, and mRNA levels of TGF-β, collagen I, collagen IV, and fibronectin were quantitated by real-time PCR. Correlations, along with 95% confidence intervals (CI), between all variables tested at time biopsy were determined. To assess their prognostic value, these variables were correlated with the slope of GFR within several time intervals after biopsy. In addition, to evaluate the predictive value of the variables for outcome in individual patients, differences for each variable were tested between patients showing progressive decline in renal function (slope GFR < 0) and patients showing stable or improving renal function over time (slope GFR ≥ 0). In chronic renal diseases, the extent of histologic damage correlated with the GFR at the time of biopsy (r = −0.44; CI −0.68 to −0.11), but it did not correlate with the slope expressing a change in GFR after the biopsy. Tubulointerstitial TGF-β mRNA levels correlated with the rate of change in GFR between time of biopsy and 1 mo later (r = 0.41; CI, 0.07 to 0.67). The GFR at the time of biopsy correlated with the slope of change in GFR between time of biopsy and 12 mo later (r = −0.50; CI, −0.73 to −0.18). In chronic renal diseases, glomerular fibronectin mRNA levels, in comparison with the GFR at time of biopsy, correlated relatively strongly with the slope of change in GFR between 3 and 12 mo (r = 0.50; CI, 0.16 to 0.74). Patients with favorable renal outcome after 12 mo showed significantly higher TGF-β mRNA levels and lower proteinuria levels at time of biopsy (P < 0.05) than patients with a progressive decline in renal function. This study shows that mRNA levels measured in kidney biopsies can function as prognostic tools in human renal diseases. In particular, relatively high levels of tubulointerstitial TGF-β mRNA and glomerular fibronectin mRNA are associated with less deterioration in renal function. E-mail: M.Eikmans@LUMC.NL
Chronic renal disease may progress to end-stage renal failure. This event leads to requirement of dialysis or transplantation and is accompanied by high mortality rates and high costs. To determine the individual prognosis of a patient with a renal disease, physicians currently make use of functional parameters such as serum creatinine, GFR, proteinuria, and the histologic information obtained from the renal biopsy. Several studies have shown that the current functional measurements are inaccurate as measures of progression in chronic renal failure (1,2 ⇓). An extensive number of studies have shown that certain histologic changes in the biopsy are associated with an adverse outcome. However, questions have been raised to the use of histologic alterations as predictors of outcome in the individual patient (3–5 ⇓ ⇓). The reason for this especially is that most of these alterations reflect existing tissue damage, which seems to exclude any form of progression. In addition, some renal lesions may be difficult to classify, resulting in a large interobserver and intraobserver variation.
The progression of chronic renal failure is accompanied by a fall in GFR and by an increase in accumulation of extracellular matrix (ECM) in the renal parenchyma, leading to glomerulosclerosis and interstitial fibrosis. Collagen IV, collagen I, and fibronectin are prominent components of the ECM in diseased kidneys (6–8 ⇓ ⇓). Transforming growth factor-β (TGF-β) plays a prime role in renal disease because it is both involved in tissue repair and tissue scarring, via the induction of ECM expression (9–11 ⇓ ⇓). In animal models, mRNA levels for TGF-β and ECM components are upregulated early after the induction of renal disease, and they precede the development of glomerulosclerosis and interstitial fibrosis (12,13 ⇓). Moreover, using renal mRNA levels of molecules involved in matrix deposition as predictors of disease progression has proven to be promising (14–17 ⇓ ⇓ ⇓). In experimental glomerulonephritis, the predictive power of early collagen I mRNA levels for late histologic damage was higher than that of conventional functional and histologic parameters (17). These results warrant an evaluation of the prognostic value of mRNA levels in patients suffering from renal disease.
Materials and Methods
Patients
Between February 1998 and April 2000, 52 native kidney biopsies were collected from 52 patients suffering from renal diseases. Informed consent was given by the patients for use of part of the biopsy for scientific purposes. Patients were divided into two groups: those with chronic renal diseases (n = 37) and those with other renal diseases (n = 15). Diagnoses in the former group were: IgA nephropathy (n = 10), lupus nephritis (n = 7), focal and segmental glomerulosclerosis (n = 5), diabetic nephropathy (n = 2), pauci-immune glomerulonephritis (n = 3), light chain deposit disease (n = 3, two of whom additionally had interstitial nephritis), membranoproliferative glomerulonephritis type I (n = 1), membranous glomerulopathy (n = 1), interstitial nephritis (n = 1), rheumatoid arthritis-related renal disease (n = 1), hypertension-related renal disease (n = 1), end-stage renal disease (n = 2; as a result of diabetic nephropathy or malignant hypertension). Patients with other renal diseases were diagnosed having: minimal change disease (MCD, n = 10), no specific lesions that could account for the sudden deterioration in renal function (n = 3), post-infectious glomerulonephritis (n = 1), or radiation nephropathy (n = 1). Patients in the latter three groups showed a sharp increase in serum creatinine levels before biopsy, which returned to normal within several weeks after biopsy. Patients with MCD showed no specific lesions by light microscopy, and they retained normal renal function during the whole follow-up period. In the study, the patients with chronic renal disease and the patients with other renal diseases were analyzed separately, because we consider the kinetics of these particular disease entities different in terms of underlying molecular processes.
As controls, renal tissue was studied from 16 individuals with normal kidney function and histology. These individuals had no history of renal disease, and kidneys were obtained from Eurotransplant (cadaveric donor kidneys, n = 6) or at autopsy (n = 10).
Clinical Data
Renal function was measured as the GFR by using the Cockcroft-Gault equation (18). The GFR was followed for 12 mo. In the whole patient group, the mean number of available GFR data points for calculating slopes within the intervals between time of biopsy and 1 mo (T0 to 1 mo), T0 to 3 mo, T0 to 6 mo, T0 to 12 mo, and 3 mo to 12 mo were: 6.5 ± 5.1, 10.2 ± 7.4, 13.3 ± 8.8, 15.7 ± 9.6, and 8.0 ± 3.9, respectively. In the group of chronic renal diseases, these numbers were: 7.2 ± 5.4, 11.5 ± 7.5, 14.8 ± 9.1, 15.7 ± 10.1, and 8.0 ± 4.0, respectively. Urine protein was evaluated at time of biopsy and expressed as mg/24 h.
Assessment of Histologic Damage
Glomerulosclerosis.
Paraffin-embedded kidney tissue was sectioned at 4-μm thickness. Glomerulosclerosis was quantitated by evaluation of the glomerular deposition of periodic acid-Schiff (PAS)-positive material in PAS stainings. Digital image analysis was performed using a Zeiss microscope equipped with a full-color 3CCD camera (Sony DXC 950p) and KS-400 software version 3.0 (Zeiss-Kontron, Eichen, Germany). Further details on the method of image analysis have been described in a previous report (19). To calculate the amount of glomerulosclerosis in the PAS stainings, at least six glomeruli were randomly selected and analyzed at 400× magnification. Bowman’s capsules were left out of the analysis. Data are represented as the PAS-positive percentage of the total glomerular area measured.
Interstitial Fibrosis.
The Sirius red staining was used as a measure of the extent of interstitial fibrosis, as described in earlier studies (20,21 ⇓). All sections were stained simultaneously in one session. Digital image analysis was applied to calculate the extent of surface staining by the Sirius red in the sections. Five adjacent microscopic fields were evaluated at 200× magnification. Glomeruli were left out of the analysis. Data are represented as the Sirius red-positive percentage of the “total tubulointerstitial area” measured.
Microdissection and RNA Extraction
For separation of glomeruli and tubulointerstitium in each biopsy, microdissection was performed under a stereomicroscope and according to a previously described protocol (22). For RNA extractions, Trizol Reagent (Life Technologies BRL) was used. RNA from the glomerular and the tubulointerstitial samples was extracted with 200 μl and 500 μl Trizol, respectively. This protocol has been described in a previous report (19).
cDNA Synthesis
RNA extracted from the glomerular and the tubulointerstitial tissues was converted into cDNA with the aid of the Sensiscript-RT kit and the Omniscript-RT kit (Qiagen, Westburg BV, Leusden, The Netherlands), respectively, according to the supplier’s manual. All cDNA reactions were performed in a total volume of 20 μl.
Real-Time PCR
Glomerular cDNA samples were diluted 25 times. Five microliters of each sample was used to measure mRNA levels of TGF-β1, collagen α1(IV), fibronectin, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) by real-time PCR (ABI Prism 7700, Perkin Elmer). Tubulointerstitial cDNA samples were diluted 50 times. Five microliters of each sample was used to measure mRNA levels of TGF-β1, collagen α1(I), collagen α1(IV), fibronectin, and GAPDH by real-time PCR. Sequences for TGF-β1 and fibronectin of the forward primer, reverse primer (Biosource), and probe (Perkin Elmer) are: CCC AGC ATC TGC AAA GCT C, GTC AAT GTA CAG CTG CCG CA, ACA CCA ACT ATT GCT TCA GCT CCA CGG A [TGF-β1]; GGA GAA TTC AAG TGT GAC CCT CA, AGG CAA CGT GTT ACG ATG ATG GGA AGA CAT, and TGC CAC TGT TCT CCT ACG TGG [fibronectin]. Sequences for collagen α1(I), collagen α1(IV), and GADPH have been described in a previous report (19). All measurements were done in duplicate. The real-time PCR protocol has been described in an earlier report (19). The mRNA level of each transcript was measured in one PCR run on a 96-wells plate. A standard was measured in duplicate on the same plate. The accompanying duplicate measurements of the patient samples were performed in a second PCR run on another plate and also included the standard in duplicate.
Correlations at Time of Biopsy
For the patient group with chronic renal diseases (n = 37) and for the patient group with other renal diseases (n = 15), correlations were tested between GFR, proteinuria, the extent of glomerulosclerosis, the extent of interstitial fibrosis, and the mRNA levels, all measured at time of biopsy.
Comparisons of Variables at Time of Biopsy with Outcome
Both in the whole patient group (n = 52) and in patients with chronic renal disease (n = 37), the prognostic value of the mRNA levels, histologic parameters, and renal function at time of biopsy was evaluated.
Correlations with Rate of Changes in GFR.
The mRNA levels, the extent of histologic damage, proteinuria, and the GFR at time of biopsy were correlated with the slopes of the regression lines through GFR data points within the following time intervals: T0 to 1 mo, T0 to 3 mo, T0 to 6 mo, T0 to 12 mo, and 3 mo to 12 mo.
Differences between Patients with Various Outcomes.
We wanted to evaluate the predictive value of clinical, morphologic, and molecular variables at time of biopsy for outcome in individual patients. For this, for each time interval indicated above, patients were separated into a group showing progressive decline in renal function (slope GFR < 0) and a group showing stable or improving renal function over time (slope GFR ≥ 0). Then, GFR, proteinuria, histology, and the mRNA levels were compared between the two groups, and statistical analyses were performed as indicated below.
Statistical Analyses
SPSS software version 10.0 was used for statistical analyses. Data are presented as means ± SD. Differences between controls and patients, and between patient groups with different outcomes, were evaluated by independent t tests. Correlations were evaluated by Pearson bivariate correlation tests (23) and presented with their 95% confidence intervals (CI) in Tables 2 and 3.
Table 2. Correlations at time of biopsya
Table 3. Prognostic value of variables at time of biopsya
Results
Characteristics of Controls and Patients
Table 1 shows clinical variables, histologic variables, and mRNA levels in controls and in patients with renal diseases. GFR is significantly lower (P < 0.05) and proteinuria is significantly higher (P < 0.001) in patients compared with controls. The extent of glomerulosclerosis and interstitial fibrosis (P < 0.001), glomerular mRNA levels of TGF-β (P < 0.001), collagen IV (P < 0.05), and fibronectin (P < 0.001), and tubulointerstitial mRNA levels of TGF-β (P < 0.05) are significantly higher in renal patients than those in controls.
Table 1. Variables at time of biopsya
Correlations at Time of Biopsy
Correlations, with 95% CI, between variables at time of biopsy within the patient group with chronic renal diseases and the patient group with other renal diseases are shown in Table 2. In chronic renal diseases, relatively strong correlations were found between the extent of glomerulosclerosis and interstitial fibrosis (r = 0.56; CI, 0.27 to 0.77), between the extent of interstitial fibrosis and GFR (r = −0.44; CI, −0.68 to −0.11), between glomerular fibronectin mRNA and GFR (r = −0.41; CI, −0.66 to −0.08), and between tubulointerstitial TGF-β mRNA and the extent of glomerulosclerosis (r = −0.38; CI −0.66 to −0.01). In the group with other renal diseases, the strongest correlations found were between glomerular TGF-β mRNA levels and GFR (r = 0.71; CI, 0.09 to 0.94), between glomerular TGF-β mRNA levels and the extent of interstitial fibrosis (r = −0.64; CI, −0.91 to −0.01), between glomerular fibronectin mRNA levels and proteinuria (r = 0.67; CI, 0.12 to 0.91), and between tubulointerstitial TGF-β mRNA levels and GFR (r = −0.67; CI, −0.90 to −0.15).
Correlations of Variables at Time of Biopsy with Rate of Change in GFR
Table 3 shows correlations with 95% CI of the variables at time of biopsy with rate of change in GFR within different time intervals after biopsy, both in the whole patient group and in chronic renal diseases. In the whole patient group, renal function at time of biopsy correlated strongly with GFR slope values between time of biopsy and 1 mo later (Figure 1A; r = −0.63; CI, −0.78 to −0.42), with GFR slope between time of biopsy and 3 mo later (r = −0.60; CI, −0.77 to −0.38), with GFR slope between time of biopsy and 6 mo later (r = −0.51; CI, −0.71 to −0.26), and with GFR slope between time of biopsy and 12 mo later (r = −0.43; CI, −0.65 to −0.15). Tubulointerstitial TGF-mRNA levels were significantly correlated with GFR slope between time of biopsy and 1 mo later (Figure 1B; r = 0.39; CI, 0.10 to 0.63). In chronic renal diseases, several strong correlations were found. For instance, renal function at time of biopsy significantly correlated with GFR slope values between time of biopsy and 1 mo later (Figure 1C; r = −0.57; CI, −0.77 to −0.29), with GFR slope between time of biopsy and 3 mo later (r = −0.45; CI, −0.70 to −0.12), with GFR slope between time of biopsy and 6 mo later (r = −0.44; CI, −0.69 to −0.11), and with GFR slope between time of biopsy and 12 mo later (Figure 1D; r = −0.50; CI, −0.73 to −0.18). Tubulointerstitial TGF-β mRNA levels correlated with the GFR slopes between time of biopsy and 1 mo later (Figure 1E; r = 0.41; CI, 0.07 to 0.67). The extent of histologic damage in the biopsies did not correlate with the change of renal function in time. Glomerular fibronectin mRNA levels, in comparison with the renal function at time of biopsy, correlated relatively strong with the rate of changes in GFR between 3 mo and 12 mo after the biopsy (Figure 1F; r = 0.50; CI, 0.16 to 0.74).
Figure 1. Correlations of variables at time of biopsy with rate of changes in GFR. For each patient, the slope of the regression line through the GFR values between different time intervals after biopsy was calculated. Correlations with 95% confidence intervals (CI) evaluated in the whole patient group (n = 52; A and B) and in chronic renal diseases (n = 37; C, D, E, and F) are shown. (A) Correlation of the GFR at time of biopsy with the rate of change in GFR between time of biopsy and 1 mo later (r = −0.63; CI, −0.78 to −0.42). (B) Correlation of tubulointerstitial TGF-mRNA levels with rate of change in GFR between time of biopsy and 1 mo later (r = 0.39; CI, 0.10 to 0.63). Correlations of the GFR at time of biopsy with (C) the rate of change in GFR between time of biopsy and 1 mo later (r = −0.57; CI, −0.77 to −0.29) and with (D) the rate of change in GFR between time of biopsy and 12 mo later (r = −0.50; CI, −0.73 to −0.18). (E) Correlation of tubulointerstitial TGF-β mRNA levels with rate of change in GFR between time of biopsy and 1 mo later (r = 0.41; CI, 0.07 to 0.67). (F) Correlation of glomerular fibronectin mRNA levels with the rate of change in GFR between 3 mo and 12 mo after biopsy (r = 0.50; CI, 0.16 to 0.74).
Differences in Variables between Patients with Various Outcomes
Patients with stable or improving renal function between T0 and 1 mo (slope GFR ≥ 0) had significantly higher tubulointerstitial TGF-β mRNA levels (Figure 2A; relative mean level of 3.6 ± 2.9 versus 1.6 ± 1.0; P < 0.05), lower urine protein levels at time of biopsy (Figure 2B; 1.9 ± 1.9 mg/24 h versus 4.4 ± 3.2 mg/24 h; P < 0.05), and lower GFR at time of biopsy (Figure 2C; 53.6 ± 32.6 ml/min versus 85.6 ± 44.3 ml/min; P < 0.05) than patients showing progressive decline in renal function (slope GFR < 0). Within the time interval between T0 and 6 mo, patients with favorable prognosis had significantly lower urine protein levels (1.9 ± 1.7 mg/24 h versus 4.7 ± 3.4 mg/24 h; P < 0.01). Patients with stable or improving renal function between T0 and 12 mo had significantly higher TGF-β mRNA levels (Figure 2D; relative level of 3.7 ± 2.9 versus 1.4 ± 1.2; P < 0.05) and lower urine protein levels at time of biopsy (Figure 2E; 2.1 ± 1.7 mg/24 h versus 4.3 ± 3.5 mg/24 h; P < 0.05).
Figure 2. Predictive value of variables at time of biopsy in chronic renal disease for outcome in individual patients. For different time intervals after biopsy, patients with chronic renal diseases were separated into a group showing progressive decline in renal function (slope GFR < 0) and a group showing stable or improving renal function over time (slope GFR ≥ 0). For the time interval between time of biopsy and 1 mo, patient groups differed in (A) tubulointerstitial TGF-β mRNA levels (P < 0.05), in (B) urine protein levels (P < 0.05), and in (C) GFR (P < 0.05). For the time interval between time of biopsy and 12 mo, patient groups differed in (D) tubulointerstitial TGF-β mRNA levels (P < 0.05) and in (E) urine protein levels (P < 0.05).
Discussion
Adequate assessment of the rate of renal disease progression in the individual patient is essential in determining the risk of progression toward end-stage renal failure. The prognostic information obtained from clinical parameters and from renal biopsies has been questioned (2,4,5 ⇓ ⇓). This demands a refinement of the prospective analysis of progressive renal disease by the use of additional, perhaps more suitable, molecular techniques (24,25 ⇓). In experimental kidney disease, measurement of mRNA levels in the renal tissue has proven to be a promising predictor of outcome. Alterations in mRNA synthesis for extracellular matrix (ECM) components and ECM-regulating molecules precede the appearance of histologic lesions (12,26 ⇓). Moreover, levels of collagen mRNA predict for the amount of scarring (17) and for the slope of renal function deterioration (15). Studies in early biopsies taken from patients with renal transplants have demonstrated that cortical mRNA levels are associated with renal function deterioration and might be used as prognostic indicators (27–29 ⇓ ⇓). In sequential protocol transplant biopsies, levels of mRNA for TGF-β, but not renal function, reflect disease progression (29). More recently, a study was conducted evaluating the use of mRNA levels as markers for risk for diabetic nephropathy (30). Although the data are limited, the study shows that glomerular mRNA for collagen IV and connective tissue growth factor can be used to predict progression from normoalbuminuria to microalbuminuria. The findings mentioned above are the rationales for evaluating the applicability of mRNA measurement in biopsy tissue from native kidneys in the clinical setting.
Our main objective was to determine the predictive value of mRNA levels in comparison with functional parameters and histologic data, irrespective of the patient’s diagnosis. It has been hypothesized that progressive kidney diseases develop via a final common pathway irrespective of the original etiology (31,32 ⇓). This common pathway is the process of tissue scar formation (33), which is a result of alterations in ECM synthesis, and is mediated by TGF-β. For this reason, we chose to measure mRNA levels of TGF-β and ECM components in a heterogeneous group of patients with diverse renal diseases. For our study, the entire patient group of 52 patients was divided into a group of 37 patients with chronic renal diseases and a group of 15 patients with other renal diseases. In the latter group, the majority of patients were diagnosed having minimal change disease and did not show any deterioration of GFR over time. Other patients in this group had an acute form of renal disease, showing a renal function which rapidly deteriorated before biopsy, but which normalized within several weeks after biopsy. We believe that the kinetics of these particular entities are different in terms of underlying molecular processes; therefore, correlations between variables at time of biopsy and correlations with prognosis were analyzed separately in patients with chronic renal diseases and in patients with other renal diseases. Separate assessment of correlations at time of biopsy within subgroups indeed showed different results. The finding that in the group of chronic renal diseases the extent of interstitial fibrosis, in comparison with the extent of glomerulosclerosis, relatively strongly correlated with the GFR at time of biopsy confirms earlier reports that tubulointerstitial alterations correlate stronger with loss of renal function at time of biopsy than glomerular alterations (34,35 ⇓).
As an outcome measurement for each patient, we used the rate of change in GFR, which is often used in clinical studies (36,37 ⇓). The rate of change in renal function was determined according to the slope values of regression lines through GFR data points between set time intervals after taking the biopsy. For this study, we have performed a considerable number of pairwise correlation tests. When many non-independent correlations are being made, the possibility of finding significant differences by chance increases as the number of comparisons increases. This means that a multiple comparison problem exists. The use of a Bonferroni correction for multiple correlation testing is rather conservative; we have therefore chosen to present estimated correlation values with corresponding 95% CI rather than with P-values. We assessed correlations of clinical, histologic, and molecular variables with the change in renal function after biopsy. Both in patients with chronic renal diseases and in the whole patient group, TGF-β mRNA levels in the tubulointerstitium correlated positively with the rate of change in GFR between time of biopsy and 1 mo later. In addition, patients who showed stable or improving renal function between time of biopsy and 1 mo had significantly higher levels of tubulointerstitial TGF-β mRNA than patients who showed progressive decline in renal function. A similar observation was made for tubulointerstitial TGF-β mRNA between groups when renal outcome was considered over a time interval between time of biopsy and 12 mo. This means that relatively high tubulointerstitial TGF-β mRNA levels are associated with a favorable prognosis. This finding is in concert with observations from a study in transplanted kidneys, in which we showed that relatively high renal cortical mRNA levels of TGF-β early after transplantation are associated with a stable graft function at later time points (28). Various beneficial effects have been ascribed to TGF-β, such as its role in tissue repair and as an immunosuppressive agent (10,38,39 ⇓ ⇓). The TGF-β mRNA levels measured in the tubulointerstitium may reflect the repair process after tissue injury, but it may also reflect the responsiveness of the patient to therapy within the first month after biopsy. Only in patients with chronic renal diseases, glomerular fibronectin mRNA levels correlated positively with the rate of change in GFR between 3 mo and 12 mo after the biopsy was taken. This correlation was very weak when evaluated in the whole patient group. This shows that the mRNA levels we assessed might be used as prognostic indicators only in certain subsets of renal disease entities. Although no significant difference was found in glomerular fibronectin mRNA levels between patients who showed progressive decline in renal function and patients with favorable outcome of renal function, glomerular fibronectin mRNA levels in the biopsies might still be associated with a relatively favorable prognosis. For example, fibronectin, regulated by the actions of TGF-β (40), may play an important role in compensatory repair mechanisms of the tissue. Indeed, fibronectin has been found to be of major importance during wound healing (41,42 ⇓). In particular, fibronectin mRNA is upregulated in regenerating tissue of damaged kidneys (43).
This study shows application of quantitative real-time PCR on fresh kidney biopsies for assessing the prognostic value of mRNA levels in patients with chronic renal diseases. The method of mRNA measurement may be applicable in other organ disorders, as has been hypothesized and found for instance in the liver (44), the lungs (45), the heart (46), and in cancer research (47,48 ⇓). In conclusion, the extent of histologic damage is negatively correlated with the GFR at time of biopsy, but it does not predict the rate of change in GFR after taking of the biopsy. On the contrary, TGF-β mRNA levels in the tubulointerstitium correlate with the rate of change in renal function of the patient between the time of biopsy and 1 mo later. Above all, patients with a favorable renal outcome between time of biopsy and 12 mo later show significantly higher tubulointerstitial TGF-β mRNA levels than patients who show progressive decline in renal function during the same time period. In chronic renal diseases, glomerular fibronectin correlated better with the rate of changes in GFR between 3 mo and 12 mo than the GFR at time of biopsy did. The current findings indicate that renal mRNA levels may constitute a powerful prognostic tool for outcome in human kidney diseases.
Acknowledgments
This work was supported by the Dutch ‘Praeventiefonds’ (Grant 28–2184–1). The authors thank Dr. I. M. Bajema and Dr. E. L. Lagaaij for critically reading the manuscript. Conclusions derived from this paper have been incorporated into a previous review article (Eikmans et al., Kidney Int 62, 1125–1135, 2002).
- © 2003 American Society of Nephrology
References
Jump to section
More in this TOC Section
Cited By...
- Blood cell mRNAs and microRNAs: optimized protocols for extraction and preservation
- Quantitative Polymerase Chain Reaction-Based Analysis of Podocyturia Is a Feasible Diagnostic Tool in Preeclampsia
- Molecular Correlates of Renal Function in Kidney Transplant Biopsies
- Expression of Surfactant Protein-C, S100A8, S100A9, and B Cell Markers in Renal Allografts: Investigation of the Prognostic Value
- Windows on Renal Biopsy Interpretation: Does mRNA Analysis Represent a New Gold Standard?