Background
Autosomal dominant polycystic kidney disease (ADPKD) is one of the most commonly inherited disorders in humans, with an estimated prevalence of 1:400 to 1:1,000 [
1]. ADPKD is the fourth most common cause of renal replacement therapy (i.e., dialysis or transplant) [
2‐
4] and is generally diagnosed by imaging of the kidney using ultrasonography (US), computed tomography (CT), or magnetic resonance imaging (MRI). For ADPKD, approximately 75–85% of cases are caused by variants in
PKD1, which encodes the protein known as polycystin-1 [
5‐
7]. Most human families with ADPKD have novel variants and over 1,273 causal variants for
PKD1 are catalogued in the Autosomal Dominant Polycystic Kidney Disease Mutation Database: PKDB [
8]. Because of this genetic heterogeneity, cohorts with the same
PKD1 variant or cohorts with the same genetic backgrounds are limited in humans, which inhibits the power of studies focused on genetic modifiers that would influence interfamilial, intrafamilial and sex differences in disease progression and responses to therapeutics. Genetic screening is also complicated in humans since
PKD1 includes 46 exons, has a large, ~ 14 kb mRNA [
9] spanning a 47.2 kb genomic region, and six pseudogenes are present in the human genome [
10,
11].
ADPKD is widely recognized as the most commonly inherited renal disease in the domestic cat, specifically cats of the Persian breed [
12‐
16]. The feline
PKD1 variant (c.10063C > A) causes a stop codon at position 3284 in exon 29 (C3284X) [
17] and is the only variant causing ADPKD in cats known to date. This variant is found in Persian-related breeds as well [
17‐
19]. Hepatic and pancreatic cysts are present in some cats with ADPKD [
14,
16,
20], however, hypertension is noted to be minor [
21], and other vascular or systemic complications are not documented. Although many ADPKD cats remain subclinical through-out their lives, some show rapid disease progression, developing chronic kidney disease (CKD) secondary to ADPKD, and succumb to disease within 7 years of life or earlier, which is only mid-life for a cat [
16,
22]. These younger cats that succumb to disease have consistent disease progressing to humans who have truncating
PKD1 variants [
5]. Cats should be instrumental for identifying genetic modifiers of cystic progression and for deciphering variation in therapeutic responses since the solitary causal variant and control of the genetic background via colony of ADPKD cats will reduce variables in the analyses.
US, CT, and MRI are commonly used imaging modalities to evaluate disease progression and the therapeutic efficacy in humans with ADPKD [
23], however, detailed and comparative imaging in feline ADPKD using different modalities is limited. US is routinely used to diagnose feline ADPKD, however, a study assessing the progression of the disease over time (approximately 1 year) showed an apparent improvement in a small number of cats [
24], thus, the accuracy of US is questionable. In humans, US is used to screen ADPKD-suspected patients to follow changes over long periods of time, while CT and MRI are used to quantify changes in kidney parenchyma over shorter intervals [
23]. In rodents, imaging accurately reflects kidney volumes, however, due to the small sizes of rodent kidneys, accurate evaluations of fractional cyst volume (FCV) are difficult and limited without the use of ultra-high field MRI (7 T and above) [
25]. CT and MRI are the routine modalities used to evaluate the interventional efficacy of newly developed drugs in humans. MRI has not been used to evaluate ADPKD in cats, thus, baseline imaging studies could support the cats’ role for evaluating therapeutics.
Three imaging modalities, US, CT, and MRI, are used to examine variation in disease presentation and disease progression in feline ADPKD. Imaging was compared to well-known biomarkers for CKD, such as urine specific gravity (USG), blood urea nitrogen (BUN), serum creatinine (sCr), glomerular filtration rate (GFR) and symmetric dimethylarginine (SDMA), which is a novel biomarker for feline CKD at the earlier stage [
26]. Total kidney volume (TKV), total cystic volume (TCV) and FCV were determined for the first time in ADPKD cats, demonstrating the wide variation in disease presentation, the potential identification of rapid and slow progression individual cats and the potential to evaluate therapeutic interventions.
Methods
Subjects
The cats represented an ADPKD cat colony housed at the University of Missouri (MU), which has been maintained for over 20 years. Five Persian cats with ADPKD were originally donated by private owners to establish the colony. All animal procedures were conducted in accordance with the National Research Council Guide for the Care and Use of Laboratory Animals and were approved by the MU Animal Care and Use Committee (protocol #8787). Cat housing and husbandry was overseen by MU Office of Animal Research. No cats were euthanized as part of this study. However, cats were euthanized by barbiturate overdose after sedation as dictated by poor health (renal failure). All cats were genotyped for the
PKD1 c.10063C > A that causes feline ADPKD (data not shown) [
17,
27]. Diagnostic imaging studies were performed at MU during normal clinic hours (8:00 am – 6:00 pm) from November 2016 to May 2017, for the initial imaging, and June 2018 for the follow-up CT and MRI imaging of two cats. Cats were fasted for at least 12 h prior to diagnostic imaging. At the time of sedation for imaging, blood was collected by jugular venipuncture for a complete blood count, serum chemistry including BUN, sCr, SDMA. Urine was collected for urinalysis via cystocentesis using ultrasound guidance.
Ultrasonography
US examinations were performed using an 8 MHz micro-convex transducer on a dedicated ultrasound unit (Logiq 9, GE Healthcare, Wauwatosa, WI, USA). Each kidney was scanned in sagittal and transverse planes by either a first-year veterinary radiology resident (K.L.S.) or a board-certified veterinary radiologist (J.S.M.). The length was measured, using internal digital calipers, as the longest point between the cranial and caudal poles in the sagittal plane. Width and height were measured in the transverse plane. TKV was calculated using the prolate ellipsoid formula [
28].
Glomerular filtration rate
GFRs were determined on the same day as the US imaging. Approximately 3 mCi of
99mTc-diethylenetriaminepentaacetic acid was administered intravenously. Images were obtained using a gamma camera (Equistand II, Diagnostic Services, Middlesex, NJ.) with a low energy all-purpose collimator and using a 256 × 256 matrix. The GFR was calculated for each kidney by a single observer (K.L.S.), using Mirage software and a previously described method for scintigraphic uptake [
29].
Magnetic resonance imaging
MRI was performed using a 3 T unit (Vantage Titan 3 T, Canon Medical Systems, Tustin, CA, USA) and a transmit/receive coil. Scans obtained included dorsal T2 (TR 2469–5675, TE 120) weighted imaging (T2WI) with 2 mm slices, a 0.2 mm interspace gap, matrix = 320 × 320–352, and number of acquisitions = 1–3. The scans were performed using respiratory gating to eliminate respiratory motion.
Computed tomography
CT scans were performed using a third generation 64 slice instrument (Aquillion 64, Canon Medical Systems, Tustin, CA, USA) under the same anesthetic event as the MRI scan. Follow-up imaging was performed for two cats at 12 months and 15 months, which were thought to have fast and slow cyst progression (i.e., high and low FCV at the youngest age), respectively.
Calculation of imaging parameters from CT and MRI data
TKVs were calculated from US as described above. TKVs as determined by CT and MRI was measured using a planimetry method with Fiji software common to veterinary practice [
30]. To determine the area per slice, the outline of the kidney was traced by free-hand by a single investigator (K.L.S.) on each contiguous slice. Total volumes of the kidneys were obtained by summing the areas of the slices and multiplying by slice thickness. TKV and TCV were also calculated by a single analyst (M.E.E.) using the minimal interaction rapid organ segmentation (MIROS) method [
31]. Cyst progression rate was predicted from the FCV per months (age) and from follow-up CT and MRI for two cats at 12 months and 15 months as described above.
Statistical analysis
Means and standard deviations were calculated for 11 ADPKD cats for variables imaging indexes. All statistical analyses were performed using R software (version 3.3.3; R Foundation for Statistical Computing, Vienna, Austria). Comparisons of kidney volumes between male and female cats were analyzed using Mann–Whitney U test. The Spearman’s rank correlation test was used for the correlation analyses of imaging parameters and clinical data such as age, body weight and serum biomarkers. Kendall’s coefficient of concordance was calculated to evaluate the reliability between or among each modality. High agreement is indicated when Kendall’s coefficient of concordance (W) is higher than 0.75. P values < 0.05 were considered statistically significant. Data are expressed as mean ± standard deviation (SD).
Discussion
ADPKD is caused by
PKD1 variants and is a common genetic and life-threatening disease for both humans and domestic cats. Although a variety of rodent models support PKD research, none support long-term trials of therapeutics. ADPKD domestic cats have a stop codon in exon 46, disrupting ~ 30% of polycystin-1 [
17]. The disease is autosomal dominant, the homozygous state is lethal in utero, and cysts develop prior 8 months of age. Thus, this cat models mimics the human condition genetically. However, little is known about the changes in kidney imaging parameters in ADPKD cats, especially TKV, TCV, and FCV.
The rate and extent of cystic progression has not been examined in cats, although generally, cysts worsen as the cat ages. The average life span of cats is ~ 13–17 years of age and many die of CKD [
33,
34]. One of the factors leading to the discovery of ADPKD was early CKD as cats were dying at 3–4 years old. However, many cats with ADPKD live a normal life span. Thus, the variation in disease severity and rate of progression is recognized but undocumented. Imaging is required to determine disease severity and to monitor progression even though a genetic test exists for feline ADPKD [
17,
27]. The range in severity suggests additional genetic and non-genetic factors influence disease progression.
Eleven ADPKD cats were examined by different modalities to quantify the variation in the cats and to compare imaging modalities. Ten of eleven cats had no suggestion of renal compromise other than the presence of cysts, thus renal biomarkers, including SDMA, sCr and BUN, are not predictive risk indicators of feline ADPKD and disease severity. GFR was also not indicative of kidney disease in the ADPKD cats. Significant parenchymal loss is likely to be required before abnormal GFRs are observed in APDKD cats [
32]. The GFRs were within normal limits or low for the ADPKD cats, including the oldest cat and five ADPKD cats with ~ 15% FCV. Three cats with lower FCV had mildly lowered GFRs, which may be partially due to reduced renal artery pressure secondary to the anesthesia. The age of the cat cohort is relatively young (mean: 39.7 months [range: 16.4–97.1 months]. Many human ADPKD patients have no obvious clinical symptoms until the third or fourth decade of life [
35] and renal function usually remains normal until the fourth to sixth decade of life [
36]. A recent study of 377 cats from Japan indicated of cats with the cat
PKD1 mutation, the incidence of a high concentration of plasma Cre (>1.6 mg/dl: ≥IRIS-CKD stage 2) was greater in cats older than 3 years old, and especially in those older than 7 years. In contrast, a few cats aged ≥9 years had low plasma Cre concentrations (≤1.6 mg/dl) [
37]. Therefore, the correlation between TKV and GFR may improve in older ADPKD cats.
TKV, rather than renal function, is suggested as the more appropriate biomarker for monitoring and predicting disease progression in human medicine [
38]. In humans, TKV measurements are obtained by MRI or CT to assess the efficacy of therapeutic interventions [
23,
39,
40]. US, CT and MRI are all minimally invasive and highly diagnostic for ADPKD in cats, with US being the most rapid, least expensive, and most accessible. MRI T2WI is sensitive and sufficient for volume measurement and CT is associated with radiation exposure, hence CT is less favored in human medicine [
2,
40]. However, CT is favored in cats due to decreased anesthesia requirements and the minimal concern of long-term consequences of radiation exposure.
US, CT and MRI were conducted in ADPKD cats to estimate TKV. Normal cat kidney volumes have been estimated using water displacement at 18.99 ± 7.68 cm
3, and US and CT imaging from 14.8 ± 2.9 ml to 19.01 ± 7.55 ml, respectively [
32,
41]. For ADPKD cats, total kidney volumes have been estimated as 27.4 ± 10.3 ml using US and 29.3 ± 13.4 ml by CT [
32], ~ 30% larger than normal cat kidneys. In this study, the 11 ADPKD cats had kidney volume estimates from 23.0 ± 6.91 ml using US to as high as 34.69 ± 6.36 ml using the planimetric method for MRI thereby overlapping within normal limits but also have significantly larger TKVs. Overall, US and planimetry CT TKV estimates were slightly smaller, however, within the ranges of previous study. The cats in the previous study (mean age: 59 ± 10 months) [
32] were an average of 20 months older than this study (mean age: 39.67 ± 23.89 months), thus would be expected to have larger TKVs. The TKV estimates for a given cat were increasing larger over US-based TKV estimates by 20, 22, 27, 34%, estimating by planimetry CT, MIROS CT, MIROS MRI, and planimetry MRI, respectively. The TKV for case 4 that was estimated by water displacement was consistent with CT estimates, but case 5 estimates were over-estimated by CT and MRI, potentially due to the adipose tissue in this overweight cat.
In one study, cyst growth in humans is reported as relatively symmetrically and at a steady rate [
42]. However, other research has indicated variable expression of cyst progression, even in the same family [
43,
44]. In the cats, FCVs were highly correlated with age, supporting cyst size increasing with age and a steady rate. However, Case 3 showed 1.08 and 1.84% increase in FCV per month, by CT and MRI, respectively, which is approximately six times the average increase than the other ADPKD cats. Once this “outlier” was removed from the correlation analysis, FCV per month showed an average of 0.19 and 0.29% by CT and MRI, respectively, and was more highly correlated with age. Thus, for cats, most cats have a consistent rate of disease progression, however, some cats are highly variability. Furthermore, the value on TCV and FCV showed wide ranges as shown in Table
3. Although ages were not uniform, these findings also indicated that variability of cyst progression speed among feline ADPKD cats, regardless of genetic homogeneity for the mutation. Feline ADPKD is individually variable between kidneys of a given cat, even though all cats had the same germline mutation (
PKD1 c.10063C > A), suggesting other factors modify disease expression and progression.
Cats as a large animal model are clearly an asset to evaluate the efficacy and safety of the development of drugs and gene therapies [
45,
46]. Although the exact same mutation is not found in cats and humans, cats and humans have similarly disruptive mutations that truncate approximately 30% of polycystin-1. Feline ADPKD may fill a void of translational research between rodent and human ADPKD. The rodent models do not perfectly recapitulate human ADPKD in terms of differences in lifespan, metabolism, and renal anatomy [
47]. Feline ADPKD has the potential to overcome these differences. Although feline ADPKD is caused by a single
PKD1 c.10063C > A, considering progression variability, disease progression of feline ADPKD is likely influenced by other genetic and/or environmental factors, such as those identified in humans [
48‐
51]. In addition, cats could support studies focusing on pleiotrophic effects of ADPKD. A recent study showed co-occurrence of hepatic and renal cysts was found in 20 (12.6%) out of 159 cat cases with renal cyst(s), and all cases were positive for the
PKD1 mutation [
37]. Identification of genetic modifiers could lead to selection of appropriate cats with ADPKD for therapeutic trials and reducing animal use and improving study design.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.