Alström syndrome: Renal findings in correlation with obesity, insulin resistance, dyslipidemia and cardiomyopathy in 38 patients prospectively evaluated at the NIH clinical center

Abstract

Alström Syndrome is a ciliopathy associated with obesity, insulin resistance/type 2 diabetes mellitus, cardiomyopathy, retinal degeneration, hearing loss, progressive liver and kidney disease, and normal cognitive function. ALMS1, the protein defective in this disorder, localizes to the cytoskeleton, microtubule organizing center, as well as the centrosomes and ciliary basal bodies and plays roles in formation and maintenance of cilia, cell cycle regulation, and endosomal trafficking. Kidney disease in this disorder has not been well characterized. We performed comprehensive multisystem evaluations on 38 patients. Kidney function decreased progressively; eGFR varied inversely with age (p = 0.002). Eighteen percent met the definition for chronic kidney disease (eGFR < 60 mL/min/1.73 m² and proteinuria); all were adults with median age of 32.8 (20.6–37.9) years. After adjusting for age, there were no significant associations of kidney dysfunction with type 2 diabetes mellitus, dyslipidemia, hypertension, cardiomyopathy or portal hypertension suggesting that kidney disease in AS is a primary manifestation of the syndrome due to lack of ALMS1 protein. Approximately one-third of patients had hyperechogenicity of the renal parenchyma on imaging. While strict control of type 2 diabetes mellitus may decrease kidney-related morbidity and mortality in Alström syndrome, identification of novel targeted therapies is needed.

1. Introduction

Alström Syndrome (AS) (ALMS; OMIM#203800) is a ciliopathy associated with obesity, severe insulin resistance/type 2 diabetes mellitus (T2DM), hypertriglyceridemia, retinal degeneration, hearing loss, cardiomyopathy, progressive liver and kidney disease, and normal cognitive function. AS is extremely rare, with an estimated prevalence of 1 to 10 per 1,000,000 persons [1, 2]. AS is caused by bi-allelic mutations in the ALMS1 gene [3–5], which encodes multiple isoforms of ALMS1, ranging in size from 3859 to 4169 amino acids and ubiquitously expressed in fetal and postnatal tissues [6, 7]. At the subcellular level, ALMS1 localizes to the cytoplasm, cytoskeleton, microtubule organizing center, centrosomes and ciliary basal bodies [3, 4]. Although its function has not been fully elucidated, ALMS1 is known to play roles in the formation and maintenance of cilia, cell cycle regulation, endosomal trafficking, and extracellular matrix production [8–13].

Progressive retinal cone-rod dystrophy is associated with AS, begins in infancy, and leads to blindness in childhood. Sensorineural hearing loss is often diagnosed in childhood, although onset is variable [14]. Obesity and insulin resistance begin to develop during the first year; most patients ultimately develop T2DM [1, 15]. Marked hypertriglyceridemia is typical and results in pancreatitis in some cases. Hepatic steatosis is almost universal; portal hypertension may occur [16]. Some AS patients present in infancy with severe dilated cardiomyopathy characterized by mitogenic histopathology with cardiomyocyte hyperplasia and proliferation [11]; many succumb to cardiac failure before the diagnosis of AS is made. In those who survive, myocardial dysfunction resolves to low-normal within a few years. Others develop restrictive cardiomyopathy later in childhood to adulthood [2, 17].

Chronic progressive nephropathy with eventual kidney failure is a common complication of AS [1, 18, 19]. Due to the extreme rarity of AS, the literature regarding renal disease is largely based on retrospective reviews and small case series [18–20].

Here we present comprehensive data on renal parameters in an AS cohort of 38 patients prospectively evaluated at a single center. We describe characteristics of renal function and renal imaging in detail and present these findings in the context of other organ system disease including hepatic, endocrine, cardiac, ophthalmologic and auditory involvement as well as molecular genetic findings.

2. Methods

2.1. Patients

All 38 patients with AS were evaluated at the National Institutes of Health (NIH) Clinical Center between February 2013 and June 2014, under the NIH protocol “Clinical and Molecular Investigations into Ciliopathies” (www.clinicaltrials.gov, trial NCT00068224), approved by the Institutional Review Board of the National Human Genome Research Institute (NHGRI). All patients fulfilled the clinical diagnostic criteria for AS. In addition, mutations in ALMS1 were identified in all 38 patients. Patients or their parents gave written informed consent. Evaluations performed at The NIH Clinical Center included review of past medical records, detailed past medical and family history, physical examinations, comprehensive biochemical blood and urine testing, abdominal ultrasonography (USG), abdominal magnetic resonance imaging (MRI), audiology and ophthalmology evaluations, bone density scans, echocardiography and cardiac MRI.

2.2. Molecular genetic testing

All patients with AS had prior genetic testing using Sanger sequencing, microarray-based APEX (arrayed primer extension), or Targeted Gene Sequencing and Custom Analysis (TaGSCAN) test as previously described [5].

2.3. Laboratory assessments

Serum creatinine, cystatin C, glucose, insulin, hemoglobin A1c (HbA1c), triglycerides, high-density lipoprotein cholesterol (HDL), and low-density lipoprotein cholesterol (LDL), uric acid, parathyoroid hormone (PTH), calcium, phosphorus, 25-hydroxyvitamin D (25OHD), and 1,25-hydroxyvitamin D (1,25OHD) concentrations and osmolality were measured by the NIH Department of Laboratory Medicine. Urinary studies included protein to creatinine (UPCR) (mg/g), albumin to creatinine (UACR) (mg/g) and calcium/creatinine (Ca:Cre) (mg/mg) ratios, osmolality, and 24-h urine collections for glucose, electrolytes, protein, albumin and creatinine excretion. Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated by the following equation: [(Fasting insulin (μIU/mL) × fasting glucose (mg/dL)/405)] [21]. Insulin resistance was defined as HOMA-IR > 2.8 in adults [22] and > 90th centile of sex, age and body mass index (BMI)-matched controls in children [23]. Estimated glomerular filtration rate (eGFR) was calculated using pediatric [24] and adult [25] formulas. Serum cystatin C was not available in two children due to problems in handling of blood specimens. For these patients, eGFR was calculated using an alternative formula based on height and serum creatinine [26].

Proteinuria was defined as UPCR > 500 (mg/g) in infants and toddlers, 6 to 24 months, > 200 (mg/g) in children > 24 months of age and adults or 24-h urinary excretion of > 300 mg/day. Albuminuria was defined as UACR > 30 (mg/g). Hypercalciuria was defined as urinary Ca:Cr ratio of > 0.21 in a spot urine collection or 24-h urine calcium excretion > 4 mg/kg or 200 mg/day. Chronic kidney disease was defined as GFR < 60 mL/min/1.73 m² with proteinuria. Prehypertension was defined as systolic or diastolic blood pressure ≥90th percentile for age/sex/height or > 120/80 mmHg. Stage 1 hypertension was defined as ≥95th percentile or ≥130/80 mmHg. Stage 2 hypertension was defined as ≥5 mmHg above 99th percentile or ≥140/90. Urinary concentrating ability was assessed by calculating the ratio of urine osmolality to serum osmolality; normal values are 1.0 to 3.0.

2.4. Imaging studies

Complete abdominal ultrasonography (USG) evaluations were performed by a single technologist using standard (4 MHz) and high-resolution (7 MHz) USG probes on all patients (AVI Sequoia Inc., Mountain View, CA). Kidney length measured by USG was converted to standard deviation based on normative data [27]. MRI was performed on a 1.5- or 3-Tesla machine (Philips Medical Systems, NA, Bothell, WA; General Electric Healthcare, Waukesha, WI) without intravenous contrast media.

2.5. Cardiac evaluations

Standard 12-lead electrocardiograms were recorded using commercially available systems with measurement of PR interval, QRS duration and QT. Transthoracic echocardiograms were performed using commercially available systems; standard views were obtained with patients in the left lateral recumbent position.

2.6. Statistical methods

Data were summarized using counts and percentages or medians and ranges. Non-parametric methods were used for comparisons because most of the continuous measures were not normally distributed. Group comparisons of categorical data were made using Fisher’s exact tests, and for continuous measures the Wilcoxon rank-sum test or the Kruskal-Wallis test as appropriate. Spearman rank correlations were used to test the association of continuous measures including BMI percentile, Z scores of systolic and diastolic blood pressures, low (LDL) and high density lipoprotein (HDL), cholesterol, triglycerides, HOMA-IR, HbA1C and duration of T2DM. General linear model regression was performed to determine if, after adjusting for age, measures of kidney involvement differed by percentile of BMI, Z scores of systolic and diastolic blood pressures, duration of T2DM, LDL and HDL cholesterol, triglycerides, HOMA-IR, and by presence of T2DM, cardiomyopathy and portal hypertension. Analyses were performed using SAS version 9.4 (SAS Institute, Inc., Cary, NC). All tests were two-sided and significance was set at p < 0.05.

3. Results

3.1. Patient characteristics

The patients included 18 males and 20 females, with a median (range) age of 12.4 (1.7–37.9) years (Table 1). One family contributed 3 siblings, and 4 families contributed two siblings each including 2 sets of dizygotic twins, leaving 32 independent families. Twelve (32%) patients were younger children (≤10 y), 13 (34%) were older children and adolescents (11–17 y) and 13 (34%) were adults (≥18 y). Thirty-four (89%) patients were Caucasian, 3 (8%) were African-American and 1 (3%) was Asian.

Table 1.

Summary of Clinical and Laboratory Features of Kidney Disease and Measures of Other Multiorgan System Involvement in 38 Patients with Alström syndrome evaluated at the NIH Clinical Center.

Patient No	Family No	Age (y)	Sex	Race	eGFR (mL/ min/1.73sm)		Serum Creatinine	Serum Cystatin C	Urine Qualitative Protein	Urine protein excretion (mg/24 h)	Urine Protein/Creatinine Ratio (mg/g)	Urine Albumin/Creatinine Ratio (mg/g)
1	1	1.7	M	W	85		0.27	0.91	Neg	NA	419	33.3
2	1	5.0	M	W	111		0.3	0.81	Neg	75	349	6.7
3	1	8.6	M	w	107		0.41	0.77	Trace	226	377	150
4	2	3.0	F	w	95		0.3	0.66	NA	NA	NA	NA
5	3	3.8	F	w	124		0.21	0.57	Neg	NA	351	8.4
6	4	4.4	F	w	103		0.32	0.73	Neg	NA	197	9.5
7	5	5.1	F	w	106		0.37	0.66	Neg	95	188	18
8	6	5.2	M	w	117		0.29	0.74	Neg	NA	250	19
9	7	5.7	M	w	127		0.37	NA	Neg	74	136	4
10	7	13.0	M	w	101		0.61	0.8	Neg	202	207	18
11	8	8.3	F	w	61		0.79	1.15	Trace	290	566	113
12	9	9.0	M	w	102		0.53	0.84	Neg	232	400	140
13	10	10.1	F	w	100		0.48	0.72	Neg	135	268	15
14	11	11.3	M	w	107		0.6	0.65	Neg	84	216	12
15	12	11.5	F	AA	88		0.55	0.99	2 +	893	1406	159
16	13	11.5	F	W	79		0.68	0.9	Neg	181	153	15
17	14	11.7	F	As	95		0.38	1.07	Neg	372	830	184
18	15	12.4	F	AA	114		0.54	0.59	Neg	129	107	4
19	15	12.4	M	AA	96		0.62	0.78	Neg	NA	NA	NA
20	16	12.4	M	W	109		0.56	NA	Trace	255	58	2
21	16	12.4	M	W	121		0.42	0.71	Trace	112	193	4
22	17	13.9	F	W	88		0.72	0.91	NA	371	296	40
23	18	15.7	F	W	76		0.63	1.21	Neg	178	207	41
24	19	17.0	F	W	74		0.79	1.01	Neg	831	213	7
25	20	17.6	M	W	93		0.71	0.76	NA	NA	133	3
26	21	18.9	M	W	68		1.18	1.46	Neg	781	137	59
27	22	19.2	F	W	143		0.41	0.69	Neg	254	415	95
28	23	20.6	F	W	37		2.13	1.58	Neg	NA	214	83
29	24	20.9	M	W	116		0.93	0.83	2 +	518	173	57
30	25	24.0	F	W	44		1.4	1.73	Neg	549	231	7
31	26	24.1	F	W	95		0.83	0.92	Neg	NA	80	3
32	27	27.0	F	W	49		1.42	1.43	Neg	287	391	74
33	28	32.8	M	W	55		1.62	1.35	1 +	1073	784	470
34	29	34.2	M	W	84		1.02	1.1	2 +	2801	1935	1250
35	30	35.0	F	W	110		0.67	0.81	Trace	196	115	42
36	30	36.6	M	W	46		1.91	1.49	1 +	679	576	300
37	31	35.7	F	W	53		1.15	1.44	1 +	995	1256	365
38	32	37.9	M	W	25		2.89	2.46	Trace	758	287	96
Patient No	Serum albumin	Urine osmolality	Serum Osmolality	U Osm/S Osm Ratio	R Kidney Length USG (Z)	L Kidney Length USG (Z)	Kidney USG	Kidney MRI	Systolic BP (Z)	Diastolic BP (Z)	Anti-Hypertensive Medications	BMI (%)
1	4.2	1036	298	3.5	−2.15	−1.25	Normal	NA	2.2	2.0	N	97.86
2	4.1	943	293	3.2	0.99	2.2	Normal	NA	0.8	0.4	N	99.99
3	4.2	869	294	3.0	1.58	2.11	Normal	NA	1.5	−0.1	Y	98.23
4	4.1	NA	NA	NA	−0.12	0.87	Normal	NA	−0.6	−1.3	N	99.97
5	4.2	1156	296	3.9	−0.87	−0.51	Normal	NA	NA	NA	N	99.99
6	4.3	858	294	2.9	−0.19	1.19	Normal	NA	0.4	0.5	Y	99.32
7	4.3	NA	NA	NA	2.16	1.3	Normal	NA	0.4	−0.2	N	99.86
8	4.5	525	302	1.7	0.25	1.29	Normal	NA	NA	NA	N	99.99
9	4.0	450	NA	NA	0.92	2.68	Normal	NA	1.6	0.8	N	99.94
10	4.2	764	NA	NA	2.64	2.08	Normal	Normal	0.4	−0.6	Y	96.36
11	4.4	619	297	2.1	2.68	3.49	Medullary hyperechogenicity	Medullary hyperintensities, cysts	2.6	0.6	N	99.63
12	4.1	846	NA	NA	4.4	3.74	Medullary hyperechogenicity	NA	1.0	−0.2	N	97.31
13	3.6	932	294	3.2	1.7	0.45	Medullary hyperechogenicity	Medullary hyperintensi ties	0.9	0.0	N	99.22
14	3.6	885	294	3.0	1.93	1.18	Normal	Normal	0.8	0.5	N	98.73
15	3.8	564	299	1.9	1.33	2.6	Normal	Normal	1.7	0.6	Y	99.16
16	4.5	755	296	2.6	−0.06	0.32	Medullary hyperechogenicity	Medullary hyperintensi ties	0.9	−0.3	N	96.81
17	4.2	789	301	2.6	1.41	0.39	Normal	Normal	1.5	0.8	N	81.22
18	3.4	583	291	2.0	1.9	0.96	Normal	Normal	1.6	0.8	N	91.61
19	4.0	NA	NA	NA	0.01	−0.25	Corticomedullary hyperechogenicity	Normal	2.0	0.9	N	76.80
20	4.4	948	NA	NA	−0.54	0.54	Normal	Normal	0.8	0.4	N	70.11
21	4.3	869	287	3.0	0	0	Normal	Normal	1.4	0.4	N	65.28
22	4.2	456	294	1.6	4.92	3.16	Medullary hyperechogenicity	Medullary hyperintensities	2.3	0.8	N	99.28
23	3.5	678	298	2.3	1.83	2.54	Normal	NA	1.7	0.4	N	96.64
24	3.2	455	296	1.5	1	1.69	Normal	Cysts	0.8	−0.5	Y	93.91
25	4.0	583	291	2.0	−0.41	−0.3	Corticomedullary hyperechogenicity	Normal	−0.7	−0.5	N	57.15
26	3.9	504	305	1.7	2.42	−0.85	Medullary hyperechogenicity	Normal	−0.4	−0.6	Y	99.19
27	4.0	467	294	1.6	1.97	0.25	Normal	Normal	0.9	0.8	N	94.01
28	3.6	311	309	1.0	−0.75	−0.75	Normal	Normal	−0.6	−0.6	Y	93.22
29	4.1	491	301	1.6	1	0.87	Normal	Medullary hyperintensities	5.0	3.1	Y	85.44
30	3.5	330	312	1.1	0.5	1.75	Normal	Normal	−0.6	−0.8	Y	93.97
31	3.7	659	300	2.2	−0.75	−0.75	Normal	Medullary hyperintensities	1.2	−0.4	N	98.89
32	3.0	362	NA	NA	NA	NA	NA	Normal	1.0	−0.5	Y	99.52
33	3.8	424	308	1.4	NA	−2.3	Normal	Normal	1.2	1.4	Y	99.24
34	4.1	305	303	1.0	1.2	1.2	Normal	Cysts	0.3	0.1	Y	91.94
35	3.6	533	295	1.8	−1.7	−1.7	Corticomedullary hyperechogenicity	Medullary hyperintensi ties	2.3	0.3	N	93.39
36	4.0	304	304	1.0	−2.2	−1.75	Corticomedullary hyperechogenicity	Medullary hyperintensities	−0.2	0.5	Y	94.29
37	3.5	276	304	0.9	2	−1.7	Medullary hyperechogenicity	Medullary hyperintensities, cysts	1.4	0.3	Y	95.88
38	3.9	457	313	1.5	−0.25	−0.75	Normal	NA	1.0	0.2	Y	99.98
Patient No	T2DM	Duration of T2DM (y)	HbA1c (%)	HOMA-insulin Resistance	LDL Cholesterol (mg/dL)	HDL Cholesterol (mg/dL)	Trigylcerides (mg/dL)	Portal Hypertension	Infantile-onset Cardiomyopa thy	Late-onset Cardiomyopa thy	Visual Acuity	Hearing Loss
1	N	NA	4.8	0.49	82	37	58	N	N	N	No LP	Und
2	N	NA	4.8	4.6*	115	29	143	N	N	N	10/600	Slight
3	N	NA	4.7	2.30	NA	NA	408	Y	Y	N	CUSM	Mod
4	N	NA	5.2	NA	75	35	87	N	N	N	CUSM	Und
5	N	NA	4.9	1.24	124	54	73	N	Y	N	20/300	Und
6	N	NA	5	2.35*	55	43	43	N	N	N	10/200	Und
7	N	NA	5.3	4.07*	82	43	169	NA	Y	N	20/400	Normal
8	N	NA	5.2	1.07	81	74	89	N	Y	N	20/500	Und
9	N	NA		NA	66	30	113	NA	N	N	20/200	Mod
10	N	NA	5.7	NA	NA	NA	501	N	N	N	20/320	Mild
11	N	NA	4.9	2.82	73	27	106	Y	Y	N	20/800	Mod
12	N	NA	5.8	7.25*	54	54	83	N	N	N	20/250	Mild
13	N	NA	5.9	NA	143	36	121	N	Y	N	20/400	Normal
14	N	NA	5.8	NA	94	31	114	N	N	N	LP	Mod
15	Y	5	12	33.82*	NA	NA	3292	Y	N	N	LP	Mod
16	N	NA	5.7	NA	62	32	203	N	N	Y	LP	Mod
17	Y	0.8	5	35.71*	66	29	206	N	N	N	20/800	Mod
18	N	NA	5.1	10.48*	84	33	69	N	Y	N	20/400	Mild
19	N	NA	5.3	2.28	97	28	70	N	Y	N	NA	NA
20	N	NA	5.3	1.45	55	37	239	N	Y	N	20/320	Normal
21	N	NA	5.2	3.45*	60	41	107	N	Y	N	20/250	Slight
22	Y	0.6	6.2	14.44*	131	30	296	N	N	N	20/160	Mod
23	Y	8.7	5.3	19.89*	107	38	155	Y	N	N	LP	Mod
24	Y	11	7.7	49.85*	163	11	332	Y	N	Y	No LP	Mod
25	N	NA	5.3	0.84	71	46	43	N	Y	N	LP	Mild
26	Y	10.9	8	34.10*	NA	NA	593	Y	N	Y	LP	Mod
27	N	NA	5.9	17.71*	NA	NA	508	N	N	N	LP	Mod
28	Y	7.6		19.11*	NA	NA	481	N	Y	N	No LP	Mod
29	Y	12.9	11.8	NA	167	29	275	Y	N	Y	No LP	Mod
30	Y	13.5	10.2	110.82*	72	18	355	Y	N	Y	LP	Mod
31	N	NA	5.7	8.34*	91	33	202	Y	N	N	LP	Mod
32	Y	14	5.5	12.74*	38	36	81	Y	N	N	No LP	Mod
33	N	NA	5.2	NA	74	26	159	N	Y	N	No LP	Severe
34	Y	17.2	8.2	NA	NA	NA	559	Y	N	N	No LP	Severe
35	Y	23	5.7	12.61*	81	50	179	N	N	N	No LP	Mod
36	Y	20.6	6.4	17.82*	69	53	140	N	N	N	No LP	Mod
37	Y	17.7	6.6	5.64*	NA	NA	497	Y	N	N	No LP	Mod
38	N	NA	5.1	13.73*	97	33	125	N	N	N	No LP	Severe

y:years; eGFR: estimated glomerular filtration rate; h:hour; USG:ultrasonography; MRI: magnetic resonance imaging; BP:blood pressure; BMI: body mass index; T2DM:type 2 diabetes mellitus; HOMA-insulin resistance: homeostasis model assessment of insulin resistance; LP;light perception; CUSM: central, unsteady and maintained; NA:not available.

^*Indicates increased HOMA-insulin Resistance values; N:no, Y:yes; Mod:moderate; Und:undetermined.

In all patients, vision loss preceded hearing loss. The median age of detection of hearing loss was 8.0 (1.5–15) years [14], and median age of AS diagnosis was 7.7 (prenatal-27) years. Severe metabolic abnormalities were evident in early childhood. Among the 25 children, weight was normal in only 5 (20%); two (8%) were overweight (BMI ≥85th percentile) and the majority (72%) were obese (BMI ≥ 95th percentile) (Table 1). Among the 13 adult patients, two (15%) were overweight (BMI ≥25 kg/m²) and 11 (85%) were obese (BMI ≥ 30 kg/m²). Insulin resistance was identified in 58% of children and 100% of adults (Table 1). T2DM was diagnosed in 37% of the cohort including 20% of children and 69% of adults. (Table 1). The median HbA1C of patients with T2DM, 86% of whom were receiving glucose lowering medications, was 6.6% (5.0–12.0%). Blood pressure was in the prehypertension range in five children (20%) and 6 adults (46%), and in the stage 1 hypertension range in 6 children (24%) and 1 adult (8%). Ten children (40%) and 10 adults (77%) had hypertriglyceridemia (≥150 mg/dL). Liver echogenicity based on ultrasound was normal in 21%, mildly increased in 26% and moderately to severely increased in 50% and 3% of patients, respectively. Approximately half of patients (17/38) had cardiomyopathy: in the majority (13 patients, 76%), the onset was in infancy and 4 patients were diagnosed during adolescence or early adulthood [17]. The 13 patients with infantile onset cardiomyopathy responded to heart failure treatment with improvement of ejection fractions to the lower end of normal within the first 1–2 years with no recurrence of cardiomyopathy. At the time of their evaluation at the NIH Clinical Center, all 38 patients were clinically stable. Four patients were on beta blockers, 14 were on angiotensin converting enzyme inhibitors, two were taking angiotensin receptor blockers, and three were on other anti-hypertensive agents.

3.2. Molecular genetic findings

We identified bi-allelic mutations in ALMS1 in 37 (97%) patients; in one patient a single mutation was identified (Table 2). All mutations were nonsense or frameshifting resulting in protein truncation.

Table 2.

ALMS1 mutations identified in 38 patients with Alström syndrome evaluated at the NIH Clinical Center.

Patient No	Family No	ALMS1 Mutation 1 c.DNA	ALMS1 Mutation 1 Protein	ALMS1 Mutation 2 c.DNA	ALMS1 Mutation 2 Protein
1	1	c.11316_11319delAGAG	p.Glu3773Trpfs*18	c.11416C > T	p.Arg3806*
2	1	c.11316_11319delAGAG	p.Glu3773Trpfs*18	c.11416C > T	p.Arg3806*
3	1	c.11316_11319delAGAG	p.Glu3773Trpfs*18	c.11416C > T	p.Arg3806*
4	2	c.10483C > T	p.Gln3495*	c.10483C > T	p.Gln3495*
5	3	c.10535G > A	p.Trp3512*	c.11291G > A	p.Ser 3764*
6	4	c.11316_11319delAGAG	p.Glu3773Trpfs*18	c.7771_7772insT	p.Thr2592Asnfs*3
7	5	c.4156dupA	p.Thr1386Asnfs*15	Not identified	Not identified
8	6	c.10775delC	p.Thr3592Lysfs*6	c.2234C > G	p.Ser 745*
9	7	c.10265delC	p.Pro3422Glnfs*2	c.2930_2933dupGAGA	p.Ser979fs
10	7	c.10265delC	p.Pro3422Glnfs*2	c.2930_2933dupGAGA	p.Ser979fs
11	8	c.10775delC	p.Thr3592Lysfs*6	c.10775delC	p.Thr3592Lysfs*6
12	9	c.4156dupA	p.Thr1386Asnfs*15	c.4156dupA	p. Thr1386Asnfs*15
13	10	c.5145T > G	p.Tyr1715*	c.3754C > T	p.Gln1252*
14	11	c.8352_8355delAGAA	p.Glu2785*	c.6436C > T	p.Arg2146*
15	12	c.9328C > T	p.Gln3110*	c.10549C > T	p.Gln3517*
16	13	c.10539_10557ins(n)19	p.Lys3545Asnfs*18	c.10539_10557ins(n)19	p.Lys3545Asnfs*18
17	14	c.6436C > T	p.Arg2146*	c.6436C > T	p.Arg2146*
18	15	c.4885C > T	p.Gln1629*	c. 5923C > T	p.Gln 1975*
19	15	c.4885C > T	p.Gln1629*	c. 5923C > T	p.Gln 1975*
20	16	c.592C > T	p.Gln198*	c.1610_1611delTC	p.Leu538Glnfs22
21	16	c.592C > T	p.Gln198*	c.1610_1611delTC	p.Leu538Glnfs22
22	17	c.11651_11652insGTTA	p.Asn3885Leufs*9	c.9900dupC	p.Ser3301Leufs*7
23	18	c.10539_10557ins(n)19	p.His3512fs	c.11416C > T	p.Arg3806*
24	19	c.6305C > A	p.Ser2102*	c.10775delC	p.Thr3592Lysfs*6
25	20	c.10849G > T	p.Glu3617*	c.10483C > T	p.Gln3495*
26	21	c.10775delC	p.Thr3592Lysfs*6	c.3716_3719del	p.Ser1240Thrfs*23
27	22	c.4180C > T	p.Gln1394*	c.4180C > T	p.Gln1394*
28	23	c.11314dupA	p.Arg3772Trpfs*10	c.10885C > T	p.Arg3629*
29	24	c.5311C > T	p.Gln1769*	c.5311C > T	p.Gln1769*
30	25	c.11651_11652insGTTA	p.Asn3885LeufsX9	c.4817delA	p.Lys1608ArgfsX9
31	26	c.11313_11316delTAGA	p.Asp3771Glufs*20	c.2329C > T	p.Gln777*
32	27	c.8394_8395insA	p.Leu2799Ilefs*4	c.9194T > G	p.Leu3065*
33	28	c.1903C > T	p.Gln635*	c.3579C > G	p.Tyr1193*
34	29	c.10849G > T	p.Glu3617*	c.3019dupA	p.Arg1007Lysfs*15
35	30	c.5283delA	p.His1762Ifs*24	c.10483C > T	p.Gln3495*
36	30	c.5283delA	p.His1762Ifs*24	c.10483C > T	p.Gln3495*
37	31	c.4039C > T	p.Gln1347*	c.5145T > G	p.Tyr1715*
38	32	c.7374_7375delAG	p.Asp2459*	c.7374_7375delAG	p.Asp2459*

^*indicates stop codon.

3.3. Renal findings

Table 1 displays the kidney-related serum and urine chemistries, kidney USG and MRI findings as well as measures of other organ system involvement including BMI, systolic and diastolic blood pressures, HbA1c, HOMA-IR, LDL and HDL cholesterol, triglycerides, T2DM, cardiomyopathy, portal hypertension, retinal degeneration and hearing loss.

Renal function decreased progressively with increased patient age; there was an inverse correlation between eGFR and age (p = 0.002) (Supplementary Table 1, Fig. 1A). More than half (21/37 = 57%) of the patients had proteinuria that was low grade in most (median 255, range 74–2801 mg/day) (Table 1); only two patients (Table 1, patients # 33 and 34) excreted > 1 g/day protein (1.1 and 2.8 g/day). Fifteen patients (42%) were receiving either an angiotensin converting enzyme inhibitor or an angiotensin II receptor antagonist, which can lower proteinuria. Urinalysis was negative for proteinuria in 66%, trace in 16%, 1+ in 9% and 2+ in 9%; adults were more likely to have proteinuria detectable on urinalysis, but the presence and degree of proteinuria did not correlate with age (Supplemental Table 1, p = 0.83). In patients with proteinuria, urinary albumin excretion was low (< 50% of total protein excretion) in the majority (78%) (Table 1). When patients were analyzed in four age groups (≤10 y, 11–20 y, 21–30 y, 31–40 y), the percentage of patients with proteinuria did not significantly increase with age (p = 0.79) (Fig. 1B). However, the percentage of patients with albuminuria increased significantly with age (p = 0.03) (Fig. 1C). Serum albumin (median 4.0, 3.0–4.5) was preserved in most patients; 6 had only mild hypoalbuminemia (Table 1). Seven patients (18%) met the definition for chronic kidney disease defined as eGFR < 60 mL/min/1.73 m² and proteinuria; all 7 were adults with median age of 32.8 (20.6–37.9) years (Table 1).

Fig. 1.

Changes in Markers of Kidney Disease in AS from childhood to adulthood.

A. eGFR declines significantly with older age (y = −1.5285× +114.01, R² = 0.334, p = 0.002). B. UPCR in AS was not correlated with age (p = 0.83), in contrast to UACR (p = 0.04) which is more likely to be increased in older children and adults (C). D. Urine/serum osmolality ratio in AS decreases with older age (p < 0.001).

We explored associations between markers of kidney disease (serum creatinine, serum cystatin C, eGFR, UPCR, UACR) and other multisystem manifestations of AS including the continuous measures of BMI percentile, HbA1C, HOMA-IR, duration of T2DM, triglycerides, LDL and HDL cholesterol, Z scores of systolic and diastolic blood pressures and the dichotomous measures of abnormal kidney imaging findings, T2DM, portal hypertension and cardiomyopathy. Before adjusting for age, we identified significant associations between markers of kidney disease and HbA1c, HOMA-IR, T2DM, HDL cholesterol, triglycerides and portal hypertension (Supplementary Table 1). Patients with T2DM had significantly lower eGFR (75 [37–116] mL/min/1.73 m²) than those without T2DM (102 [25–143] mL/min/1.73 m²) (p = 0.0035; Supplementary Table 1). In patients with T2DM, duration of T2DM was not associated with eGFR (r = −18; p = 0.53), UPCR (r = −0.03; p = 0.92) or UACR (r = 0.24; p = 0.41). Similarly, patients with portal hypertension had significantly lower eGFR (p = 0.04), higher serum cystatin C (p = 0.006) and higher serum creatinine (p = 0.02) in comparison to those without portal hypertension. However, after adjusting for age, there were no significant associations between kidney disease and HOMA-IR, T2DM, triglycerides and portal hypertension. The presence or degree of proteinuria was not significantly associated with age, BMI percentile, insulin resistance, T2DM, hypertriglyceridemia, Z scores of systolic and diastolic blood pressures, portal hypertension or cardiomyopathy.

Hematuria was uncommon. Urine dipstick was positive for hemoglobin and increased number of red blood cells in one patient (Table 1, patient # 30) who had low eGFR (32 mL/min/1.73 m2) poorly controlled T2DM, portal hypertension and late onset cardiomyopathy. Normoglycemic glucosuria was present in two patients. There were no other abnormalities in serum or urine that suggested a generalized tubulopathy in these two patients or in the cohort in general. Median urine/serum osmolality ratio, a marker of ability to concentrate urine, decreased with age (3.06 in children 10 years and younger, in comparison to 1.19 among adults ages 31–40 years, normal 1–3; p = 0.002) (Fig. 1D). While some patients reported polyuria and polydipsia, we were unable to collect sufficiently reliable data to further characterize these symptoms.

Renal USG images were obtained in 37 patients; in one patient (Table 1, # 32) satisfactory USG images could not be obtained due to a large amount of subcutaneous fat. In 30% of the patients (11/37), USG showed renal parenchymal hyperechogenicity that was corticomedullary in 4 and exclusively medullary in 7 patients (Fig. 2A). Kidney size was normal in most patients; 3 patients had small kidneys (1 of these 3 had corticomedullary hyperechogenicity) and 6 patients had enlarged kidneys (4 of these 6 had medullary hyperechogenicity) including 2 patients with kidney length above 4 SD (Table 1, patients 12 and 22). The youngest patient with medullary hyperechogenicity was 8 years old. Abdominal MRI was performed in 26 patients. Medullary hyperintensity on MRI was present in 9 patients (36%) including 2 with normal USG; 2 patients who had corticomedullary hyperechogenicity on USG had no evidence of hyperintensity on MRI and 2 others with corticomedullary hyperechogenicity on USG had only medullary hyperintensity on MRI (Fig. 2B). Four patients (17%) had several cysts (4–20 mm) in both kidneys (Fig. 2C), and two of them had concomitant medullary hyperintensity. Parathyroid hormone levels, serum calcium, phosphorus, uric acid and vitamin D levels and urinary calcium excretion were normal in these patients with renal parenchymal hyperechogenicity. Only one patient had normocalcemic hypercalciuria but had no evidence of medullary hyperechogenicity or hyperintensity of renal USG or MRI.

Fig. 2.

Renal imaging findings of patients with Alström syndrome. A. Renal ultrasonography image of right kidney of 35 years old female (Table 1, # 35) showing hyperechogenicity of renal medulla. B. Coronal abdominal MRI image of 18 years of male patient (Table 1, #26) showing hyperintensities in the medulla of both kidneys. C. Coronal abdominal MRI image of 34 years old male patient (Table 1, # 34) showing multiple discrete cysts (arrows).

When patients with and without renal parenchymal hyperechogenicity were compared, no significant statistical differences were identified in markers of kidney disease (serum creatinine, serum cystatin C, eGFR, UPCR, UACR), or presence of other systemic disease including T2DM, cardiomyopathy, and liver hyperechogenicity on USG (Supplementary Table 1). Similarly, no correlations were identified between kidney size and markers of kidney disease, T2DM, cardiomyopathy, or liver hyperechogenicity on USG (Supplementary Table 1).

4. Discussion

Renal disease in AS has not been well characterized, although it is recognized as a major cause of morbidity in these patients. In a retrospective review of 161 patients, Marshall et al. reported that 50% of patients with AS (ages 5–42 years) had decreased renal function (defined as serum creatinine > 2 mg/dl, blood urea nitrogen > 20 mg/dL or proteinuria); furthermore, end-stage renal disease was the cause of death in 23% of patients between the ages of 22 and 48 years [20]. Although our definition of renal dysfunction differs from that in the aforementioned study, our data confirm that chronic progressive renal disease is an important cause of morbidity in AS; almost 20% of the cohort between the ages of 20 to 38 years met the definition for chronic kidney disease.

Given the extreme rarity of AS, it is not surprising that detailed biochemical and imaging characteristics of kidney disease were not previously analyzed in the context of other organ dysfunction. Hence, the pathogenic mechanisms that contribute to the progression of renal disease in AS are unknown. Initial analysis of our multisystem data revealed significant associations between kidney disease and insulin resistance/T2DM and portal hypertension. Given that the likelihood of kidney disease, T2DM, liver disease in AS all increase with age, we suspected that these associations may be a reflection of the aging process. Hence, we performed multivariable regression analysis to remove the possible effect of age; there was no significant association between kidney disease and insulin resistance/T2DM or portal hypertension after adjusting for age. We cannot eliminate the possibility that insulin resistance/T2DM, portal hypertension, hypertension, dyslipidemia and cardiomyopathy contribute to renal disease by an additive effect.

Renal histopathology has been reported in the literature in only a few patients with AS. Interstitial fibrosis and tubular atrophy associated with varying degrees of glomerular sclerosis are consistent findings [18–20]. A similar pattern of fibrosis is seen in other organs of patients with AS, including heart, lungs and liver [20]. This suggests that a common causal relationship exists, possibly directly related to the absence of the ALMS1 protein or to a secondary chronic inflammatory response to a cellular insult that results from loss of function of ALMS1. At least two lines of evidence support the former. First, in vitro knockdown of Alms1 in murine-derived inner-medullary collecting duct cells causes stunted cilia and prevents these cells from increasing calcium influx in response to mechanical stimuli [7]. Second, the Alms^{L2131X/L2131X} homozygous truncating mutation mouse model shows age-dependent cilia loss in the proximal tubules, and kidneys of older mutant mice show foci of apoptosis and proliferation in areas restricted to tubules that have lost cilia; this is accompanied by tubular dilation and mild proteinuria [7]. It is notable that histologic findings consistent with diabetic nephropathy (such as mesangial expansion, glomerular basement membrane thickening, podocyte injury, and glomerular sclerosis) have not been described in reports of the rare renal biopsies performed in AS patients. This is consistent with our finding of no significant association of kidney disease with T2DM when effects of age were removed. These 2 lines of evidence suggest that diabetes is not the major contributor to renal dysfunction in AS. Hypertension, which is also prevalent in AS, may cause tubular injury but is usually accompanied by simultaneous and sometimes severe changes in the vasculature and glomeruli. These hypertension-related findings have not been described in AS kidney biopsies. However, it is possible that the presence of diabetes and hypertension have an additive effect on renal disease progression in AS patients. Furthermore, because patients with AS generally have more frequent screening for metabolic complications compared to the general population, initiation of antihypertensive and antihyperglycemic medications early in the course of disease development may modulate the course of renal pathology development.

Proteinuria was present in more than half of the patients and was consistent with tubular rather than glomerular origin in most patients, based on the low percentage of albumin excretion. Not surprisingly, the urinalysis was negative in most cases, since urinary dipstick is not sensitive for the detection of proteins other than albumin. This low level of tubular-derived proteinuria would be expected in the context of the tubulointerstitial damage seen on renal biopsies of patients with AS. Only four patients (Table 1, patients # 15, 33, 34, and 36) had values of albuminuria that were consistent with glomerular proteinuria and had total daily protein excretion > 500 mg. Since patients 34 and 36 had T2DM for 17.2 and 20.6 years, respectively, their proteinuria may be partially attributable to long term effects of diabetes. However, patient 15 (age 11.5 y) had T2DM for 5 years and patient 33 (age 32.8 y) did not have T2DM; they both had hypertension (for 5 and 7 years, respectively), which may have contributed to their proteinuria.

On USG, one third of the patients had renal parenchymal hyperechogenicity that was limited to the medulla in most patients. The differential diagnosis of the USG finding of medullary hyperechogenicity is broad and includes true nephrocalcinosis, hyperoxaluria, Bartter’s syndrome, Dent’s disease, distal renal tubular acidosis, deposition of urate or proteins, medullary sponge kidney, and autosomal recessive polycystic kidney disease [28–30]. We did not identify evidence for hyperparathyroidism, vitamin D intoxication, hypercalciuria, or generalized tubular dysfunction in our patient cohort. In addition, we found no correlation between the renal hyperechogenicity and eGFR or proteinuria. The radiologic abnormalities of the kidneys in AS might be primarily caused by the absence of ALMS1 protein resulting in cilia dysfunction, with subsequent alterations in the flux of ions, changes in interstitial milieu, or cell death. Disorders of the non-motile primary cilia are often associated with fibrocystic disease of kidneys [31–34]. Under the same NIH protocol, we previously identified similar radiologic findings in the kidneys of patients with autosomal recessive polycystic kidney disease [29] as well as in some obligate heterozygotes for the same disorder [28]. The types of fibrocystic kidney disease observed in ciliopathies include polycystic kidneys and nephronophthisis, which are both associated with decreased ability to concentrate urine [34]. Similarly, the urine/serum osmolality ratio of this cohort of AS patients decreased with age, suggesting decreased ability to concentrate urine as patients got older. While other metabolic disturbances in AS may also be contributing to this decline in urinary concentration ability (e.g., hyperglycemia), ciliary dysfunction secondary to lack of ALMS1 protein affecting tubule and collecting duct function may also play a role.

Pathogenic variants in ALMS1 that cause AS are almost all truncating mutations resulting in the early termination of ALMS1 [5]. Consistently, the ALMS1 mutations we identified in our AS cohort were all either nonsense or frameshifting variants. Truncating mutations abolish protein expression, or may produce stable truncated proteins with perturbed function. If the truncated versions of ALMS1 protein were expressed, they would be visualized in a diffuse pattern due to the loss of the centrosome-targeting domain at the C terminus [9]. However, in 14 of the 16 fibroblast cell lines from AS patients with various bi-allelic truncating ALMS1 mutations, no ALMS1 staining was detected, consistent with absent protein expression [35]. Given these published data, and the fact that all our patients had bi-allelic truncation mutations, we have not aimed to evaluate genotype-phenotype correlations in this study. It is possible that in patients with certain mutations, shorter isoforms of ALMS1 may be expressed in some tissues potentially explaining the variability of severity of clinical features.

There are limitations of this study. Although data were collected prospectively, we present data from a single time point rather than longitudinally, and renal biopsies were not performed. In addition, even though this study represents the largest cohort of AS patients who underwent comprehensive prospective multisystem evaluations, the relatively small number of patients (due to the rarity of the disease) limits statistical significance of some analysis.

In summary, kidney disease in AS is most likely a primary manifestation of the syndrome caused by the primary disease process due to lack of ALMS1 protein, while additive effects of other manifestations including T2DM and liver disease complicated with portal hypertension are possible. Approximately one-third of patients with AS have diffuse parenchymal hyperechogenicity on USG most commonly affecting the medulla. These radiologic findings are not associated with other markers of kidney disease or T2DM, or cardiomyopathy. While strict control of T2DM may decrease kidney-related morbidity and mortality in AS, identification of novel therapies that target the primary intracellular defects in kidneys as well other involved organs in AS is needed.