Age-related changes in hematological biomarkers in common marmosets

Abstract

Researchers and veterinarians often use hematology and clinical chemistry to evaluate animal health. These biomarkers are relatively easy to obtain, and understanding how they change across healthy aging is critical to clinical care and diagnostics for these animals. We aimed to evaluate how clinical biomarkers from a chemistry profile and complete blood count (CBC) change with age in common marmosets (Callithrix jacchus). We assessed blood samples collected during routine physical exams at the Southwest National Primate Research Center and the University of Texas Health San Antonio marmoset colonies from November 2020 - November 2021. We found that chemistry and CBC profiles varied based on facility, sex, and age. Significant changes in albumin, phosphorus/creatinine ratio, albumin/globulin ratio, amylase, creatinine, lymphocyte percent, hematocrit, granulocytes percent, lymphocytes, hemoglobin, red cell distribution width (RDWs), and platelet distribution width (PDWc) were all reported with advancing age. Aged individuals also demonstrated evidence for changes in liver, kidney, and immune system function compared to younger individuals. Our results suggest there may be regular changes associated with healthy aging in marmosets that are outside of the range typically considered as normal values for healthy young individuals, indicating the potential need for redefined healthy ranges for clinical biomarkers in aged animals. Identifying animals that exhibit values outside of this defined healthy aging reference will allow more accurate diagnostics and treatments for aging colonies.

Graphical Abstract

Introduction:

Clinicians, both for human and animal populations, often use standard metrics of physiological status, such as blood chemistry, to evaluate and diagnose significant health concerns. Reference values for normal ranges of these biomarkers are imperative for accurate and efficient assessments (Gounden et al., 2023; Lala et al., 2023; Nah et al., 2018). For many species there are differences in these reference ranges associated with sex and age of the individuals. Thus, dividing the analyses by age, sex, and facility allows for a more complete understanding of how demographic factors change serum chemistry and complete blood count parameters (Buchl & Howard, 1997; Chichester et al., 2015; Karal-ogly et al., 2023; Kramer et al., 2022; Li et al., 2022). Healthy aging has been found to be associated with changes in hematology and clinical chemistry values (Johnson, 2006). Reference values have long been used to determine what normal ranges are in humans and non-human primates (Ceriotti et al., 2009; Geffré et al., 2009). However, age, sex, and pregnancy have all been shown to alter reference ranges for hematology and clinical chemistry across humans and non-human primates (Bakrim et al., 2018; Buchl & Howard, 1997; Chichester et al., 2015). While these changes have long been documented, understanding how these changes represent normal healthy aging has yet to be examined in common marmosets (Callithrix jacchus). In addition, these clinical ranges are a significant means to determine health status by animal care staff. Clarification of typical healthy ranges along all stages of life will improve standard of care throughout the lifespan.

Common marmosets are small-bodied, South American primates that have become an important translational model of human aging and development (Tardif, 2019). Their increased demand in research, in part, can be attributed to their shorter life span and smaller size compared to many other primates (Ross et al., 2012; Ross & Salmon, 2019). Additionally, their closer phylogenetic relationship with humans often makes them a more translatable model compared to murine models (Marini et al., 2018). Despite this increasing popularity of use in biomedical studies, only one recent study has defined clinical parameters for chemistry and complete blood count (CBC) panels at different stages of life for this species (Kramer et al., 2022). Previous papers which have examined marmoset serum chemistry and CBC in marmosets addressed very small numbers of animals that failed to accurately capture the range of health values for clinical parameters (McNees et al., 1982). Evaluations of serum chemistry and CBC values in other nonhuman primate species have revealed significant differences associated with age, sex, and often facility (Buchl & Howard, 1997; Chichester et al., 2015; Karal-ogly et al., 2023; Kramer et al., 2022; Li et al., 2022). Given this, there is a need to further evaluate variance in value ranges that may be associated with factors such as age, sex, or the facility that the marmosets are housed in.

Healthy and expected changes occur in metabolic demand, cognation, and immune system that are a function of a marmoset’s changing needs with age (Goodroe et al., 2021; Phillips et al., 2019), and mirror what is reported for human aging (Nah et al., 2018). For example, resting energy expenditure has been found to decrease as individuals age (Frisard et al., 2007; Palmer & Jensen, 2022; Vaughan et al., 1991); and marmosets perform more poorly in cognitive tasks such as detoured reach task (Phillips et al., 2019; Ross et al., 2019). These physiological changes with age can influence the resulting serum chemistry and CBC values that are taken as a part of routine bloodwork (Hoffman et al., 2016; Kramer et al., 2022; McNees et al., 1982, 1984), which are often used as indicators for health issues in many marmoset colonies. Yet, the ways these parameters change with normal healthy aging have not been well established to allow for accurate diagnostics.

Only a few small changes in blood urea nitrogen, calcium, total protein, alkaline phosphatase, and serum albumin biomarkers have previously been associated with age in marmosets (Kramer et al., 2022; McNees et al., 1982, 1984). However, these examinations were small in sample size (McNees et al., 1982, 1984), utilized a limited age range (i.e. the oldest age of marmoset evaluated for these studies was 12 years of age), or the marmosets only came from a single colony (Kramer et al., 2022). Recent work in aging marmosets has found that the marmoset can live substantially longer than was previously identified in captivity, with a maximum life span of 21 years reported (Nishijima et al., 2012; Ross et al., 2019). Using this expanded knowledge of the normal lifespan of marmosets, we aimed to identify differences in the average ranges in blood biomarkers associated with healthy aging marmosets at ages across nearly all of the maximum lifespan of this species, in the colonies at the Southwest National Primate Research Center (SNPRC) and University of Texas Health San Antonio (UT Health).

Materials and Methods:

Subjects:

Two hundred and fifty-two common marmosets from the Southwest Primate Research Center (SNPRC) and 80 common marmosets (Callithrix jacchus) from the Barshop Institute for Longevity and Aging Studies at University of Texas Health San Antonio (UT Health) received physical exams between November 2020 - November 2021. Ages ranged from 0.77 – 14.71 years for SNPRC and 0.6 – 16.2 for UT Health (Table 1). Weight ranges were similar across the two colonies, and none of the females evaluated were pregnant at the time of assessment or immediately post-partum (within 30 days). The UT Health colony was derived in 2006 from SNPRC animals and both colonies follow the same husbandry operating procedures es (Layne & Power, 2003; Ross et al., 2017). Marmosets at both facilities were typically pair-or family-housed (male-female breeding pair with offspring). Marmosets were fed a base diet that was either a mix of Purina Mazuri^® marmoset diet (St. Louis, MO) and Harlan-Teklad^™ marmoset purified diet (Indianapolis, IN), or only the purified diet. Food enrichment items were offered daily and included things such as Zupreem^® marmoset diet (Mission, KS), dried cranberries, raisins, or Teklad^™ primate enrichment mix, (Indianapolis, IN). The UT Health colony is housed with barrier protocols which include irradiated base diet, irradiated enrichment mix, autoclaved water, and increased PPE, in order to reduce the risk of pathogen introduction (Layne & Power, 2003; Ross et al., 2017). Although the institutions have some facility differences, the marmosets at both SNPRC (mean age = 4.71; oldest for this dataset = 13.0) and UT Health (mean age = 5.22; oldest for this dataset = 16.2) have similar survival curves following the physical exam (Supplementary Figure 1). Marmoset procedures were approved by the IACUCs at each institution and adhered to the American Society of Primatologists Principles for the ethical treatment of non-human primates.

Table 1.

Demographics for marmosets included in this study from the Southwest National Primate Research Center and UT Health San Antonio colonies.

Facility	Age range (years)	Age category ratio (juvenile: adult: peri- geriatric: geriatric)	Sex ratio (M: F)	Weight range (g)
SNPRC	0.77 – 13.71	61: 160: 15: 14	129: 122	258 – 637
UT Health	0.6 – 16.2	16 :43: 6: 12	44: 33	265 – 612

Sample collection:

At both facilities each marmoset received an annual physical exam as a part of routine colony health monitoring. For the physical exam animals are fasted in the morning (removal of food at 8 am), given a dose of 20 mg/kg of ketamine intra-muscular for sedation, and evaluated by a veterinarian. To evaluate CBC and blood chemistry 1.5 ml of blood was collected from each animal from the femoral vein with 0.5 mL placed in a tube with EDTA for CBC, and the remaining 1.0 ml placed in a serum separator tube for serum chemistry. Blood tubes were transported to either the SNPRC veterinary clinical pathology core or the UT Health Lab animal services lab. Samples at the SNPRC were analyzed using the UniCel DxH 800 Coulter Cellular Analysis System for complete blood counts. From the beginning of the study until March 29^th of 2021 the UniCel DxC 600 Synchron Clinical System was used for serum chemistry panels. On March 29, 2021, SNPRC switched to using the DxC700 AU Chemistry Analyzer, which was used through the end of study. Both machines were used for a short time at SNPRC to establish reliability and the clinical pathology team found no significant differences between these two periods. For UT Health the complete blood counts were performed using Vetscan HM5 and chemistry panel by Vetscan VS2 machines. The results from the two sites have previously been compared and validated for consistency (Ross et al 2019). Exclusion criteria for this study were any animals that were flagged for a follow-up exam by the veterinarians, or if an animal had more than eight parameters that were found to be significant outliers during the analysis. Animals were also excluded if they were on any long-term clinical care support, e.g. extra protein, or iron for anemia. Animals who were pregnant were removed from the analysis. Animals were included if they were deemed healthy or “within normal limits” by the veterinarian at the time of physical exam.

Statistical Analysis:

We removed outliers for individual parameters that were more than ± 2 standard deviations from the mean and individuals who had more than eight parameters outside of ± 2 standard deviations away from the mean, leaving 251 SNPRC animals and 77 UT Health animals (Table 1). To test if age was associated with chemistry and CBC profiles, we divided age into four age categories: juvenile, adult, peri-geriatric, and geriatric (Ross et al., 2012, 2019; Tardif, 2019). Juvenile individuals are those 6-months-old through 1.99 years old that are still growing and maturing (Ross et al., 2012, 2019; Tardif, 2019). Adult individuals are sexually mature and over 2 years old and younger than 7.99 years old (Ross et al., 2012, 2019; Tardif, 2019). Peri-geriatric individuals are those in the 8 - 10 year range; individuals in this age range typically exhibit a transitional phenotype for many physiological parameters (Ross et al., 2012, 2019; Tardif, 2019; Mustoe et al. 2023). Animals over the age of 10 typically exhibit geriatric phenotypes (Ross et al., 2012, 2019; Tardif, 2019; Mustoe et al. 2023). We ran multivariate ANOVAs with Bonferroni corrections to test for age, sex, and site main and interaction effects for each parameter in a chemistry panel and CBC panel.

Results:

Three-way interactions (comparison of age, sex, and facility):

There were only two parameters found to have significant three-way interactions; potassium [df = 2, 268, F = 4.112, P-value = 0.0017] and hematocrit [df = 3, 293, F = 3.453, P-value = 0.016] (Table 2; Supplementary Figure 2).

Table 2.

Mean, standard deviation, minimum, and maximum by age category. Bold are the variables which were only assessed at SNPRC, italics are variables only assessed at UT Health.

	Juvenile				Adult				Peri-geriatric				Geriatric
Parameter	n	Mean ± SD	Min. – Max.		n	Mean ± SD	Min. – Max		n	Mean ± SD	Min. – Max		n	Mean ± SD	Min. – Max
Albumin	68	4.06 ± 0.51	2.7 – 4.8		190	3.75 ± 0.44	2.7 – 4.7		20	3.52 ± 0.43	2.7 – 4.2		21	3.40 ± 0.44	2.7 – 4.4
Alkaline phosphatase	67	151.19 ± 57.60	43.0 – 267.0		194	71.07 ± 30.19	5.0 – 231.0		19	76.10 ± 43.51	34.0 – 223.0		22	86.90 ± 35.57	41.0 – 168.0
Alanine transaminase	72	43.04 ± 57.64	4.0 – 375.0		191	39.92 ± 54.38	4.0 – 319.0		20	25.50 ± 22.87	6.0 – 91.0		23	21.52 ± 16.21	7.0 – 65
Amylase	14	71.71 ± 9.98	59.0 – 91.0		39	86.33 ± 12.47	66.0 – 114.0		6	97.50 ± 17.49	77.0 – 116.0		10	92.50 ± 16.22	69.0 – 122.0
Total Bilirubin	71	0.20 ± 0.09	0.1 – 0.4		191	0.19 ± 0.09	0.0 – 0.4		20	0.17 ± 0.10	0.1 – 0.4		23	0.19 ± 0.10	0.1 -0.4
Blood Urea Nitrogen	72	20.27 ± 5.29	6.0 – 35.0		193	19.39 ± 5.06	8.0 – 38.0		18	20.00 ± 5.62	10.0 – 29.0		17	19.52 ± 6.67	10.0 – 36.0
Calcium	69	10.27 ± 0.67	9.0 – 11.6		190	9.76 ± 0.55	8.1 – 11.2		20	9.61 ± 0.63	8.0 – 10.8		23	9.67 ± 0.76	8.3 – 11.1
Phosphorus	66	4.12 ± 0.88	1.7 – 5.8		191	3.61 ± 0.77	1.7 – 5.7		18	3.92 ± 0.92	2.5 – 6.0		23	3.99 ± 0.96	2.3 – 5.9
Creatinine	72	0.30 ± 0.10	0.2 – 0.6		194	0.32 ± 0.12	0.1 – 0.7		19	0.40 ± 0.12	0.2 –0.7		19	0.46 ± 0.14	0.2 – 0.7
Phosphorus Creatinine Ratio	66	14.91 ± 5.65	4.25 – 25.0		191	12.58 ± 4.93	2.67 – 24.5		20	10.63 ± 3.50	5.00 – 20.0		23	8.18 ± 3.85	2.63 – 15.0
Glucose	67	190.43 ± 62.38	62.0 – 315.0		191	162.44 ± 61.00	29.0 – 317.0		19	150.10 ± 46.55	42.0 – 280.0		23	141.86 ± 39.32	79.0 – 231.0
Sodium	71	150.28 ± 3.91	143.0 – 161.0		184	149.79 ± 4.22	139.0 – 161.0		17	150.41 ± 3.96	145.0 – 159.0		20	151.50 ± 4.12	142.0 – 159.0
Potassium	66	3.36 ± 0.80	2.4 – 5.8		178	3.35 ± 0.78	2.1 – 5.9		19	3.62 ± 0.96	2.6 – 5.9		20	3.70 ± 0.99	2.8 – 5.8
Total Protein	68	6.52 ± 0.61	5.0 – 8.0		190	6.31 ± 0.72	4.7 – 7.9		19	6.43 ± 0.86	5.0 – 8.0		23	6.57 ± 0.88	5.0 – 7.9
Globulin	71	2.46 ± 0.49	1.7 – 3.7		192	2.56 ± 0.57	1.3 – 4.1		17	2.74 ± 0.55	2.2 – 4.0		18	2.85 ± 0.60	2.1 -3.9
Creatine Phosphokinase	28	1235.50 ± 734.91	471.0 – 3316.0		78	930.16 ± 564.19	95.0 – 3216.0		4	1192.75 ± 1169.58	427.0 -2913.0		11	881.45 ± 555.12	314.0 – 2106.0
Triglycerides	27	429.03 ± 331.31	69.0 – 1620.0		77	286.29 ± 251.00	36.0 – 1197.0		5	500.20 ± 294.83	66.0 – 818.0		12	365.00 ± 235.55	61.0 -762.0
Lactate dehydrogenase	29	698.72 ± 600.17	215.0 – 2083.0		78	566.39 ± 532.61	135.0 – 2287.0		5	641.20 ± 361.39	197.0 – 1088.0		12	398.91 ± 225.11	140.0 – 859.0
Gammaglutamyl transferase	29	12.96 ± 12.28	3.0 – 64.0		11	10.23 ± 9.67	2.0 – 54.0		5	10.80 ± 6.87	4.0 – 20.0		11	10.63 ± 7.01	4.0 - 26
Chloride	57	104.64 ± 3.40	98.0 – 112.0		147	105.65 ± 2.86	99.0 – 112.0		14	106.07 ± 2.30	102.0 – 109.0		13	105.46 ± 3.59	100.0 – 111.0
Albumin/ Globulin Ratio	52	1.79 ± 0.30	1.0 – 2.3		148	1.61 ± 0.28	0.9 – 2.3		14	1.40 ± 0.18	1.1 – 1.7		13	1.37 ± 0.21	0.9 – 1.7
Cholesterol	55	151.67 ± 61.63	68.0 – 340.0		148	145.58 ± 42.83	64.0 – 287.0		14	147.35 ± 51.83	55.0 – 236.0		13	171.61 ± 51.02	71.0 – 275.0
Blood Urea Nitrogen/Creatinine Ratio	30	54.78 ± 18.26	22.5 – 93.3		99	54.57 ± 17.28	15.0 – 93.0		13	51.86 ± 8.94	37.5 – 65.0		13	62.14 ± 20.89	25.0 – 90.0
Bicarbonate	56	24.01 ± 2.93	17.0 – 30.0		148	25.10 ± 3.22	18.0 – 32.0		12	26.00 ± 2.79	21.0 – 30.0		12	24.83 ± 4.62	18.0 – 32.0
Anion Gap	53	22.69 ± 3.02	16.8 – 30.2		148	20.65 ± 3.58	13.2 – 30.5		13	19.49 ± 2.68	15.0 – 22.9		12	21.36 ± 3.12	16.1 – 26.1
Nucleated Red Blood Cells	55	0.72 ± 0.54	0.1 - 2.5		150	0.74 ± 0.62	0.0 – 3.2		14	0.59 ± 0.44	0.1 – 1.4		11	0.40 ± 0.19	0.2 – 0.7
White Blood Cells	66	7.46 ± 2.76	1.7 – 14.08		186	5.81 ± 2.26	1.7 – 13.91		20	6.18 ± 2.49	2.9 – 12.18		23	6.10 ± 3.13	2.1 – 14.07
Lymphocytes	62	4.44 ± 1.75	0.49 – 8.32		193	3.00 ± 1.61	0.16 – 7.63		20	2.56 ± 1.21	0.61 – 5.90		22	2.36 ± 1.58	0.08 – 6.14
Monocytes	70	0.38 ± 0.80	0.00 – 3.86		182	0.43 ± 0.80	0.00 – 4.14		20	0.65 ± 0.92	0.10 – 3.39		22	0.62 ± 1.03	0.05 – 3.27
Neutrophils	15	3.43 ± 1.25	1.21 – 5.35		40	3.47 ± 1.65	1.50 – 7.45		6	4.33 ± 2.29	1.37 – 7.38		12	3.06 ± 1.20	1.40 – 5.17
Eosinophil	55	0.10 ± 0.10	0.0 – 0.4		145	0.09 ± 0.09	0.0 – 0.4		14	0.07 ± 0.07	0.0 – 0.2		11	0.09 ± 0.09	0.0 – 0.3
Basophils	53	0.01 ± 0.03	0.0 – 0.1		142	0.02 ± 0.04	0.0 – 0.1		14	0.02 ± 0.04	0.0 – 0.1		11	0.01 ± 0.04	0.0 – 0.1
Lymphocytes percent	72	68.01 ± 18.59	11.6 – 91.0		186	54.75 ± 19.07	10.7 – 87.8		20	43.28 ± 14.30	11.4 – 63.7		21	48.06 ± 18.60	13.2 – 71.0
Monocytes percent	70	3.85 ± 6.39	0.0 – 32.9		178	4.81 ± 6.80	0.5 – 37.5		19	6.41 ± 6.51	1.4 – 26.3		20	5.42 ± 6.91	0.8 – 28.9
Neutrophils percent	16	33.31 ± 14.64	14.4 – 58.1		41	38.18 ± 14.13	13.1 – 70.5		5	44.32 ± 13.19	25.5 – 60.6		12	40.10 ± 11.04	26.2 – 61.1
Eosinophil percent	54	1.57 ± 1.35	0.0 – 5.9		142	1.62 ± 1.27	0.0 – 7.0		14	1.25 ± 0.84	0.1 – 2.5		9	1.35 ± 0.69	0.6 – 2.3
Basophils percent	56	0.45 ± 0.42	0.0 – 2.1		143	0.59 ± 0.45	0.0 – 2.0		14	0.71 ± 0.49	0.2 – 2.2		10	0.57 ± 0.32	0.2 – 1.2
Granulocyte percent	56	22.54 ± 9.81	7.9 – 49.7		145	33.01 ± 12.86	8.0 – 63.3		14	44.33 ± 7.57	29.6 – 52.2		10	40.47 ± 13.70	26.1 – 62.5
Red Blood Cells	70	7.21 ± 0.71	5.37 – 8.86		188	6.81 ± 0.78	4.67 – 8.86		17	6.51 ± 0.69	5.16 – 7.71		18	6.57 ± 1.06	4.75 – 8.10
Hemoglobin	72	14.96 ± 1.56	10.5 – 18.1		189	14.85 ± 1.81	9.8 – 19.0		17	13.96 ± 1.42	11.7 – 16.6		18	13.83 ± 2.12	9.8 – 17.8
Hematocrit	73	46.26 ± 5.27	22.20 – 56.71		195	45.22 ± 7.03	2.88 – 59.30		20	40.70 ± 7.93	21.80 – 51.50		21	39.60 ± 10.25	13.43 – 53.97
Mean corpuscular volume	68	64.90 ± 3.19	59.0 – 72.4		188	67.09 ± 3.12	59.1 – 74.6		19	67.63 ± 3.40	61.0 – 72.4		22	66.08 ± 3.69	59.0 – 73.0
Mean corpuscular hemoglobin	73	20.75 ± 1.41	16.7 – 24.3		195	21.80 ± 1.63	18.3 – 29.7		20	21.44 ± 1.71	17.1 – 25.3		23	21.07 ± 1.42	18.6 – 23.5
Mean corpuscular hemoglobin concentration	72	32.07 ± 0.85	30.0 – 34.2		187	32.10 ± 0.84	29.9 – 34.5		19	31.77 ± 0.46	31.1 – 32.6		23	32.08 ± 0.91	30.7 – 34.3
Red cell distribution width variation	67	16.04 ± 1.490	13.0 – 19.6		190	14.90 ± 1.34	12.8 – 19.6		20	15.59 ± 1.50	13.7 – 19.1		23	15.11 ± 1.26	12.9 – 18.0
Red cell distribution width standard deviation	15	49.28 ± 2.88	45.3 – 54.7		39	48.27 ± 3.36	41.4 – 55.5		5	47.48 ± 3.86	42.2 – 52.3		12	44.08 ± 2.71	40.6 – 50.8
Platelet	70	426.00 ± 118.71	220.0 – 842.0		186	529.08 ± 135.04	198.0 – 838.0		18	535.00 ± 153.15	322.0 – 780.0		21	525.61 ± 124.57	309.0 – 733.0
Mean platelet volume	70	8.90 ± 0.70	7.3 – 10.3		186	8.48 ± 0.73	7.0 – 10.1		19	8.46 ± 1.00	6.8 – 9.9		21	8.33 ± 0.92	6.8 – 10.2
Plateletcrit	16	0.44 ± 0.08	0.27 – 0.60		40	0.55 ± 0.13	0.32 – 0.80		5	0.61 ± 0.11	0.47 – 0.77		12	0.50 ± 0.11	0.35 – 0.75
Platelet distribution width	16	17.51 ± 2.30	12.2 – 20.4		41	16.02 ± 2.24	12.2 – 20.2		6	16.13 ± 1.81	13.7 – 18.8		12	15.11 ± 2.54	12.0 – 20.2
Platelet distribution width	16	40.75 ± 1.50	36.9 – 42.7		41	39.93 ± 1.63	36.9 – 42.8		6	39.63 ± 0.78	38.3 – 40.6		12	38.91 ± 1.51	36.7 – 41.8

Two-way interactions:

Several parameters were identified as having significant two-way interactions. Significant interactions between age and site were found for the clinical chemistry markers, alkaline phosphatase [df = 3, 287, F = 3.163, P-value = 0.024], sodium [df = 3, 278, F = 2.722, P-value = 0.044], and total protein [df = 3, 285, F = 4.016, P-value = 0.008] (Supplementary Table 1; Supplementary Figure 3). Significant interactions between site and age were also identified for blood count variables monocytes [df = 3, 279, F = 4.152, P-value = 0.006], lymphocyte percent [df = 3, 283, F = 2.906, P-value = 0.035] , monocyte percent [df = 3, 271, F = 8.211, P-value = <0.001] , hemoglobin [df = 3, 281, F = 3.674, P-value = 0.012] , mean corpuscular volume [df = 3, 281, F = 3.066, P-value = 0.028], and mean corpuscular hemoglobin [df = 3, 295, F = 6.688, P-value = <0.001]. There was one factor, hematocrit, that had a significant two-way interaction between age category and sex [df = 3, 293, F = 3.103, P-value = 0.026] (Supplementary Table 2; Supplementary Figure 3).

Main effects – age:

For clinical chemistry markers, the mean, standard deviation, and range are broken down by age category and listed in Table 2. The parameters listed here were significant main effects for age category: amylase [df = 3, 62, F = 7.99 , P-value = <0.001], creatinine [df = 3, 289, F = 13.21, P-value = <0.001], and anion gap [df = 3, 218, F = 5.75, P-value = <0.001], were all highest in geriatric individuals while albumin [df = 3, 284, F = 15.50, P-value = <0.001], phosphorus/creatine ratio [df = 3, 285 , F = 12.14, P-value = <0.001], albumin/globulin ratio [df = 3, 219, F = 12.403, P-value = <0.001], calcium [df = 3, 287, F = 16.29, P-value = <0.001], and glucose [df = 3, 285, F = 6.02, P-value = <0.001] were all lowest in geriatric individuals. Juveniles had the highest vales for phosphorus [df = 3, 283, F = 7.13, P-value = <0.001***] (Figure 1A – I; Supplementary Table 1). The pairwise comparison results with Bonferroni correction are represented in Figures 1A – I, with significant pairwise comparisons and their p-values represented in number of stars above the age categories being compared. No parameter was found to differ significantly between the peri-geriatric and geriatric age categories.

Figure 1.

Significant parameters from the chemistry ANOVAs for age category. The comparison of: A. albumin, B. phosphorus/creatine ratio, C. albumin/globulin ratio, D. amylase, E. calcium, F. creatinine, G. anion gap, and H. glucose for juveniles, adults, peri-geriatric, and geriatric marmosets. The number of symbols indicate the significance level, * = 0.05, ** = 0.01 and *** = <0.001.

Potassium [df = 3, 268, F = 5.691, P-value = <0.001] and globulin [df = 3, 287, F = 7.175, P-value = <0.001] were significantly different in the main effect of age category but there were no significant post-hoc comparisons. Alkaline phosphatase [df = 3, 287, F = 73.756, P-value = <0.001] had significant main effects of age but after a significant interaction effect the main effects are not simple to interpret but the means for this parameter show juveniles have the highest values (Table 2). Total protein [df = 3, 285, F = 3.111, P-value = 0.026] and potassium [df = 3, 268, F = 4.112, P-value = 0.017] had significant main effects for age but after a significant interaction effect the main effects are not simple to interpret but means for this parameter show geriatric individuals have the highest values ((Supplementary Figure 2 & 3; Supplementary Table 1).

For complete blood cell counts, the mean, standard deviation, and range are broken down by age category and listed in Table 2. All parameters listed here were significant main effects for age category: granulocyte percent [df = 3, 217 , F = 18.23, P-value = <0.001] and platelets [df = 3, 279, F = 11.67, P-value = <0.001] were all highest in geriatric individuals while lymphocytes [df = 3, 281 , F = 16.01, P-value = <0.001], red blood cells [df = 3, 278, F = 8.03 , P-value = <0.001], red cell distribution width [df = 3, 284 , F = 14.51 , P-value = <0.001], mean platelet volume [df = 3, 280, F = 7.55 , P-value = <0.001], red cell distribution standard deviation [df = 3, 63, F = 6.87, P-value = <0.001], plateletcrit [df = 3, 65, F = 4.00 , P-value = 0.01], platelet distribution width [df = 3, 67, F = 3.48, P-value = 0.02], and platelet distribution standard deviation [df = 3, 67, F = 2.92, P-value = 0.04] were all lowest in geriatric individuals. Adults had the lowest values for white blood cells [df = 3, 279, F = 9.30, P-value = <0.001]. (Figures 2A – K; Supplementary Table 2). The pairwise comparison results with Bonferroni correction are represented in Figures 2A – K, with significant pairwise comparisons and their p-values represented in number of stars above the age categories being compared. None of the complete blood count parameters differed between the peri-geriatric and geriatric age categories.

Figure 2.

Significant parameters from the complete blood count ANOVAs for age category. The comparison of: A. granulocyte percent B. lymphocytes, C. red blood cells, D. platelets, E. red cell distribution width, F. mean platelet volume, G. red cell distribution standard deviation, H. white blood cells, I. platelet distribution width, and J. platelet distribution standard deviation for juveniles, adults, peri-geriatric, and geriatric marmosets. The number of symbols indicate the significance level, * = 0.05, ** = 0.01 and *** = <0.001.

Lymphocytes percent [df = 3, 283, F = 19.164, P-value = <0.001], hemoglobin [df = 3, 281, F = 4.048, P-value = 0.007], hematocrit [df = 3, 293, F = 8.663, P-value = <0.001], mean corpuscular volume [df = 3, 281, F = 8.960, P-value = <0.001], and mean corpuscular hemoglobin [df = 3, 295, F = 9.126, P-value = <0.001] had significant interaction effects that make interpreting the significant main effects difficult ((Supplementary Figure 2 & 3; Supplementary Table 2).

Main effects – facility:

The blood analysis machines at the two facilities produced mostly overlapping analyses with a few unique parameters at each institution which were excluded from this analysis. Parameters that were facility specific for SNPRC were creatine phosphokinase, triglycerides, lactate dehydrogenase, gamma-glutamyl transferase, chlorine, albumin/glutamine ratio, cholesterol, blood urea nitrogen/creatinine ratio, bicarbonate, nucleated red blood cells, neutrophils, eosinophils, basophils, neutrophil percent, eosinophil percent, basophils, and granulocyte percent (Table 3). For UT Health amylase, red blood cell width, platelet distribution width, platelet distribution width, and plateletcrit were facility specific (Table 3).

Table 3.

Mean, standard deviation, minimum, and maximum by site. Bold are values measured only at SNPRC, italics are values only measured at UT Health.

		SNPRC				UT Health
Parameter	Units	n	Mean ± SD	Min. – Max.		n	Mean ± SD	Min. – Max.
Albumin	g/dL	233	3.80 ± 0.49	2.7 – 4.8		66	3.72 ± 0.48	2.7 – 4.7
Alkaline phosphatase	IU/L	232	91.37 ± 50.79	25 - 267		70	86.814 ± 51.457	5 - 254
Alanine transaminase	U/L	234	44.38 ± 56.17	4 - 375		72	18.63 ± 27.54	5 - 220
Amylase	U/L	-	-	-		69	85.23 ± 14.92	59.0 – 122.0
Total Bilirubin	mg/dL	236	0.16 ± 0.08	0.0 – 0.4		69	0.31 ± 0.04	0.2 – 0.4
Blood Urea Nitrogen	mg/dL	230	19.12 ± 5.21	6 - 38		70	21.385 ± 4.964	12 - 37
Calcium	mg/dL	232	9.75 ± 0.60	8.0 – 11.3		70	10.22 ± 0.66	8.8 – 11.6
Phosphorus	mg/dL	233	3.78 ± 0.81	1.7 – 6.0		65	3.74 ± 0.97	1.7 – 5.6
Creatinine	mg/dL	234	0.30 ± 0.11	0.1 – 0.7		70	0.40 ± 0.15	0.2 – 0.7
Phosphorus Creatinine Ratio		231	13.15 ± 4.86	2.83 – 25.0		69	10.86 ± 5.97	2.63 – 25.0
Glucose	mg/dL	229	167.90 ± 61.95	29.0 – 317.0		71	161.28 ± 56.06	42.0 – 307.0
Sodium	meq/L	236	148.75 ± 2.88	139.0 – 155.0		56	155.58 ± 4.07	142.0 – 161.0
Potassium	meq/L	237	3.09 ± 0.38	2.1 – 4.4		46	4.96 ± 0.65	3.2 – 5.9
Total Protein	g/dL	235	6.17 ± 0.60	4.7 – 8.0		65	7.17 ± 0.58	5.7 – 8.0
Globulin	g/dL	236	2.34 ± 0.36	1.3 – 3.5		62	3.40 ± 0.36	2.4 – 4.1
Creatine Phosphokinase	U/L	121	1005.07 ± 635.81	95.0 – 3316.0		-	-	-
Triglycerides	mg/dL	121	334.79 ± 276.41	36.0 – 1620.0		-	-	-
Lactate dehydrogenase	U/L	124	584.15 ± 524.53	135.0 – 2287.0		-	-	-
Gammaglutamyl transferase	U/L	122	10.94 ± 10.03	2.0 – 64.0		-	-	-
Chloride	meq/L	231	105.41 ± 3.03	98.0 – 112.0		-	-	-
Albumin/Globulin Ratio		227	1.62 ± 0.30	0.9 – 2.3		-	-	-
Cholesterol	mg/dL	239	148.62 ± 48.99	55.0 – 340.0		-	-	-
Blood Urea Nitrogen/Creatinine Ratio		155	55.02 ± 17.27	15.0 – 93.3		-	-	-
Bicarbonate	meq/L	228	24.86 ± 3.24	17.0 – 32.0		-	-	-
Anion Gap	meq/L	226	21.10 ± 3.50	13.2 – 30.5		-	-	-
Nucleated Red Blood Cells	%	230	0.71 ± 0.58	0.0 – 3.2		-	-	-
White Blood Cells	(10^3/uL)	232	5.63 ± 2.21	1.7 – 13.30		63	8.41 ± 2.54	4.1 – 14.08
Lymphocytes	(10^3/uL)	227	3.37 ± 1.57	0.70 – 8.20		70	2.75 ± 2.14	0.08 – 8.32
Monocytes	(10^3/uL)	235	0.15 ± 0.10	0.00 – 0.50		60	1.60 ± 1.31	0.04 – 4.14
Neutrophils	(10^3/uL)	-	-	-		73	3.47 ± 1.57	1.21 – 7.45
Eosinophils	(10^3/uL)	225	0.09 ± 0.09	0.00 – 0.40		-	-	-
Basophils	(10^3/uL)	220	0.02 ± 0.04	0.00 – 0.10		-	-	-
Lymphocytes percent	%	236	61.54 ± 16.30	17.3 – 91.0		63	38.58 ± 21.66	10.7 – 84.8
Monocytes percent	%	235	2.66 ± 1.39	0.0 – 7.0		52	14.05 ± 11.58	0.5 – 37.5
Neutrophils percent	%	-	-	-		74	37.85 ± 13.77	13.1 – 70.5
Eosinophils percent	%	219	1.57 ± 1.24	0.0 – 7.0		-	-	-
Basophils percent	%	223	0.56 ± 0.44	0.0 – 2.2		-	-	-
Granulocyte percent	%	225	31.44 ± 13.27	7.9 – 63.3		-	-	-
Red Blood Cells	(10^6/uL)	221	6.79 ± 0.74	4.67 – 8.86		72	7.15 ± 0.90	4.70 – 8.86
Hemoglobin	g/dL	223	14.44 ± 1.56	9.8 – 18.2		73	15.74 ± 2.02	10.7 – 19.0
Hematocrit	%	234	44.15 ± 6.55	20.80 – 59.30		75	46.78 ± 8.68	2.88 – 57.94
Mean corpuscular volume	fL	230	66.67 ± 3.24	59.0 – 74.6		67	66.13 ± 3.57	59.0 – 74.0
Mean corpuscular hemoglobin	pg	236	21.31 ± 1.21	16.7 – 24.4		75	21.99 ± 2.47	17.1 – 29.7
Mean corpuscular hemoglobin concentration	g/dL	236	31.98 ± 0.75	29.9 – 33.9		65	32.41 ± 1.01	30.7 – 34.5
Red cell distribution width variation	%	226	14.86 ± 1.34	12.8 – 19.1		74	16.32 ± 1.30	14.1 – 19.6
Red cell distribution width variation	%	-	-	-		71	47.72 ± 3.57	40.6 – 55.5
Platelets	(10^3/uL)	225	488.09 ± 137.75	198.0 – 842.0		70	558.24 ± 126.92	311.0 – 780.0
Mean platelet volume	fL	227	8.41 ± 0.74	6.8 – 10.3		69	9.09 ± 0.66	7.8 – 10.2
Plateletcrit	(10^3/uL)	-	-	-		73	0.52 ± 0.13	0.27 – 0.80
Platelet distribution standard deviation	fL	-	-	-		75	16.20 ± 2.36	12.0 – 20.4
Platelet distribution width	fL	-	-	-		75	39.92 ± 1.61	36.7 – 42.8

When comparing values between institutions, for clinical chemistry markers, the mean, standard deviation, and range are broken down by institution and listed in Table 3. The parameters listed here were significant main effects for facility: alanine transaminase [df = 1, 291, F = 12.072, P-value = <0.001] and phosphatase/creatinine ratio [df = 1, 285, F = 7.421, P-value = 0.006] were all highest in SNPRC individuals while total bilirubin [df = 1, 290, F = 237.797, P-value = <0.001], blood urea nitrogen [df = 1, 285, F = 10.427, P-value = 0.001], calcium [df = 1, 287, F = 40.972, P-value = <0.001], creatinine [df = 1, 289, F = 24.871 P-value = <0.001], globulin [df = 1, 283, F = 432.808, P-value = <0.001], and total protein [df = 1, 285, F = 143.178, P-value = <0.001] were all highest in UT Health individuals (Figure 3A - H, Supplementary Table 1). The pairwise comparison results with Bonferroni correction are represented in Figures 3A – H, with significant pairwise comparisons and their p-values represented in number of stars above the age categories being compared. Sodium [df = 1, 278, F = 230.138, P-value = <0.001] and potassium [df = 1, 268, F = 691.242, P-value = <0.001] had significant interaction with mixed outcomes (Supplementary Figure 2 & 3).

Figure 3.

Significant parameters from the chemistry ANOVAs for facility. The comparison of: A. alanine transaminase, B. blood urea nitrogen, C. calcium, D. creatinine, E. phosphatase/creatinine ratio, F. total protein, and G. globulin between SNPRC and UT Health. The number of symbols indicate the significance level, * = 0.05, ** = 0.01 and *** = <0.001.

For the complete blood count values, the mean, standard deviation, and range are broken down by institution and listed in Table 3. The parameters listed here were significant main effects for facility: lymphocytes [df = 1, 281, F = 5.465, P-value = 0.020] were highest in SNPRC individuals while white blood cells [df = 1, 279, F = 86.612, P-value = <0.001], red blood cells [df = 1, 278, F = 17.107, P-value = <0.001], mean corpuscular hemoglobin concentration [df = 1, 285, F = 14.967, P-value = <0.001], red blood cell width [df = 1, 284, F = 78.798, P-value = <0.001], platelets [df = 1, 279, F = 15.809, P-value = <0.001], and mean platelet volume [df = 1, 280, F = 60.127, P-value = <0.001] were all highest in UT Health individuals (Figure 4A - G, Supplementary Table 2). The pairwise comparison results with Bonferroni correction are represented in Figure 4A - G, with significant pairwise comparisons and their p-values represented in number of stars above the age categories being compared. Monocytes [df = 1, 279, F = 296.817, P-value = <0.001], lymphocyte percent [df = 1, 283, F = 93.760, P-value = <0.001], monocyte percent [df = 1, 271 F = 244.690, P-value = <0.001], hemoglobin [df = 1, 281, F = 45.628, P-value = <0.001], hematocrit [df = 1, 293, F = 14.495, P-value = <0.001], and mean corpuscular hemoglobin [df = 1, 295, F = 14.216, P-value = <0.001] had significant interaction effects with mixed outcomes (Supplementary Figure 2 & 3).

Figure 4.

Significant parameters from the complete blood count ANOVAs for facility. The comparison of: A. white blood cells, B. lymphocytes, C. red blood cells, D. mean corpuscular hemoglobin concentration, E. red cell distribution width, F. platelets, and G. mean platelet volume between SNPRC and UT Health. The number of symbols indicate the significance level, * = 0.05, ** = 0.01 and *** = <0.001.

Most of the differences between the SNPRC and UT Health are relatively small, only about 10 – 20% variance in the mean, and thus these differences may not be biologically significant. However, for parameters like monocytes and monocyte percent where the difference between SNPRC and UT Health is large, about a 10-fold difference, these results are more likely to be biologically significant, and may reflect husbandry differences. For example, all the UT Health marmosets consume irradiated food and receive reduced enrichment option.

Main effects – sex:

The mean, standard deviation, minimum, and maximum values are listed by sex for each parameter in Table 4. Alkaline phosphatase [df = 1, 287, F = 9.407, P-value = 0.002], blood urea nitrogen [df = 1, 285, F = 10.427, P-value = 0.001], sodium [df = 1, 278, F = 20.819, P-value = <0.001], chlorine [df = 1, 225, F = 5.132, P-value = 0.024], cholesterol [df = 1, 222, F = 9.522, P-value = 0.002], and bicarbonate [df = 1, 220, F = 4.883, P-value = 0.028] were significantly higher in males while glucose [df = 1, 285, F = 7.184, P-value = 0.007] was higher in females (Figure 5A - G, Supplementary Table 1).

Table 4.

The sample size, mean, standard deviation, minimum, and maximum by sex. Bold are values measured only at SNPRC, italics are values only measured at UT Health.

		Female				Male
Parameter	Units	n	Mean ± SD	Min. – Max.		n	Mean ± SD	Min. – Max.
Albumin	g/dL	141	3.76 ± 0.48	2.7 – 4.8		158	3.80 ± 0.50	2.7 – 4.8
Alkaline phosphatase	IU/L	142	79.68 ± 47.06	25.0 – 254.0		160	99.76 ± 52.43	5.0 – 267.0
Alanine transaminase	U/L	145	38.69 ± 50.34	4.0 – 271.0		161	38.00 ± 53.65	4.0 – 375.0
Amylase	U/L	29	84.65 ± 17.37	59.0 – 116.0		40	85.65 ± 13.07	60.0 – 122.0
Total Bilirubin	mg/dL	146	0.18 ± 0.09	0.1 – 0.4		159	0.20 ± 0.10	0.0 – 0.4
Blood Urea Nitrogen	mg/dL	141	18.87 ± 5.01	6.0 – 38.0		159	20.33 ± 5.34	8.0 – 37.0
Calcium	mg/dL	143	9.81 ± 0.65	8.0 – 11.5		159	9.90 ± 0.63	8.3 – 11.6
Phosphorus	mg/dL	141	3.69 ± 0.88	1.7 – 6.0		157	3.85 ± 0.82	1.7 – 5.9
Creatinine	mg/dL	144	0.32 ± 0.12	0.2 – 0.7		160	0.33 ± 0.13	0.1 – 0.7
Phosphorus Creatinine Ratio		141	12.29 ± 4.98	2.83 - 24.5		159	12.93 ± 5.41	2.63 -25.0
Glucose	mg/dL	145	175.41 ± 59.37	68.0 – 317.0		155	157.84 ± 60.66	29.0 – 316.0
Sodium	meq/L	144	149.21 ± 3.92	139.0 – 159.0		148	150.89 ± 4.17	141.0 – 161.0
Potassium	meq/L	137	3.35 ± 0.79	2.4 – 5.8		146	3.43 ± 0.84	2.1 – 5.9
Total Protein	g/dL	141	6.32 ± 0.74	4.7 – 8.0		159	6.44 ± 0.71	4.8 – 8.0
Globulin	g/dL	143	2.54 ± 0.53	1.6 – 4.1		155	2.58 ± 0.58	1.3 – 4.0
Creatine Phosphokinase	U/L	58	1023.948 ± 616.38	95.0 – 3216.0		63	987.69 ± 657.66	164.0 – 3316.0
Triglycerides	mg/dL	60	324.81 ± 272.27	36.0 – 1197.0		62	344.60 ± 282.34	50.0 – 1620.0
Lactate dehydrogenase	U/L	59	605.79 ± 577.68	136.0 – 2287.0		65	564.50 ± 474.89	135.0 – 2083.0
Gammaglutamyl transferase	U/L	59	10.18 ± 8.22	2.0 – 40.0		63	11.65 ± 11.49	3.0 – 64.0
Chloride	meq/L	113	105.00 ± 3.08	98.0 – 112.0		118	105.81 ± 2.94	99.0 – 112.0
Albumin/ Globulin Ratio		113	1.59 ± 0.29	0.9 – 2.3		114	1.65 ± 0.31	1.0 – 2.3
Cholesterol	mg/dL	116	138.64 ± 43.16	55.0 – 287.0		114	158.77 ± 52.56	64.0 – 340.0
Blood Urea Nitrogen/Creatinine Ratio		76	54.12 ± 17.48	15.0 – 93.3		79	55.88 ± 17.13	22.5 – 93.3
Bicarbonate	meq/L	114	24.44 ± 3.12	18.0 – 32.0		114	25.28 ± 3.31	17.0 - 32
ANION GAP	meq/L	111	21.26 ± 3.45	13.2 – 30.5		115	20.94 ± 3.56	14.1 – 30.2
Nucleated Red Blood Cells	%	112	0.79 ± 0.65	0.0 – 3.2		118	0.63 ± 0.50	0.1 – 2.4
White Blood Cells	(10^3/uL)	140	5.97 ± 2.47	1.7 – 13.91		155	6.46 ± 2.60	1.7 – 14.08
Lymphocytes	(10^3/uL)	142	3.09 ± 1.71	0.46 – 7.80		155	3.35 ± 1.75	0.08 – 8.32
Monocytes	(10^3/uL)	141	0.42 ± 0.86	0.0 – 4.14		154	0.47 ± 0.81	0.0 – 3.53
Neutrophils	(10^3/uL)	31	3.35 ± 1.63	1.21 – 7.38		42	3.55 ± 1.54	1.40 – 7.45
Eosinophil	(10^3/uL)	113	0.09 ± 0.09	0.0 – 0.4		112	0.09 ± 0.09	0.0 – 0.4
Basophils	(10^3/uL)	107	0.02 ± 0.04	0.0 – 0.1		113	0.01 ± 0.03	0.0 – 0.1
Lymphocytes percent	%	141	55.98 ± 20.42	11.4 – 91.0		158	57.35 ± 19.42	10.7 – 89.7
Monocytes percent	%	136	4.28 ± 6.72	0.0 – 37.5		151	5.12 ± 6.66	0.4 – 32.9
Neutrophils percent	%	30	36.54 ± 15.30	13.1 – 66.6		44	38.75 ± 12.73	13.2 – 70.5
Eosinophil percent	%	108	1.48 ± 1.25	0.0 – 7.0		111	1.67 ± 1.23	0.0 – 5.9
Basophils percent	%	109	0.63 ± 0.49	0.0 – 2.2		114	0.50 ± 0.38	0.0 – 1.9
Granulocyte percent	%	107	31.84 ± 13.15	9.0 – 63.3		118	31.08 ± 13.43	7.9 – 62.1
Red Blood Cells	(10^6/uL)	134	6.63 ± 0.82	4.67 – 8.29		159	7.09 ± 0.71	4.70 – 8.86
Hemoglobin	g/dL	136	14.41 ± 1.81	10.1 – 19.0		160	15.07 ± 1.69	9.8 – 19.0
Hematocrit	%	145	42.99 ± 8.22	2.88 – 57.94		164	46.37 ± 5.74	22.20 – 59.30
Mean corpuscular volume	fL	139	67.22 ± 3.21	61 – 74.6		158	65.95 ± 3.31	59 – 73.5
Mean corpuscular hemoglobin	pg	147	21.67 ± 1.52	16.7 – 29.7		164	21.30 ± 1.71	17.1 – 29.6
Mean corpuscular hemoglobin concentration	g/dL	143	32.08 ± 0.78	29.9 – 34.3		158	32.07 ± 0.87	30.0 – 34.5
Red cell distribution width variation	%	142	14.97 ± 1.31	12.8 – 18.4		158	15.45 ± 1.57	12.8 – 19.6
Red cell distribution width standard deviation	%	31	47.98 ± 3.03	42.2 – 55.5		40	47.52 ± 3.97	40.6 – 55.5
Platelet	(10^3/uL)	139	485.46 ± 146.09	198.0 – 842.0		156	521.91 ± 129.06	220.0 – 838.0
Mean platelet volume	fL	139	8.48 ± 0.80	6.8 – 10.2		157	8.64 ± 0.75	6.9 – 10.3
Plateletcrit	(10^3/uL)	31	0.51 ± 0.14	0.27 – 0.79		42	0.54 ± 0.12	0.35 – 0.80
Platelet distribution width	fL	32	39.60 ± 1.70	36.9 – 42.4		43	40.15 ± 1.51	36.7 – 42.8
Platelet distribution width	fL	32	15.81 ± 2.47	12.2 – 19.9		43	16.50 ± 2.25	12.0 – 20.4

Figure 5.

Significant parameters from the chemistry ANOVAs for sex. The comparison of: A. alkaline phosphatase, B. blood urea nitrogen, C. glucose, D. sodium, E. chlorine, F. cholesterol, and G. bicarbonate between females and males. The number of symbols indicate the significance level, * = 0.05, ** = 0.01 and *** = <0.001.

The mean, standard deviation, minimum, and maximum values are listed by sex for each parameter in Table 4. Nucleated red blood cells [df = 1, 222, F = 4.177, P-value = 0.042], basophil percent [df = 1, 215, F = 4.286, P-value = 0.039], were all higher in females while red blood cells [df = 1, 278, F = 28.741, P-value = <0.001], red cell distribution width [df = 1, 284, F = 8.215, P-value = 0.004], platelet [df = 1, 279, F = 8.906, P-value = 0.003], and platelet distribution width [df = 1, 67, F = 5.408, P-value = 0.023] were higher in males (Figure 6A - F, Supplementary Table 2). Hemoglobin [df = 1, 281, F = 15.151, P-value = <0.001], hematocrit [df = 1, 293, F = 24.033, P-value = <0.001], and mean corpuscular volume [df = 1, 281, F = 8.047, P-value = 0.004] had significant interaction effects (Supplementary Figure 2 & 3). All other parameters were not significantly different between males and females.

Figure 6.

Significant parameters from the complete blood count ANOVAs for sex. The comparison of: A. nucleated red blood cells, B. basophil percent, C. red blood cells, D. red cell distribution width, E. platelets, and F. platelet distribution width between females and males. The number of symbols indicate the significance level, * = 0.05, ** = 0.01 and *** = <0.001.

Comparison with previous report on aging blood chemistry values in aging marmosets

A previously published report outlined the median ranges for blood chemistry and CBC from the Massachusetts Institute of Technology (MIT) marmoset colony (Kramer et al., 2022). A comparison with the median values for SNPRC and UT Health colonies displayed in Table 5, demonstrate that for the most part the medians are consistent across datasets. However, the ranges of many parameters differ slightly between those outlined from MIT and those reported here. Note that the age groups of interest in this report differed slightly to evaluate any potential changes in the per-geriatric time frame, and the maximum age assessed in our geriatric cohort was higher than the MIT study. For a direct comparison with the MIT work, we divided our age groups up into young (0-2yrs.), adult (2-8 yrs.), and geriatric (8+yrs.) to match Kramer et al., (2022).

Table 5.

MIT comparison table, ranges and medians of young (0-2yrs.), adult (2-8 yrs.), and geriatric (8+yrs.). Medians are in parenthesis.

	MIT (F, M)			SNPRC			UT Health
	Young	Adult	Geriatric	Young	Adult	Geriatric	Young	Adult	Geriatric
Albumin	3.3 – 5.2; (4.4, 4.3)	3.1 – 5.1; (4.2, 4.1)	2.7 – 4.5; (3.6, 3.5)	2.9 – 4.8; (4.20)	2.7 – 4.7; (3.70)	2.7 – 4.4; (3.50)	2.7 – 4.7; (4.05)	2.8 – 4.5; (3.80)	2.7 – 4.2; (3.30)
Alkaline phosphatase	34.4 – 170.3; (88, 101)	13.1 – 121.9; (54, 64)	22.6 – 109.8; (52.5, 64.5)	43 – 267; (150)	25 – 231; (68)	41 – 223; (66)	52 – 254; (140)	5 – 110; (61)	34 – 168; (90)
Alanine transaminase	0.0 – 25.3; (7.5, 6)	0.0 – 21.6; (7, 6.5)	0.0 – 21.8; (8.5, 4.5)	4 – 375; (23.0)	4 – 319; (21.5)	6 – 91; (21.0)	8 – 220; (16.0)	5 – 72; (11.0)	7 – 29; (11.0)
Total Bilirubin	0.0 – 0.3; (0.1, 0.2)	0.0 – 0.2; (0.1, 0.1)	0.0 – 0.2; (0.1, 0.1)	0.1 – 0.4; (0.2)	0.0 – 0.4; (0.1)	0.1 – 0.2; (0.1)	0.3 – 0.4; (0.3)	0.2 – 0.4; (0.3)	0.2 – 0.4; (0.3)
Blood Urea Nitrogen	12.8 – 32.7; (22, 23)	9.7 – 31.1; (20, 22)	8.8 – 29.9; (19, 19.5)	6 – 35; (20)	8 – 38; (18)	10 – 36; (19)	13 – 34; (23)	12 – 37; (22)	14 – 28; (19)
Calcium	8.7 – 11.4; (10.2, 9.9)	8.5 – 11.2; (9.9, 9.7)	8.2 – 10.4; (9.3, 9.15)	9.1 – 11.3; (10.10)	8.1 – 11.2; (9.60)	8.0 – 11.1; (9.60)	9.0 – 11.6; (11.10)	8.8 – 10.9; (10.20)	8.9 – 10.8; (9.85)
Phosphorus	2.0 – 5.6; (4.1, 3.8)	1.8 – 5.4; (3.6, 3.2)	1.9 – 6.0; (3.6, 3.6)	2.0 – 5.8; (4.20)	1.7 – 5.7; (3.50)	2.3 – 6.0; (4.15)	1.7 – 5.6; (4.40)	1.7 – 5.6; (3.80)	2.5 – 5.0; (3.40)
Creatinine	0.1 – 0.4; (0.3, 0.3)	0.1 – 0.4; (0.2, 0.3)	0.0 – 0.4; (0.2, 0.2)	0.2 – 0.5; (0.20)	0.1 – 0.7; (0.30)	0.2 – 0.7; (0.40)	0.2 – 0.6; (0.40)	0.2 – 0.7; (0.40)	0.2 – 0.7; (0.55)
Glucose	31.0 – 254.0; (120, 132.5)	23.5 – 248.7; (125, 122.5)	5.8 – 247.1; (107, 109.5)	62 – 315; (183.5)	29 – 317; (164.0)	84 – 280; (143.5)	131 – 307; (201.0)	63 – 296; (154.5)	42 - 190; (135.0)
Sodium	145.7 – 155.0; (150, 150)	145.8 – 155.4; (150, 151)	146.0 – 155.5; (151, 151)	143 – 154; (149.0)	139 – 155; (148.5)	145 – 154; (150.0)	151 – 161; (156.5)	145 – 161; (157.0)	142 – 159; (156.5)
Potassium	2.2 – 4.0; (3, 3.1)	2.3 – 3.9; (3.2, 3.1)	2.3 – 3.9; (3.1, 3)	2.4 – 4.4; (3.0)	2.1 – 4.1; (3.1)	2.6 – 4.1; (3.1)	4.0 – 5.8; (5.0)	3.6 – 5.9; (5.2)	3.2 – 5.9; (4.9)
Total Protein	5.0 – 7.6; (6.5, 6.25)	5.0 – 7.5; (6.4, 6.2)	4.8 – 6.9; (5.9, 5.85)	5.0 – 8.0; (6.40)	4.7 – 7.5; (6.10)	5.0 – 7.3; (6.00)	5.7 – 7.9; (7.10)	5.9 – 7.9; (7.25)	6.0 – 8.0; (7.60)
Globulin	1.5 –2.6; (2.1, 2)	1.6 – 2.7; (2.1, 2.1)	1.7 – 2.9; (2.1, 2.35)	1.7 – 2.9; (2.25)	1.3 – 3.4; (2.40)	2.1 – 3.5; (2.50)	2.4 – 3.7; (3.20)	2.7 – 4.1; (3.40)	3.3 – 4.0; (3.75)
White Blood Cells	1.7 – 9.4; (4.8, 5.45)	0.8 – 8.3; (4.15, 4.55)	1.5 – 9.0; (5.8, 3.3)	1.70 – 13.30; (6.60)	1.7 0 – 12.80; (5.25)	2.10 – 10.20; (4.20)	5.58 – 14.08; (10.33)	4.87 – 13.91; (7.84)	4.10 – 14.07; (7.64)
Lymphocytes	0.3 – 5.3; (2.6, 2.85)	0.0 – 4.9; (2.2, 1.8)	0.0 – 5.5; (2.05, 1.1)	1.50 – 8.20; (4.5)	0.80 – 6.90; (3.1)	0.70 – 5.90; (2.1)	0.49 – 8.32; (4.59)	0.16 – 7.63; (1.80)	0.08 –6.14; (2.63)
Monocytes	0.0 – 0.3; (0.1, 0.2)	0.0 – 0.3; (0.2, 0.1)	0.0 – 0.3; (0.1, 0.1)	0.00 – 0.50; (0.10)	0.00 – 0.50; (0.10)	0.10 – 0.50; (0.10)	0.04 – 3.86; (0.17)	0.04 – 4.14; (1.73)	0.05 – 3.39; (0.93)
Lymphocytes percent	19.9 – 91.8; (57, 54.8)	14.4 – 81.8; (45.95, 46.45)	4.4 – 73.5; (31.75, 30.35)	38.3 – 91.0; (72.50)	17.3 – 87.8; (61.90)	21.0 – 70.9; (48.90)	11.6 – 84.4; (46.30)	10.7 – 69.6; (30.75)	21.0 – 70.9; (36.50)
Monocytes percent	0.0 – 7.6; (2.25, 3.5)	0.0 – 7.0; (2.7, 2.75)	0.0 – 5.2; (1.9, 1.95)	0.0 – 6.6; (2.30)	0.6 – 6.3; (2.60)	1.2 – 7.0; (3.10)	0.6 – 32.9; (1.30)	0.5 – 37.5; (14.90)	0.8 – 28.9; (7.75)
Red Blood Cells	156.7 – 643.0; (359, 395)	130.7 – 711.9; (404.5, 455)	236.7 – 876.3; (595.5, 514)	5.37 – 8.19; (7.14)	4.67 – 8.86; (6.82)	4.75 – 7.71; (6.44)	6.62 – 8.86; (7.65)	4.70 – 8.66; (7.32)	5.41 – 8.10; (6.86)
Hemoglobin	11.7 – 16.8; (13.8, 14.5)	10.9 – 16.1; (13.2, 14.15)	10.3 – 16.6; (13.2, 13.45)	10.5 – 17.2; (14.60)	9.8 – 18.2; (14.90)	9.8 – 16.6; (14.30)	13.6 – 18.1; (16.60)	11.0 – 19.0; (16.40)	10.7 – 17.8; (13.65)
Hematocrit	32.4 – 47.7; (39.45, 41)	31.3 – 47.3; (37.5, 40.25)	29.6 – 47.8; (38.65, 38.65)	22.20 – 54.20; (45.40)	22.70 – 59.30; (45.80)	20.80 – 51.50; (43.05)	42.09 – 56.71; (51.56)	2.88 – 57.94; (50.00)	13.43 – 53.97; (42.61)
Mean corpuscular volume	49.5 – 61.9; (55.5, 56.2)	49.8 – 63.3; (55.5, 56.8)	50.6 – 66.2; (58.5, 53)	59.0 – 72.4; (64.70)	59.1 – 74.6; (66.95)	61.6 – 73.0; (67.70)	61.0 – 71.0; (64.00)	60.0 – 74.0; (67.50)	59.0 – 72.0; (65.00)
Mean corpuscular hemoglobin	16.1 – 23.0; (19.35, 19.9)	16.8 – 23.0; (19.9, 19.9)	17.6 – 23.2; (20.75, 19.15)	16.7 – 24.3; (20.5)	18.3 – 24.4; (21.4)	19.2 – 23.3; (21.8)	18.1 – 23.9; (21.3)	18.8 – 29.7; (22.9)	17.1 – 25.3; (20.3)
Mean corpuscular hemoglobin concentration	31.0 – 39.5; (35.05, 35.35)	30.8 – 39.6; (35.7, 35.4)	29.2 – 39.9; (34.55, 35.9)	30.0 – 33.8; (32.0)	29.9 – 33.9; (32.1)	31.1 – 33.4; (31.9)	31.2 – 34.2; (32.7)	31.0 – 34.5; (32.5)	30.7 – 34.3; (31.5)
Red cell distribution width variation	5.9 – 8.3; (7.005, 7.37)	5.3 – 8.1; (6.72, 7.13)	5.1 – 8.2; (6.725, 7.305)	13.0 – 18.4; (15.55)	12.8 – 19.1; (14.30)	12.9 – 16.6; (14.70)	15.7 – 19.6; (16.90)	14.5 – 19.6; (15.80)	14.1 – 19.1; (15.65)
Platelet	39.0 – 57.5; (45.4, 49)	38.6 – 54.0; (45, 47)	36.9 – 50.4; (43, 44)	220 – 842; (394.0)	198 – 838; (515.5)	309 – 733; (466.0)	311 – 670; (473.0)	317 – 765; (607.5)	342 – 780; (599.5)
Mean platelet volume	7.4 – 14.2; (9.8, 10.6)	7.2 – 13.2; (10.5, 9.75)	6.2 – 12.9; (9.5, 8.75)	7.3 – 10.3; (8.90)	7.0 – 10.0; (8.30)	6.8 – 9.9; (7.85)	7.8 – 10.2; (9.40)	7.9 – 10.1; (9.30)	8.0 – 10.2; (8.70)

Discussion:

In this study we aimed to compare blood chemistry and CBC biomarkers across two colonies that use similar marmoset husbandry and had a similar pedigree of origin to evaluate reference values for adult animals and to assess whether the reference ranges for geriatric individuals need to be shifted. Our results add to the growing body of evidence that these parameters change with normal healthy aging (Kramer et al., 2022) and suggest that samples from geriatric animals might fall out of the normal reference range even in healthy individuals. We included only animals that were deemed healthy during a routine physical exam by the veterinarian, therefore to the best of our ability, we included only healthy individuals of all ages. Ill or unhealthy animals were to the best of our ability screened out, through the elimination of outliers, and through exclusion if the animal was flagged for a follow-up exam. Therefore, we are defining these reference ranges for healthy individuals in the presence of chronic healthy aging.

Healthy aging marmosets display alterations in biomarkers associated with changes in liver function, pancreas function, kidney function, cancer and heart disease risk, infection risk and anemia. Specifically, decreases in albumin and globulin are associated with impaired liver function, while increases in creatinine are associated with changes in kidney function (Gounden et al., 2023; Lala et al., 2023). Reductions in phosphorous/creatinine ratio are associated with chronic kidney disease, while reductions in albumin/globulin ratios are associated with impaired liver function (Gounden et al., 2023; Lala et al., 2023). Here, the parameters that were reduced in peri-geriatric or geriatric marmosets compared to younger marmosets were albumin, phosphorus/creatinine ratio, albumin/globulin ratio, calcium, glucose, lymphocytes, red cell distribution width, red cell distribution standard deviation, white blood cells, platelet distribution width, and platelet distribution standard deviation. The parameters that were increased in peri-geriatric and geriatric compared to young were amylase, creatinine, granulocytes percent, and platelets.

Lower albumin levels have previously been found to be negatively correlated with age (Kramer et al., 2022; Ross et al., 2012). This study demonstrates this same pattern for albumin. Lower serum albumin has also been associated with marmosets exhibiting symptoms of bone and gastrointestinal disease. Previously, a recommendation was made that a value of less than 3.5 g/dL serum albumin identifies diseased animals, aged 0 – 16 years old (Baxter et al., 2013). However, our study found that seemingly healthy geriatric marmosets’ mean and median were both about 3.4 g/dL for serum albumin suggesting similarly to Kramer et al., (2022) that the cutoff point of 3.5 g/dL may be a valid cutoff point for young animals but does not necessarily designate disease in aged animals.

Interestingly, in the MIT colony they found that calcium was lower in geriatric animals similar to what we found in our results (Kramer et al., 2022). In the SNPRC and UT Health animals, we also found a significant difference based on facility and age which may indicate differences due to environment, UV exposure, or intake of vitamin D3. The nutrition that the animals receive is the same between facilities, but differences in water sources (e.g. SNPRC on well water and UT Health on city water) and the hard water in San Antonio may be associated with differences in calcium uptake (Layne & Power, 2003; Ross et al., 2017).

We found a significant difference in amylase for the UT Health animals based on age with peri-geriatric individuals displaying the highest amylase values. In the MIT colony this was found to be significantly different based on post-partum status, however none of the UT Health animals included in this assessment were post-partum (Kramer et al., 2022). This difference has not been previously associated with age in marmosets, but reduction in amylase concentrations in Tibetan macaques is associated with age, with young animals having higher amylase concentrations than old animals (Wu et al., 2014), and seems to reflect pancreatic function across age groups.

We did find significant differences, for age (geriatric animals had the highest values) and facility (UT Health had higher values than SNPRC), in creatinine concentration. Comparatively the MIT animals showed differences in creatinine between juvenile and adult females (Kramer et al., 2022). In our dataset geriatric marmosets had the highest levels of creatinine, indicating that this may be a colony phenomenon. Interestingly, UT Health had higher levels of creatinine than did SNPRC. SNPRC has previously reported high levels of chronic kidney disease especially as related to mortality, and the significant differences we see in the SNPRC and UT Health data may be a reflection of this as creatinine is primarily a kidney function marker (Ross et al., 2012). The significant differences we see between age categories may be indicative of sub-clinical kidney disease, but because all animals included in the study were considered healthy these differences may also be due to the normal changes associated with age. We eliminated any animals from the analysis that were getting husbandry support for kidney disease e.g., fluids, or had any clinical flags for follow-up related to kidney function, therefore any animals that were included in the analysis were to the best of our knowledge healthy aging animals. Chronic kidney disease is one of the leading causes of death in the SNPRC colony, and the difference in creatinine between adult and geriatric individuals may be the result of the loss of kidney function associated with diseases, and suggest that the marmoset model of aging may be particularly important as a model for kidney disease (Lee et al., 2019; Ross et al., 2019).

Phosphorus was not significantly different based on age for the SNPRC and UT Health animals. In the contrast for the MIT animals phosphorous was found to be different between male juveniles, adults, and geriatric animals, while for females only juvenile and adults were different. We did however find a significant difference in phosphorous/creatinine ratios across age groups and may again be due to the relationship between creatinine and chronic kidney disease seen in the SNPRC and UT Health colonies (Ross et al., 2012).

For the complete blood count parameters, the platelet distribution width decrease with age seen in this study could be related to increased cancer risk (Cheng et al., 2017). There is a relationship between colon cancer and decreases in platelet distribution width in humans, such that platelet distribution width values have been used to predict colon cancer (Cheng et al., 2017; Sakin et al., 2020) Several of these parameters are related to infection risk including lymphocyte percent which have been found to decrease with age in cynomolgus macaques (Macaca fascicularis) (Li et al., 2022). Lymphocyte percent decreased with old age in the marmosets in this study and may indicate a decrease in healthy aged animals ability to fight infection (Teissier et al., 2022). Increasing granulocyte percent changes may be related to infection, cognitive disease, reactive oxygen species, and reactive nitrogen species (Horvath & Ritz, 2015; Martins Chaves et al., 2000). Increases in hematocrit, have been found to be associated with immune system function in aging primates (Karal-ogly et al., 2023). Hematocrit, red blood cells, and hemoglobin are all related to immune system function and the ability to fight infection (Billett, 1990; Peter Klinken, 2002). Many of these immune system and complete blood count parameters changes with age are expected and assumed to be normal.

There were no significantly different parameters between peri-geriatric and geriatric marmosets. Given this we would suggest that a reference range for aged marmosets (≥8 years) be adjusted to those highlighted in gray in Table 5. These parameters which have been identified to have either significant decreases or increases in range value from young adults, should be redefined by new reference values that we suggest represent changes associated with normal healthy aging. We echo the conclusion of Kramer et al. (2022) that internal reference ranges for individual colonies are important to determine overall health. We also suggest that perhaps those values used to determine disease, or that are used to indicate poor health in young marmosets, may not be applicable in aging marmosets. We pulled these data from routine health exams for colony marmosets and excluded any animals that were flagged for ill health; thus these values represent “healthy normal” animals for their age group. Therefore, we conclude that there are age-related changes in health that are associated with many serum chemistry and CBC parameters measured here. We suggest that in addition to colony reference values there should also be a normal range shift with age as seen in human clinical assessments, and that “normal” values in geriatric marmosets do not fit exactly within the normal ranges for younger animals. This is particularly valuable with the growing movement developing tools that can clarify and identify the effects of age, or interventions to aging, on physiological outcomes in marmosets (Fernandez et al., 2019; Horvath et al., 2021; Ross et al., 2015; Sills et al., 2019)

In conclusion, we found that CBC and chemistry values vary with age, facility, and sex in marmosets. This result is like other previously published data on marmoset serum chemistry and complete blood count values. We suggest that the development of reference values within a colony, based on sex, and age is necessary. The development of these reference values provides a diagnostic and treatment tool that can aid in the care and management of aging marmoset colonies.

Supplementary Material

Fig S1

Supplementary Figure 1. Survival curves for Southwest National Primate Research Center and University of Texas Health San Antonio marmoset colonies. The survival is metric is in days post-physical exam and blood sampling. The Wald’s test result for Cox-Hazards is not significant (Wald’s = 1.22, df = 1, P-value = 0.30)

Fig S2

Supplementary Figure 2. Three-way interaction plots for A. potassium where the main interaction was between age and sex, B. hematocrit for age and sex, and C. hematocrit for age category and facility.

Fig S3

Supplementary Figure 3. Two-way interaction plots for A. alkaline phosphatase, B. sodium, C. total protein, D. monocytes, E. lymphocyte percent, F. monocyte percent, G. hemoglobin, H. mean corpuscular volume, and J. mean corpuscular hemoglobin for age category and site. One factor I. hematocrit has a significant interaction between age category and sex.

Research Highlights:

• The parameters that were reduced in peri-geriatric or geriatric marmosets compared to younger marmosets were albumin, phosphorus/creatinine ratio, albumin/globulin ratio, calcium, glucose, lymphocytes, red cell distribution width, red cell distribution standard deviation, white blood cells, platelet distribution width, and platelet distribution standard deviation.

• The parameters that were increased in peri-geriatric and geriatric compared to young marmosets were amylase, creatinine, granulocytes percent, and platelets.

• Our results suggest normal changes associated with healthy aging marmosets indicating the potential need for refinement of normal ranges for clinical biomarkers in aged marmosets.

Abbreviations:

CBC

complete blood count

SNPRC

Southwest National Primate Research Center

UT Health

University of Texas Health San Antonio

ANOVA

Analysis of Variance

MIT

Massachusetts Institute of Technology