A randomized, double-blind, placebo-controlled clinical study to evaluate the efficacy of the synbiotic medical food, SBD111, for the clinical dietary management of bone loss in menopausal women

Abstract

Summary

This 12-month study in 286 early postmenopausal women evaluated the efficacy and safety of SBD111, a synbiotic medical food, in reducing bone loss. SBD111 did not significantly reduce bone loss for the full cohort, but did produce evidence of reduced bone loss in women with osteopenia and BMI ≥ 30.

Purpose

To determine the efficacy of SBD111, a synbiotic medical food comprising probiotics and prebiotics, in reducing bone loss in women post-menopause, including prespecified subpopulations of women with osteopenia or elevated BMI.

Methods

In this prospective, multicenter, double-blind, randomized, placebo-controlled clinical food trial (NCT05009875), 286 healthy, non-osteoporotic women between 1 and 6 years post-menopause were enrolled and consumed SBD111 (4.75 × 10¹⁰ colony forming units) or placebo (maltodextrin) capsules twice daily for 12 months. The primary endpoint was change in areal BMD at the lumbar spine (LS). Secondary endpoints included change in areal BMD at the femoral neck (FN) and total hip (TH), trabecular volumetric BMD at the LS, markers of bone turnover and inflammation, and safety. Changes in gut microbiome composition were exploratory. The hypotheses being tested were formulated before data collection.

Results

Two hundred eighty-six women [age 55 ± 3 years (mean ± standard deviation)] were enrolled, with 221 (77%) completing the study. For the primary outcome, SBD111 administration was not associated with significantly less bone loss in the LS after 12 months [0.15% (− 0.52%, 0.82%), mean effect size (95% CI) by linear mixed-effects regression]. However, SBD111 was associated with reduced BMD loss in the TH for women with BMI ≥ 30 [0.97% (0.015%, 1.925%)] and modestly reduced BMD loss in the FN for women with osteopenia [0.89% (− 0.277%, 2.051%)].

Conclusions

These findings indicate SBD111 did not significantly reduce BMD loss for the full cohort. However, the trial produced evidence that SBD111 reduced bone loss in women with osteopenia and BMI ≥ 30.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00198-025-07650-7.

Introduction

Decreased estrogen production during and after menopause leads to rapid loss of bone mineral density (BMD) throughout the body, rendering women susceptible to osteopenia, osteoporosis, and associated fractures [1]. Osteoporosis and osteopenia affect over 50% of women over age 50, and women with low BMD have a high risk of hip and vertebral fractures, highlighting the need to manage bone loss during and after menopause [2].

Treatments are available to maintain or increase BMD in women, including antiresorptive and osteoanabolic drugs [3], but these therapies are indicated only for patients diagnosed with osteoporosis or at high risk of fracture [4]. Hormone therapy (HT) can be used to manage perimenopausal bone loss, but many women avoid this due to safety concerns [5, 6]. Thus, reduction of bone loss is the objective for maintaining skeletal health at the time of menopause, and there is a need to develop new strategies with wide patient acceptance.

Estrogen loss during menopause has effects on bone-resorbing osteoclasts mediated by immune signals [7, 8], including tumor necrosis factor alpha (TNF-α), interleukin (IL)−1 Beta (IL-1β), and receptor activator of nuclear factor kB ligand (RANKL) [9]. Blockade of TNF-α or IL-1β has been shown to decrease bone resorption in postmenopausal women [10], and RANKL inhibition with denosumab is commonly used for decreasing bone resorption in patients with osteoporosis [11]. These interventions underscore the importance of systemic inflammation as a driver of bone loss in menopause.

Immunity and inflammation are strongly influenced by the gut microbiota, the diverse community of microbes inhabiting the gastrointestinal tract [12–14]. Probiotic microorganisms can regulate bone remodeling and protect against bone loss through several mechanisms, including the production of anti-inflammatory short-chain fatty acids (SCFA) and enhanced gut barrier function [15–17]. Together, these observations highlight an opportunity to identify novel microbial solutions to reduce inflammatory signalling and manage bone loss after menopause.

We previously described SBD111, a synbiotic medical food containing prebiotic dietary fibers and microbial strains isolated from fresh fruits and vegetables, for the dietary management of postmenopausal bone loss [18]. The construction of SBD111 was driven by computational prediction of functions shown to modulate bone remodeling (e.g., SCFA and vitamin K2 production). SBD111’s component strains act synergistically; the combination produces a greater amount of anti-inflammatory short-chain fatty acids than the additive sum produced by its individual components [18]. Furthermore, we demonstrated that oral administration of SBD111 significantly reduced bone loss in a mouse model of postmenopausal osteoporosis and decreased the expression of TNF-α and IL-6 in vertebrae compared to vehicle controls [18]. We also demonstrated that SBD111 is safe and well tolerated in a randomized, placebo-controlled safety and tolerability clinical food trial of 32 healthy adults [19].

Here, we present the results of a clinical food trial designed to test the hypothesis that oral administration of SBD111 will reduce BMD loss in otherwise healthy postmenopausal women over a 12-month period.

Methods

Trial design

The study was a multicenter, double-blind, 1:1 randomized, placebo-controlled clinical food trial that enrolled healthy women 1–6 years after menopause. Eligible participants were randomized to receive either an oral medical food, SBD111, or placebo (oral maltodextrin capsules) twice daily for 12 months. Participants’ progression through the trial is presented in Fig. 1 (CONSORT diagram). The visit and measurement schedule is described in Table SI1 in compliance with SPIRIT guidelines. The study was conducted in compliance with applicable Australian laws and regulations after Human Research Ethics Committee (HREC) and approval from the National Institute of Integrative Medicine (NIIM). Participants were recruited to the study after all necessary approvals had been obtained.

Fig. 1.

CONSORT diagram depicting the flow of participant screening, enrolment, allocation, follow-up, and analysis throughout the reported clinical study

Participant recruitment and allocation

Two hundred eighty-six women between 1 and 6 years post-menopause were enrolled across two sites in Brisbane and Sydney, Australia, between November 2021 and July 2023. Inclusion criteria included written informed consent, at least 6 months since HT intake, Dual Energy X-Ray Absorptiometry (DXA)-derived T-score >  − 2.5 at the lumbar spine (L1–L4), femoral neck, and total hip, and BMI between 18.5 and 35 kg/m². Key exclusion criteria included a history of bone disorders, hip replacement, partial hysterectomy, and treatment with osteoporosis drugs (see Supplemental Information for all inclusion and exclusion criteria). Eligible participants were randomized to either SBD111 medical food or placebo in a parallel assignment scheme. All participants and study site personnel were blinded to test article allocation. The investigators remained blinded with no access to the randomization code until the end of study, database lock, and statistical analyses were complete.

Investigational product

The SBD111 synbiotic medical food is composed of Levilactobacillus brevis SBS04254 (3 × 10¹⁰ Colony Forming Units (CFU)/day), Lactiplantibacillus plantarum SBS04260 (3 × 10¹⁰ CFU/day), Leuconostoc mesenteroides SBS02455 (3 × 10¹⁰ CFU/day), Pichia kudriavzevii SBS04263 (5 × 10⁹ CFU/day), and prebiotics (oligofructose (640 mg/day) and dried blueberry powder (640 mg/day)). The total probiotic concentration was 9.5 × 10¹⁰ CFU/day, encased in acid-resistant, delayed-release, size zero, white, opaque capsules (4 capsules/day; Table SI2). The placebo was composed of maltodextrin (1720 mg in 4 capsules/day) in identical capsules. Vitamin D (500 IU, NanoCelle D3, Bioglan Medlab) was provided adjunctively to all participants for the duration of the trial.

Study outcomes

The primary outcome was the percent change in BMD at the lumbar spine (2–4 evaluable levels L1-L4) after 12 months of administration, attributable to intervention compared with placebo. Other measured outcomes included change in: BMD at the femur and total hip at 6 and 12 months, lumbar spine (L1 & L2) trabecular volumetric BMD (vBMD) and total vBMD at 12 months, serum c-terminal telopeptide (CTX) and procollagen 1 intact N-terminal propeptide (P1NP) at 6 and 12 months, serum inflammatory markers including C-reactive protein (CRP), interleukin-17 (IL-17), tumor necrosis factor alpha (TNFα), interleukin 1 beta (IL-1β), interleukin 4 (IL-4), receptor activator of nuclear factor-κB ligand (RANKL), and interferon gamma (IFNγ) at 6 and 12 months, gut microbiome composition and functional potential at 6 and 12 months, and safety and gastrointestinal tolerability (see Supplemental Information).

Prespecified analytic subgroups

In addition to the primary and secondary efficacy outcome measures, we prespecified several subgroup analyses: (i) duration of time since start the of menopause (1 to 3 years vs. 4 to 6 years), (ii) obesity measures at baseline visit (BMI ≥ 30 kg/m² vs. BMI < 30 kg/m²), and (iii) presence (T-score ≤ 1.0 and >  − 2.5) vs. absence (T-score >  − 1.0) of osteopenia in the lumbar spine at baseline.

Blood samples

Overnight fasted blood samples were collected at enrollment, baseline, 6, and 12 months from all participants. Full blood counts were performed, and serum samples were stored at −80 °C before analysis. Serum CTX, P1NP, CRP, IL-17, TNFα, IL-1β, IL-4, RANKL, and IFNγ were analyzed by Cardinal Bioresearch Pty. Ltd. Brisbane, Australia (see Supplemental Information).

Radiology assessment

BMD of the lumbar spine (L1–L4), total hip, and femoral neck of the left femur and body composition were measured by DXA (Lunar Prodigy, GE Medical Systems, or Hologic Horizon, Hologic, Inc) at baseline, 6, and 12 months. vBMD and lumbar spine architecture were acquired by quantitative computed tomography (QCT) of the lumbar spine (L1 & L2) at baseline and 12 months using whole-body CT scanners (Somatom Definition Edge, Siemens, Inc, Incisive, Philips Healthcare; or Lightspeed VCT, GE Medical Systems) via a custom scanning protocol as previously described [20]. DXA and QCT data were uploaded to a central imaging analysis center (Clario, USA) for image processing, quality control, and analysis by blinded technicians. QCT data were analyzed using Medical Imaging Analysis Framework (MIAF) Spine software (see Supplemental Information) [21].

Statistical analysis

The study was estimated to have 80% power to detect a difference of 1% in BMD over 12 months with an assumed standard deviation of 2.9% and an estimated evaluable sample of 133 participants per study arm. Anticipating a 10% drop out rate, investigators planned to enroll 300 participants (150 per arm). Efficacy analyses were conducted using the intention-to-treat principle and all available data on randomized participants was included. The primary analysis used mixed-effects linear regression in recognition of the repeated-measures design. The target estimate was the population-level 1-year difference in change in BMD attributable to the application of the intervention relative to control, as estimated by a regression contrast at 12 months. Parallel approaches were used for other endpoints. Mixed-effects linear regression models using 6- and 12-month data were supplemented with Students t-tests of the null hypotheses against one-sided alternatives favoring SBD111, motivated by published and unpublished preclinical evidence of consistently reduced BMD loss in animals receiving oral SBD111 across multiple studies [18]. Analyses of serum biomarkers of bone turnover and inflammatory cytokines utilized two-sided Student’s t-tests between groups, as preclinical evidence for these markers was more limited. Adverse and serious adverse events totals were computed for each study arm, as were the proportion of participants experiencing one or more event. See Supplemental Information for a comprehensive description of statistical analyses utilized for stool metagenome analysis.

Results

Participant recruitment and randomization

Two hundred eighty-six women age 55 ± 3 years, height 1.7 ± 0.1 m, and weight 70 ± 12 kg (mean ± standard deviation) were enrolled and randomized to receive SBD111 (n = 144) or placebo (n = 142) capsules for 12 months, in addition to daily oral Vitamin D₃ (500 IU/day) for the duration of the trial. At randomization, study groups were well balanced across age, height, weight, BMI, blood pressure, and BMD (Table 1). Two hundred twenty-one participants (77.3%) completed the study and represent the safety and efficacy population for statistical analysis (Fig. 1).

Table 1.

Demographics and participant baseline characteristics. Values represent mean (standard deviation)^a

		All participants		Participants with BMI ≥ 30 at baseline		Participants with osteopenia at baseline
Group		SBD111	Placebo	SBD111	Placebo	SBD111	Placebo
Enrolled		144	142	30	20	52	58
Completed		115	106	26	16	43	40
Withdrew		17	20	3	3	6	11
Withdrew with AE		10	11	1	1	2	5
Lost to follow-up		2	5	0	0	1	2
Age, years		55 (3.58)	55 (3.42)	56 (2.03)	56 (2.95)	55 (3.58)	56 (2.91)
Height, m		1.66 (0.06)	1.65 (0.07)	1.66 (0.06)	1.65 (0.07)	1.66 (0.06)	1.66 (0.07)
Weight, kg		71 (13)	69 (12)	89 (7)	89 (9)	68 (13)	68 (12)
BMI, kg/m2		26 (4.33)	26 (4.02)	32 (1.85)	33 (1.75)	25 (4.83)	25 (4.08)
Systolic BP, mmHG		124 (15)	119 (15)	132 (16)	127 (14)	120 (15)	119 (16)
Diastolic BP, mmHG		83 (9.6)	80 (10)	87 (7)	87 (12)	82 (10)	79 (10)
Baseline BMD, g/cm2	Lumbar spine	1.08 (0.17)	1.05 (0.14)	1.12 (0.96)	1.07 (0.14)	0.92 (0.08)	0.93 (0.08)
	Femoral neck	0.84 (0.13)	0.85 (0.13)	0.87 (0.12)	0.87 (0.15)	0.77 (0.11)	0.76 (0.09)
	Total hip	0.92 (0.11)	0.92 (0.11)	0.96 (0.10)	0.97 (0.12)	0.86 (0.10)	0.83 (0.07)

^aOsteopenia at baseline defined as a DXA-derived T-score <  − 1.0 in the lumbar spine

BMD loss and bone turnover markers for the safety and efficacy population

In the safety and efficacy population (n = 221), participants exhibited 91.23 ± 8.19% adherence (mean ± standard deviation), measured by number of capsules remaining at study completion. DXA revealed that BMD decreased at all three skeletal sites examined at 12 months compared to baseline (0.70–1.01% mean BMD decrease per site at 12 months across groups; Table 2,Table SI3). For the primary outcome of BMD change at the lumbar spine at 12 months, decline was lower in women receiving SBD111 than those receiving placebo [0.154% (− 0.516%, 0.824%) mean difference (95% CI)], but this did not achieve statistical significance (Fig. 2a, Table 3). Consistently, no significant differences were observed by QCT of the L1–L2 vertebrae for total vBMD, cortical vBMD, subcortical vBMD, or trabecular vBMD between study groups for the safety and efficacy population (Table SI4). In addition, no significant differences in change from baseline in serum bone turnover markers CTX and P1NP were observed between groups for this cohort (Table SI5).

Table 2.

Change in BMD measured by DXA for all participants by study group and visit^a

	Month 12		Month 6
	SBD111 (N = 115)	Placebo (N = 105)	SBD111 (N = 122)	Placebo (N = 115)
Lumbar spine
BMD, % change	− 0.70 (2.76)	− 0.76 (2.76)	− 0.55 (2.41)	− 0.59 (2.40)
T-score, change	− 0.070 (0.25)	− 0.069 (0.25)	− 0.058 (0.22)	− 0.053 (0.21)
Z-score, change	0.0057 (0.25)	0.0052 (0.25)	− 0.018 (0.22)	− 0.014 (0.21)
Femoral neck
BMD, % change	− 0.74 (2.74)	− 0.95 (2.71)	− 0.42 (2.60)	− 0.79 (2.45)
T-score, change	− 0.052 (0.17)	− 0.060 (0.18)	− 0.032 (0.17)	− 0.049 (0.15)
Z-score, change	− 0.0061 (0.17)	− 0.013 (0.18)	− 0.0073 (0.16)	− 0.024 (0.15)
Total hip
BMD, % change	− 0.89 (2.07)	− 1.01 (1.89)	− 0.58 (1.54)	− 0.049 (0.15)
T-score, change	− 0.052 (0.17)	− 0.060 (0.18)	− 0.032 (0.17)	− 0.024 (0.15)
Z-score, change	− 0.00061 (0.17)	− 0.013 (0.18)	− 0.0073 (0.16)	− 0.60 (1.61)

^aValues represent change from baseline [mean (standard deviation)]

Fig. 2.

BMD change from baseline for participants with osteopenia, elevated BMI, or elevated adiposity. a–c BMD change from baseline to 12 months represented as percent of baseline BMD at the femoral neck, total hip, and lumbar spine for a all participants, b participants with osteopenia, and c participants with BMI ≥ 30 that received either twice daily SBD111 or maltodextrin placebo. d Scatter plot showing the relationship between BMI (kg/m²) and body fat % at baseline for participants enrolled in the clinical study. Points represent individual participants, shaded by intervention group (SBD111 or placebo). e BMD change from baseline to 12 months represented as percent of baseline BMD at the total hip for participants that received either twice daily SBD111 or maltodextrin placebo stratified by increasing levels of body fat percentage at baseline. f BMD change from baseline to 12 months represented as percent of baseline BMD at the femoral neck, total hip, and lumbar spine for participants with > 40% body fat at baseline that received either twice daily SBD111 or maltodextrin placebo. g Change in serum CTX concentration from baseline to 12 months for participants with BMI  ≥ 30 that received either twice daily SBD111 or maltodextrin placebo. p-value represents a two-sided Student’s t-test. For a–c and e–g, values represent mean ± standard error. ^ap < 0.05 one-sided student’s t-test, ^bp = 0.05, linear mixed-effects model treatment effect at 12 months

Table 3.

Effect of SBD111 administration on BMD by visit for prespecified subgroups as determined by DXA^a

	N	Month 12 BMD, % change Mean (95% CI)	LME p-val	t-test p-val		N	Month 6 BMD, % change Mean (95% CI)	LME p-val	t-test p-val
Lumbar spine
All participants	220	0.154 (− 0.515, 0.824)	0.65	0.44		237	0.034 (− 0.621, 0.690)	0.92	0.46
BMI < 30	178	− 0.102 (− 0.852, 0.650)	0.79	0.71		194	0.049 (− 0.686, 0.784)	0.90	0.44
BMI ≥ 30	43	1.169 (− 0.298, 2.635)	0.13	0.07		43	0.076 (− 1.371, 1.523)	0.92	0.46
Osteopenic	83	0.715 (− 0.505, 1.939)	0.26	0.25		90	0.339 (− 0.859, 1.536)	0.58	0.28
Non-osteopenic	137	− 0.170 (− 0.949, 0.608)	0.67	0.66		147	− 0.143 (− 0.905, 0.619)	0.71	0.65
< 4 years in menopause	130	0.336 (− 0.511, 1.184)	0.44	0.28		142	0.152 (− 0.674, 0.979)	0.72	0.35
≥ 4 years in menopause	88	− 0.021 (− 1.089, 1.049)	0.97	0.60		95	− 0.115 (− 1.171, 0.941)	0.83	0.59
Femoral neck
All participants	220	0.163 (− 0.525, 0.851)	0.64	0.28		237	0.342 (− 0.329, 1.013)	0.32	0.13
BMI < 30	178	− 0.036 (− 0.810, 0.737)	0.93	0.48		194	0.084 (− 0.668, 0.836)	0.83	0.38
BMI ≥ 30	43	1.006 (− 0.509, 2.521)	0.20	0.10		43	1.388 (− 0.105, 2.881)	0.07	0.04*
Osteopenic	83	0.887 (− 0.274, 2.044)	0.14	0.044		90	0.295 (− 0.837, 1.426)	0.61	0.26
Non-osteopenic	137	− 0.275 (− 1.124, 0.575)	0.53	0.74		147	0.364 (− 0.463, 1.191)	0.39	0.18
< 4 years in menopause	130	0.087 (− 0.762, 0.935)	0.84	0.38		142	0.275 (− 0.548, 1.098)	0.51	0.25
≥ 4 years in menopause	88	0.272 (− 0.857, 1.399)	0.64	0.30		95	0.241 (− 0.867, 1.348)	0.67	0.28
Total hip
All participants	220	0.058 (− 0.410, 0.526)	0.81	0.34		237	0.019 (− 0.437, 0.474)	0.94	0.46
BMI < 30	178	− 0.164 (− 0.696, 0.368)	0.55	0.62		194	− 0.127 (− 0.643, 0.389)	0.63	0.70
BMI ≥ 30	43	0.970 (0.015, 1.925)	0.05	0.041		43	0.557 (− 0.385, 1.499)	0.25	0.11
Osteopenic	83	0.280 (− 0.482, 1.040)	0.47	0.22		90	− 0.190 (− 0.931, 0.550)	0.62	0.71
Non-osteopenic	137	− 0.045 (− 0.627, 0.537)	0.88	0.49		147	0.156 (− 0.411, 0.723)	0.59	0.28
< 4 years in menopause	130	0.294 (− 0.320, 0.907)	0.35	0.18		142	− 0.097 (− 0.693, 0.499)	0.75	0.64
≥ 4 years in menopause	88	− 0.339 (− 1.044, 0.363)	0.35	0.74		95	0.066 (− 0.627, 0.759)	0.85	0.41

^aMonth 12 and month 6 values represent effect size estimates and 95% confidence interval in percent change of BMD as determined by linear mixed-effects models. Positive values correspond to decreased bone loss in SBD111 recipients. LME p-values represent linear mixed-effects model-based treatment effect at 6 or 12 months. T-test p-values represent supporting analysis by one-sided Student’s t-test, pooled variance

BMD loss and markers of bone turnover and inflammation in prespecified subgroups

Although no significant effects were observed in the full safety and efficacy population, there was evidence of less bone loss in the SBD111 group by DXA in two populations: women with osteopenia at baseline and women with BMI ≥ 30 (Fig. 2b–c, Table 3). Among osteopenic women (n = 83), participants consuming SBD111 lost significantly less BMD in the femoral neck at 12 months than those consuming placebo [0.887% (− 0.277%, 2.051%) mean difference (95% CI) by linear mixed-effects regression; p = 0.044, one-sided student’s t-test] (Fig. 2b,Table 3). Similar nonsignificant trends were observed at the lumbar spine and total hip (Table 3).

Women with BMI ≥ 30 at baseline (n = 42) who were assigned to the SBD111 group also exhibited reduced BMD loss in the total hip at 12 months [0.970% (0.015%, 1.925%) mean difference (95% CI) by linear mixed-effects regression; p = 0.041, one-sided student’s t-test] (Fig. 2c,Table 3) compared to placebo. Similar nonsignificant trends were observed at the lumbar spine and femoral neck at 12 months (Table 3). Importantly, participant characteristics were comparable for women assigned to the SBD111 and placebo arms for these subgroup analyses (Table 1).

No significant differences in change of CTX or P1NP from baseline between SBD111 and placebo groups were observed for participants with osteopenia at baseline (Table SI5). However, SBD111 administration was associated with a significant decrease in serum CTX compared to placebo controls at 12 months in participants with BMI ≥ 30 (p = 0.049, two-sided Student’s t-test; Fig. 2g, Table SI5).

Obesity is associated with systemic inflammation, driven in part by endocrine and pro-inflammatory mediator production by adipose tissue, and we previously reported that administration of SBD111 reduced inflammatory biomarkers in mice [18, 22]. We aimed to determine whether SBD111 administration was associated with changes in serum inflammatory markers in women with BMI ≥ 30. However, no statistically significant differences were observed in serum concentrations of IFNγ, IL-17A, IL-1β, IL-4, IL-6, TNFα, RANKL, or CRP for participants with BMI ≥ 30 receiving SBD111 compared to placebo (p > 0.05, two-tailed student’s t-tests, Fig SI1).

No significant differences in BMD loss were observed in subgroups of women with BMI < 30, women without osteopenia at baseline, women less than four years after menopause, or those at least 4 years after menopause (Table 3). In addition, no significant differences were observed by QCT of the L1–L2 vertebrae for any metric between study groups for any prespecified study subgroup (Table SI4).

BMD loss in women with elevated body fat

The usefulness of BMI as a metric is debated because it fails to distinguish between lean mass and fat mass. Indeed, 87 women who completed this study (30.4%) exhibited ≥ 40% body fat despite having a BMI ≤ 30 (Fig. 2d). Furthermore, exploratory analysis revealed that administration of SBD111 was associated with reduced bone loss at the total hip compared to placebo for women with ≥ 40% body fat (n = 130), regardless of BMI (Fig. 2e). This effect was statistically significant at the total hip at 12 months (p = 0.035, one-sided student’s t-test), and similar nonsignificant trends were observed at the lumbar spine and femoral neck (Fig. 2f). However, no significant differences were observed in serum CTX or P1NP for women with ≥ 40% body fat (Table SI5).

Effects of SBD111 administration on the gut microbiota

To characterize the effects SBD111 on the structure and function of the gut microbiota, stool samples underwent metagenomic DNA sequencing. This analysis revealed that SBD111 administration was associated with significant enrichment of SBD111 strains and modestly increased richness of functional pathways in the gut metagenome at 12 months compared to placebo (Figs SI2-SI3; see Supplemental Information).

Previous clinical studies have reported decreased gut microbial diversity in obese participants [23]. Accordingly, we observed that women with  ≥ 30 exhibited significantly reduced species richness (p = 0.02, Welch’s t-test) and a trend of reduced Shannon diversity indices (p = 0.08) in their stool metagenomes at baseline compared to those with BMI < 30 (Fig SI4a-b). Moreover, women with BMI ≥ 30 displayed significantly reduced richness of Metacyc functional pathways, though no significant difference was observed for Shannon diversity of functional pathways (p = 0.02 and p = 0.83, respectively, Welch’s t-test; Fig SI4c-d). 29 bacterial species exhibited significantly lower abundance in women with BMI ≥ 30, including several representatives of genera shown to produce the anti-inflammatory SCFA, butyrate, including Clostridium, Butyrivibrio, and Anaerostipes (Table SI6). Similarly, women with ≥ 40% body fat exhibited lower species richness and Shannon indices (p = 0.0003 and p = 0.0017, Welch’s t-test, respectively; FigSI4e-f), Metacyc functional pathway richness (p = 0.016; Fig SI4g) and abundance of species associated with butyrate production (Table SI7) compared to women with body fat < 40%.

Gastrointestinal tolerability of SBD111

To evaluate gastrointestinal tolerability, participants completed 12-item gastrointestinal tolerability questionnaires (GITQ) on weeks 1–4, as well as at months 3, 6, 9, and 12. 30.3% of participants reported experiencing gastrointestinal (GI) symptoms in at least one survey, indicating that most participants experienced no GI symptoms. No significant differences were observed in the number of participants reporting symptoms or in the total number of GI symptoms reported per participants for mild or moderate symptoms between study groups (p > 0.05, Pearson’s chi² test; Table SI8). However, a significantly lower proportion of participants in the SBD111 group reported experiencing severe GI symptoms (p = 0.02, Pearson’s chi² test; Table SI8). Similarly, participants allocated to the SBD111 group reported a significantly lower number of severe GI symptoms compared to participants randomized to placebo (p = 0.02, Pearson’s chi² test; Table SI8).

Adverse events

One hundred twenty-six total adverse events (AEs) were reported, with 87 (69%) determined as possibly related to the interventions (SBD111 or placebo) (Table SI9). No significant differences were observed in number of AEs reported per participant between study groups or in the frequency of AEs possibly related to study intervention between study groups (p > 0.05, Pearson’s chi² test, respectively; Table SI9).

Discussion

Bone loss remains a significant challenge during and after menopause, and there are few safe and effective solutions to maintain bone mass during this period. In the present study, we observed no significant differences in the primary outcome of percent change in lumbar spine BMD between active and placebo groups after 12 months for the full safety and efficacy population. However, in our pre-specified subgroup analyses of women with osteopenia and of those with elevated BMI, SBD111 protected women against bone loss to a significantly greater extent than placebo. Exploratory analysis of women with elevated body fat composition similarly revealed evidence of decreased bone loss in women randomized to the SBD111 compared to placebo.

Managing bone loss in women with osteopenia is a key unmet need, as their fracture risk is elevated compared to individuals with healthy BMD, and management options are limited [2]. The present study suggests that women with osteopenia could manage bone loss by consuming a medical food synbiotic earlier than they would be eligible for pharmaceutical interventions, potentially reducing their lifetime fracture risk. However, fully evaluating the long-term benefits of the SBD111 medical food for women with osteopenia will require additional clinical investigation over several years of administration in large study cohorts.

The efficacy of SBD111 in obese women is supported by the multiple skeletal outcomes that we examined in this trial. First, administration of SBD111 in women with BMI ≥ 30 was consistently associated with reduced mean BMD loss at three sites: lumbar spine, total hip, and femoral neck. These BMD differences achieved marginal statistical significance only in the total hip, possibly due to the smaller number of participants in this group. Secondly, the observed attenuation of bone loss in women with BMI ≥ 30 kg/m² receiving SBD111 was accompanied by significantly reduced serum CTX, a marker of osteoclast activity and bone loss. Lastly, metagenome analysis revealed that women with BMI ≥ 30 harbored significantly reduced gut microbial diversity at baseline than those with BMI < 30, characterized by reduced abundance or absence of species known to produce anti-inflammatory SCFA and suggesting a more pro-inflammatory gut environment. This triangulation of evidence supports a true effect of SBD111 on skeletal health in the early menopausal period among women with obesity.

Approximately 43.3% of women over age 40 years in the United States have obesity and could benefit from new solutions for managing postmenopausal bone loss [24]. However, the relationship between elevated BMI and bone health is controversial [25]. Many studies have reported an association between obesity and increased BMD, and this relationship is often attributed to increased mechanical loading on the skeleton [22, 26]. Conversely, a study of over 6,000 individuals reported a negative correlation between body fat percentage and bone mass, and a recent meta-analysis found that abdominal obesity is not protective against fractures [27, 28]. These findings highlight the complexity of interactions between excess body weight and/or fat and bone health.

Obesity is associated with a state of chronic, systemic inflammation, thought to be driven by endocrine and pro-inflammatory mediators produced by excess adipose tissue [22]. This systemic inflammation is correlated with increased bone loss and fracture [8, 29]. Visceral adipose tissue, in particular, produces high levels of TNF-α and IL-6 and promotes lower trabecular bone volume, slower bone formation, and higher cortical bone porosity [30, 31]. Furthermore, obesity is associated with reduced gut microbial diversity that can impair gut-barrier integrity and potentiate local and systemic inflammation [32]. Taken together, these observations suggest a link between obesity-associated inflammation, the anti-inflammatory properties of SBD111, and evidence of reduced bone loss for postmenopausal women with obesity or those with > 40% body fat receiving the medical food. While no significant differences in serum concentrations of inflammatory cytokines were observed among obese participants in the present study, this finding may reflect differences between local and systemic concentrations of these inflammatory factors. For instance, RANKL is a key determinant of bone remodeling, but prior studies of the relationship between serum RANKL, tissue RANKL, and phenotypic outcomes suggest that the levels of these factors in the bone compartment may not be reflected at the systemic level [33], leaving open the possibility that local changes in the SBD111 group could underpin the bone effects reported here.

The composition of diet can have substantial impacts both on systemic inflammation and on bone health. Thus, diet is an important component to consider in relation to the efficacy of SBD111. In the present study, participants’ diets were not controlled, nor were dietary assessments performed during the study. Although our results suggest that SBD111 provided a benefit to early postmenopausal women with elevated BMI, osteopenia, or elevated body fat, we were not able to exclude the possibility of residual confounding by diet. It is plausible that dietary practices could modulate the efficacy of SBD111 synbiotic medical food. Indeed, nutraceuticals and functional foods, including dried plums and isoflavones like genistein, have shown promising results on bone in randomized, placebo-controlled clinical trials via distinct mechanisms. For example, consumption of dried plums, rich in several bioactive compounds with the potential to modulate bone turnover, and genistein, a natural selective estrogen receptor modulator found in soybeans and soy products have both shown benefit for the maintenance of BMD over time [34, 35]. Moreover, physical activity such as high-intensity impact training may complement these dietary interventions by stimulating osteoblast activity and improving musculoskeletal strength and balance [36]. Thus, we hypothesize that SBD111 medical food could exhibit greater benefits in combination with a plant-forward diet rich in isoflavones and other bioactive compounds with or without physical activity compared these components in isolation. These interactions are important to consider in future investigations.

This study provides evidence that SBD111 is safe, well tolerated, and reduces bone loss in women with osteopenia and those with BMI ≥ 30, groups that together comprised 50.3% of enrolled participants. Further investigation is warranted to fully characterize the effects of SBD111 in these contexts, to determine the mechanisms through which SBD111 can preserve bone mass, and to evaluate long-term efficacy. Importantly, these results provide evidence that synbiotic medical foods, such as SBD111, provide an additional opportunity for the dietary management of bone loss in early menopause.

Strengths and limitations

This study has several key strengths. First, it utilized a double-blind, randomized, and placebo controlled design to minimize bias in the study results. Secondly, the study included participants in demographics with elevated risk for bone loss or fracture, including women with osteopenia and those with elevated BMI, and the randomization procedure used ensured comparability between intervention and placebo groups across these prespecified analytic subgroups. Lastly, this study enabled examination of the effects of the synbiotic medical food intervention across several dimensions, including BMD measurements, serum biomarkers of bone turnover and inflammation, and assessment of gut microbiome structure and function. However, this study also has limitations. Most notably, while the safety and efficacy cohort was well powered to detect changes in BMD, the number of participants included in prespecified analytic subgroups is small relative to the effect sizes expected in BMD change. This necessitates confirmatory clinical studies in these populations to validate the results reported here. In addition, this study was conducted in a single country, and demonstrating that these findings generalize to other populations of women would require further investigation. Lastly, while participants in this study were excluded if they indicated a desire to change dietary habits during the study, participants’ diets were not controlled, nor routinely assessed during the trial. Thus we cannot rule out residual confounding by diet in the pre-specified subgroup analyses. High fiber diets or anti-inflammatory diets such as plant-based diets or the Mediterranean diet could lead to beneficial interactions with SBD111 and enhanced anti-bone resorption properties of the medical food. The ability to identify diets that impact the benefits associated medical food interventions like SBD111 could be important information for consumers and will represent a feature to consider in future investigations.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

The study was supported by Solarea Bio, Inc.

Data availability

Any additional information required to reanalyze the data reported in this paper is available upon request by email: eschott@solareabio.com.

Declarations

Institutional review board

The study was conducted in accordance with the Declaration of Helsinki, and approved by the National Institute of Integrative Medicine Human Research Ethics Committee (EC00436) in Brisbane, QLD, Australia (Reference Number: 0089E_2021, August 26, 2021).

Conflicts of interest

Eric Schott, Mark Charbonneau, Alicia Ballok, Gerardo Toledo, Maria Juliana Soto-Giron, Ian Wolff, and D. Davidson Easson are employees of Solarea Bio. Douglas Kiel, Susan Bukata, Michael Zuscik, and Clifford Rosen are advisors of Solarea Bio. Douglas Kiel received grant funding from Solarea Bio to conduct a human safety and tolerability clinical trial and has received an NIH RO1 (R01AG079952) to conduct a clinical trial of SBD111 in women over 60-years of age. Klaus Engelke is a part time employee of Clario. Elizabeth Steels, Harry Huntress, Amanda Rao, Phillippa Ebelt, Thomas Travison, and Luis Vitetta declare they have no conflicts of interest.

Informed consent

Informed consent was obtained from all subjects involved in the study.