Efficacy and safety of stem cells transplantation in patients with type 1 diabetes mellitus—a systematic review and meta-analysis

Abstract

Stem cells (SCs) therapy is a new promising therapeutic modality for type 1 diabetes (T1DM). We performed a systematic review and meta-analysis to evaluate the efficacy and safety of stem cells transplantation (SCT) in patients with T1DM. We searched five literature databases (MEDLINE, EMBASE, Web of Science, WanFang and CENTRAL) up to 31 October 2019. 29 studies (487 patients with T1DM) were included in our meta-analysis. There was no substantial publication bias. Meta-analysis showed the SCT had significant effect to decrease HbA1c (SMD, 1.40; 95% CI, 0.93 to 1.86; p < 0.00001; I² = 89%) and to improve C-peptide levels (SMD, –0.62; 95% CI, –1.22 to –0.02; p = 0.04; I² = 92%) at 1 year follow-up. Subgroup analyses showed the heterogeneity level of the results was high. Significant improvement of metabolic outcomes was observed in the subgroups of mesenchymal stem cells (MSCs) combined with hematopoietic stem cells (HSCs) and HSCs. The older age showed significant association with the efficacy in HSCs subgroup. The higher GADA positive rate before treatment also significantly associated with the decrease of daily insulin requirement. The transient insulin independence rate at last follow-up was 9.6 per 100 person-years (95% CI: 5.8–13.5%). The mean length of insulin independence was 15.6 months (95% CI: 12.3–18.9). The mortality of SCT was 3.4% (95% CI: 2.1–5.5%). Therefore, SCT is an efficacious and safe method for treating patients with T1DM especially in the subgroups of MSCs + HSCs and HSCs. Well designed, double blind and randomized controlled trails with large sample size and long-term follow-up are needed for further evaluation.

TYPE 1 DIABETES MELLITUS (T1DM) is an autoimmune disease resulting from the destruction of islet β cells, leading to a deficiency of insulin secreting and hyperglycemia. In severe cases, it can also lead to multi-organ diabetes-related dysfunction, especially hypercoagulability, neuropathy, kidney disease, retinopathy, and sometimes organ failure. The annual increment of T1DM was about 3%, which has highlighted the serious concerns of health care professionals worldwide.

Individuals with T1DM need insulin replacement for life. In addition, current therapies such as the cell-based therapies and immune therapies are also being used clinically [1]. As a method of the cell-based therapies, islets transplantation has been developed in the past decades for the replacement of damaged β cells [2]. However, the development of autoantibodies, immunosuppression toxicity, limited donor supply, and high costs limit this therapy to patients with T1DM.

Due to the intrinsic regenerative potential and immunomodulatory abilities, the stem cells (SCs) therapy has opened up a breakthrough approach to cure autoimmune disease. Meta-analyses of studies on SCs therapy indicated a certain potential to treat liver cirrhosis [3-5], chronic liver disease [6], inflammatory bowel disease [7] and ischemic heart disease [8]. SCs therapy also showed a potential effect to halt or abolish the autoimmune destruction of β cells and to generate functional β cells which could achieve optimized glycemic control in T1DM [9-13]. Our systematic review and meta-analysis aims to explore the efficacy and safety of stem cells transplantation (SCT) in patients with T1DM.

Materials and Methods

Literature and search strategy

Databases MEDLINE, EMBASE, Web of Science, WanFang and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched from 1977 to 31 October 2019. We used the Medical Subject Headings (MeSH) and key words to construct our search strategy. The search strategy was #1: Mesenchymal Stem Cell Transplantation; #2: Hematopoietic Stem Cell Transplantation; #3: Bone Marrow Stem Cell Transplantation; #4: Peripheral Blood Stem Cell Transplantation; #5: Multipotent Stem Cells; #6: Cord Blood Infusion; #7: Umbilical Cord Mesenchymal Stromal Cell; #8: type 1 diabetes; #9: T1D; #10: Insulin dependent diabetes mellitus; #11: IDDM; #12: ((#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7) AND (#8 OR #9 OR #10 OR #11). We imported the literature results into Endnote X7 and deleted duplicated studies. We also browsed the references of included studies and relevant systematic reviews.

Inclusion and exclusion criteria

Studies were identified carefully by two investigators working independently to determine whether an individual study was eligible for inclusion in the meta-analysis. When agreement could not be reached, a third researcher decided if the study was eligible, based on the same criteria of inclusion and exclusion. Inclusion criteria: individuals of any age, gender and population with T1DM according to the criteria of the American Diabetes Association before SCT. Exclusion criteria: Patients with malignancy, any acute or chronic infection, pregnancy, positive serology for human immunodeficiency virus, hepatitis B or C, underlying hematologic, nephrologic, cardiac, psychiatric, or hepatic disease, mental disorders, inborn or adaptive immunodeficiency and hypersensitivity. Case reported studies were excluded. Studies that did not report the detail data related to the efficacy or safety of SCT were also excluded.

Data extraction

Data were independently extracted by two investigators who reached a consensus on all of the items. Information extracted from each study was considered as follows: first author, year of publication, descriptions of patients, mode of injection and the outcomes we interested in, such as the daily insulin requirement, glycosylated hemoglobin (HbA1c) and C-peptide level in treatment group before and after treatment as well as the adverse and mortality effect. We defined severe complications contains bilateral pneumonia and pseudomonas aeruginosa sepsis, which may cause damage to vital organs. Dealing with missing data: we contacted original authors to clarify missing data or data which were not clearly reported.

Quality assessment of included studies

Two reviewers assessed methodological quality independently according to the Newcastle-Ottawa Scale (NOS). Any discrepancies were resolved by discussion, if disagreements cannot be resolved, a third reviewer were consulted.

Statistical analysis

Review Manager 5 and Comprehensive Meta Analysis software were used to perform the analyses. For studies reported quantitative data, a meta-analysis was undertaken when studies were sufficiently homogenous. If only two or three studies were to be identified, we performed a fixed-effects model meta-analysis. If more than three studies were to be identified, we performed random-effects analysis. Publication bias was assessed by the funnel plots. Statistical heterogeneity was assessed by the forest plots. Subgroups were classified based on various sources of SCs, including mesenchymal stem cells (MSCs), hematopoietic stem cells (HSCs), MSCs combined with HSCs (MSCs + HSCs), umbilical cord blood (UCB) and bone marrow mononuclear cells (BM-MNCs). We investigated the heterogeneity using the I² statistic which means the proportion of the unexplained heterogeneity by the estimates variability among studies. I² values of <25%, 25–50% and >50% represent minimal, moderate and substantial heterogeneity, respectively [14]. If the heterogeneity was high, meta-regression analyses were performed to test the association between the factors and the outcomes. All summary effects were presented with 95% CI [15].

Results

Characteristics of studies

We retrieved 885 references from the electronic database, as showed on the flow diagram (Fig. 1). The 29 eligible studies met the inclusion and exclusion criteria [16-44]. Main study characteristics were described in Table 1. In general, the total number of patients who underwent SCT was 487. Among them, 5 studies (including 45 participants) underwent MSCs transplantation, 13 studies (including 275 participants) underwent HSCs transplantation, 3 studies (including 52 participants) underwent MSCs + HSCs transplantation, 6 studies (including 83 participants) underwent UCB transplantation, and 2 studies (including 32 participants) underwent BM-MNCs transplantation. The ages of T1DM patients receiving SCT ranged from 3.02 to 40.3 years. These patients had less than 2 months to 11 years of disease duration before SCT. The follow-up periods ranged from 3 to 60 months. The overall quality of the included studies was optimal (Table 2). There was no substantial publication bias be observed (Fig. 2).

Fig. 1

Flow diagram of the references selection

Table 1 General characteristics of the included studies

Studies	SB	Country	No. of patients	Female%	Mean age of patients (years)	Mean duration of T1DM (month)	GADA+ (%)	Type of SCs	Mean dose of transplanted (cells/kg)	Mean length of follow-up (month)	Mode of injection
Couri 2009		Brazil	23	26.1	18.4	0.62	87	HSCs	10.52 × 106	29.8	IV
Haller 2009		USA	15	46.7	5.5	4.1	NA	UCB	1.5 × 107	12	IV
Gu 2010		China	18	66.7	18.4	<6	NA	HSCs	NA	13.8	IV
Vanikar 2010		India	11	36.4	21.1	98.4	45.4	MSCs + HSCs	3.15 × 106   NA	23	IP
Feng 2011		China	16	56	13	9.9	NA	HSCs	5.1 × 106	7	IV
Haller 2011		USA	24	58.3	5.1	4	NA	UCB	1.88 × 107	12	IV
Yu WL 2011		China	6	50	19.67	<3	NR	MSCs	1 × 107	9	IV
Gu 2012		China	28	50	17.6	3	93	HSCs	1 × 106	23	IV
Li 2012		China	13	23.1	14.1	<12	54	HSCs	4 × 106	42	IV
Zhang 2012		China	9	44	17.6	2	100	HSCs	10.49 × 106	12	IV
Zhao Y 2012	A1	China	6	66.7	30	72	100	UCB	1 × 106	6	IV
	B2	China	6	83.3	27	132	66.7	UCB	1 × 106	6	IV
Haller 2013		USA	10	20	7.2	3.96	NA	UCB	1.1 × 107	12	IV
Hu JX 2013		China	15	40	17.6	<6	73.3	MSCs	2.6 × 107	24	IV
Mesples 2013		China	2	100	7	<2	100	BM-MNCs	180 × 106	12	IP
Wang 2013		China	22	54.5	18.0	<6	100	HSCs	NA	24	IV
Giannopoulou 2014		Germany	7	29	3.02	3.36	NA	UCB	1.27 × 106	12	IV
D’Addio 2014		China	65	37	20.4	<3	90.7	HSCs	5.8 × 106	48	IV
Gu 2014		China	14	64	8.04	<3	NA	HSCs	NA	36–60	IV
Hou T 2014		China	15	46.7	18.95	4.22	NA	HSCs	>3 × 106	14	IV
Carlsson 2015		Sweden	9	11.1	24	<0.75	67.7	MSCs	2.75 × 106	12	IV
Jia XL 2015		China	10	90	8–31	<12	80	MSCs	1 × 106	12	IV
Thakkar 2015	A3	India	10	10	20.2	97.2	100	MSCs + HSCs	2.65 × 104 NA	33.1	IP
	B4	India	10	40	19.7	118.8	NA	MSCs + HSCs	2.07 × 104 NA	33.1	IP
Delgado 2015	A1	Spain	6	33.3	33.8	6.5	NA	UCB	1.0 × 106	12	IV
	B2	Spain	9	33.3	40.3	12.4	NA	UCB	1.0 × 106	12	IV
Cai 2016		China	21	57	18.29	111	66.7	MSCs + HSCs	1.1 × 106 106.78 × 106	12	IP
Snarski 2016		Poland	24	29	26.5	2.03	100	HSCs	4.19 × 106	52	IV
Ye 2017		China	8	62.5	18.86	<6	100	HSCs	2 × 105	12	IV
Gu 2018		China	20	75	18	2.26	NA	HSCs	NA	48	IV
Mohamed 2018	A5	Egypt	15	46.6	22.3	11.9	NA	BM-MNCs	1 × 106	3	IP
	B6	Egypt	15	53.3	22.3	11.9	NA	BM-MNCs	1 × 106	3	IP
Ulyanova 2019		Kazakhstan	5	20	20–42	NA	NA	MSCs	96 × 106	3	IV

GADA+(%), GADA positive rate before treatment; SB, subgroup of studies; SCs, stem cells; HSCs, hematopoietic stem cells; UCB, umbilical cord blood; MSCs, mesenchymal stem cells; BM-MNC, bone marrow mononuclear cells; NA, not applicable; 1, subgroup A of patients having some residual islet β cell function; 2, subgroup B of patients with no residual islet β cell function; 3, subgroup A of patients transplanted with autologous cells; 4, subgroup B of patients transplanted with allogenic cells; 5, subgroup A of patients combined with exercise program; 6, subgroup B of patients without exercise program; IP, intra-pancreatic; IV, intravenous.

Table 2 Quality of included studies

Studies	Selection	Comparability	Exposure	Total
Couri 2009	4	0	3	7
Haller 2009	4	0	3	7
Gu 2010	4	0	3	7
Vanikar 2010	4	0	3	7
Feng 2011	4	0	2	6
Haller 2011	4	0	3	7
Yu WL 2011	4	2	2	8
Gu 2012	4	0	3	7
Li 2012	4	0	3	7
Zhang 2012	4	0	3	7
Zhao Y 2012	4	2	2	8
Haller 2013	4	2	3	9
Hu JX 2013	4	2	3	9
Mesples 2013	3	1	3	7
Wang 2013	4	2	3	9
Giannopoulou 2014	4	2	3	9
D’Addio 2014	4	0	3	7
Gu 2014	4	2	3	9
Hou T 2014	4	2	3	9
Carlsson 2015	4	2	3	9
Jia XL 2015	4	0	3	7
Thakkar 2015	4	2	3	9
Delgado 2015	4	0	3	7
Cai 2016	4	2	3	9
Snarski 2016	4	0	3	7
Ye 2017	4	2	3	9
Gu 2018	4	2	3	9
Mohamed 2018	4	0	2	6
Ulyanova 2019	4	0	2	6

Fig. 2

Funnel plots of comparing the individual results of administrating stem cells transplantation in T1DM: (A) daily insulin requirement (U/kg per day); (B) HbA1c (%); (C) C-peptide (ng/mL). (MSCs, mesenchymal stem cells; HSCs, hematopoietic stem cells; UCB, umbilical cord blood; BM-MNCs, bone marrow mononuclear cells; HbA1c, glycosylated hemoglobin)

Outcomes of MSCs for TIDM

Metabolic outcomes

Fig. 3 and Table 3 showed the meta-analysis of daily insulin requirement (U/kg per day), HbA1c (%) and C-peptide (ng/mL) in patients accepting SCT after 12 months follow-up. The overall analysis demonstrated that SCT had significant effect to decrease HbA1c (SMD, 1.40; 95% CI, 0.93 to 1.86; p < 0.00001; I² = 89%) and to improve C-peptide levels (SMD, –0.62; 95% CI, –1.22 to –0.02; p = 0.04; I² = 92%). SCT showed no significant effect to reduce daily insulin requirement (SMD, 0.55; 95% CI, –0.01 to 1.10; p = 0.05; I² = 90%), but the p was on the critical value.

Fig. 3

Forest plots showing the individual results of administrating different types of stem cells transplantation in T1DM: (A) daily insulin requirement (U/kg per day); (B) HbA1c (%); (C) C-peptide (ng/mL). (MSCs, mesenchymal stem cells; HSCs, hematopoietic stem cells; UCB, umbilical cord blood; BM-MNCs, bone marrow mononuclear cells; HbA1c, glycosylated hemoglobin)

Fig. 3

Fig. 3

Table 3 The effect of stem cells therapy on the daily insulin requirement levels, HbA1c and C-peptide in T1DM

Variables	Type of stem cells	Summary estimates of mean difference 95% CI	p value	I2 (%)
Daily insulin requirement (U/kg per day)	MSCs	0.50 [–0.18, 1.18]	0.15	20
	MSCs + HSCs	1.45 [1.01, 1.89]	<0.00001	0
	UCB	–1.96 [–3.65, –0.27]	0.02	93
	HSCs	1.51 [1.02, 2.00]	<0.00001	74
	BM-MNC	0.26 [–0.25, 0.77]	0.32	0
	Total	0.55 [–0.01, 1.10]	0.05	90
HbA1c (%)	MSCs	1.08 [0.36, 1.80]	0.003	51
	MSCs + HSCs	0.95 [0.38, 1.52]	0.001	43
	UCB	0.25 [–0.06, 0.56]	0.12	0
	HSCs	2.42 [1.45, 3.38]	<0.00001	94
	BM-MNC	1.24 [0.44, 2.04]	0.002	33
	Total	1.40 [0.93, 1.86]	<0.00001	89
C-peptide (ng/mL)	MSCs	–0.20 [–1.17, 0.77]	0.69	74
	MSCs + HSCs	–1.28 [–1.71, –0.85]	<0.00001	0
	UCB	0.75 [–0.54, 2.04]	0.25	91
	HSCs	–1.64 [–2.72, –0.57]	0.003	95
	BM-MNC	–0.31 [–0.82, 0.19]	0.22	0
	Total	–0.62 [–1.22, –0.02]	0.04	92

MSCs, mesenchymal stem cells; HSCs, hematopoietic stem cells; UCB, umbilical cord blood; BM-MNCs, bone marrow mononuclear cells; HbA1c, glycosylated hemoglobin.

Due to the high heterogeneity, we performed subgroups analysis. The subgroups analysis demonstrated that SCT decreased the daily insulin requirement with statistically significance when compared with baseline after 1 year in the subgroups treated with MSCs + HSCs (SMD, 1.45; 95% CI, 1.01 to 1.89; p < 0.00001; I² = 0%) and HSCs (SMD, 1.51; 95% CI, 1.02 to 2.00; p < 0.00001; I² = 74%). There was no significant reduction on the daily insulin requirement in the subgroups of MSCs (SMD, 0.5; 95% CI, –0.18 to 1.18; p = 0.15; I² = 20%) and BM-MNC (SMD, 0.26; 95% CI, –0.25 to 0.77; p = 0.32; I² = 0%) when compared with baseline. On the contrary, daily insulin requirement was increased with statistically significance after treatment in UCB group (SMD, –1.96; 95% CI, –3.65 to –0.27; p = 0.02; I² = 93%) (Fig. 3A).

The analysis indicated that SCT significantly decreased HbA1c levels at 1 year in the MSCs group (SMD, 1.08; 95% CI, 0.36 to 1.80; p = 0.003; I² = 51%), the MSCs + HSCs group (SMD, 0.95; 95% CI, 0.38 to 1.52; p = 0.001; I² = 43%), the HSC group (SMD, 2.42; 95% CI, 1.45 to 3.38; p < 0.00001; I² = 94%) and the BM-MNC group (SMD, 1.24; 95% CI, 0.44 to 2.04; p = 0.002; I² = 33%). While patients treated by UCB did not show significant decreasing on the HbA1c levels (SMD, 0.25; 95% CI, –0.06 to 0.56; p = 0.12; I² = 0%) (Fig. 3B).

When evaluating the effect of SCT on C-peptide levels, the MSCs + HSCs group (SMD, –1.28; 95% CI, –1.71 to –0.85; p < 0.00001; I² = 0%) and HSCs group (SMD, –1.64; 95% CI, –2.72 to –0.57; p = 0.003; I² = 95%) showed statistically significant effects. However, there were no statistically significant effects in MSCs, UCB and BM-MNCs groups (SMD, –0.20; 95% CI, –1.17 to 0.77; p = 0.69; I² = 74%; SMD, 0.75; 95% CI, –0.54 to 2.04; p = 0.25; I² = 91%; SMD, –0.31; 95% CI, –0.82 to 0.19; p = 0.22; I² = 0% respectively) (Fig. 3C).

Because of the high heterogeneity still observed in HSCs group, we further used meta-regression analysis to investigate the factors which might affect the efficacy of HSCs transplantation. The results showed that the older age and the higher GADA positive rate before treatment were significantly associated with the decrease of daily insulin requirement (p = 0.00017 and p = 0.00041 respectively) (Fig. 4A and Fig. 4B). The older age was associated with the decrease of HbA1c and the improvement of C-peptide levels (p = 0.00077 and p = 0.00065 respectively) (Fig. 4C and Fig. 4D). However, the number of transplanted cells and the duration of T1DM showed no significant association with the outcomes in HSCs group (Table 4).

Fig. 4

Meta-regression analyses of the factors on the efficacy of HSCs transplantation.

(A) Meta-regression of age on Δdaily insulin requirement (U/kg per day) (p = 0.00017);

(B) Meta-regression of GADA+ (%) on Δdaily insulin requirement (U/kg per day) (p = 0.00041);

(C) Meta-regression of age on ΔHbA1c (%) (p = 0.00077);

(D) Meta regression of age on ΔC-peptide (ng/mL) (p = 0.00065).

(Δdaily insulin requirement, decrease of daily insulin requirement after treatment; GADA+ (%),GADA positive rate before treatment; ΔHbA1c, decrease of glycosylated hemoglobin after treatment; ΔC-peptide, increase of C-peptide after treatment)

Table 4 Results of meta-regression analysis between the outcomes and the factors may affect the efficacy of HSCs transplantation

Outcomes	Factors may affect the efficacy of HSCT	No. of Studies	p value
ΔDaily insulin requirement (U/kg per day)	Age (year)	11	0.00017
	Mean dose of transplanted cells (×106 cells/kg)	6	>0.05
	Mean duration of T1DM (month)	11	>0.05
	GADA+ (%)	6	0.00041
ΔHbA1c (%)	Age (year)	13	0.00077
	Mean dose of transplanted cells (×106 cells/kg)	7	>0.05
	Mean duration of T1DM (month)	13	>0.05
	GADA+ (%)	8	>0.05
ΔC-peptide (ng/mL)	Age (year)	10	0.00065
	Mean dose of transplanted cells (×106 cells/kg)	6	>0.05
	Mean duration of T1DM (month)	10	>0.05
	GADA+ (%)	7	>0.05

No., number; Δdaily insulin requirement, decrease of daily insulin requirement after treatment; GADA+ (%), GADA positive rate before treatment; ΔHbA1c, decrease of HbA1c after treatment; ΔC-peptide, increase of C-peptide after treatment.

Insulin independent rate

The incidence of patients who obtained transient insulin independence after SCT was 9.6 per 100 person-years (95% CI: 5.8 to 13.5%, I² = 72). 10 studies reported the mean period of transient insulin independence after SCT [16, 20, 23-25, 28, 31, 33, 40, 42] which was 15.6 months (95% CI: 12.3 to 18.9, I² = 95) (Table 5). Yu WL et al. [22] also observed 3 patients achieving insulin independence after SCT, but did not reported the mean period. Among the 11 studies, only 4 were randomized controlled trials (RCT) [22, 28, 33, 42], the others were self-controlled trails. When analyzing the insulin independence rate in the 4 RCT studies, the SCT showed the significant effects to improve the insulin independence rate compared to control group which treated with insulin (risk ratio (RR), 9.67; 95% CI, 2.62 to 35.62; p = 0.001; I² = 0%).

Table 5 Transient insulin independence rate and adverse events assessment

	Studies	Events	Total	Rate	95% CI	I2
Transient insulin independence rate (100-PY)	27	138	447	9.6	5.8–13.5	72
Mean length of transient insulin independence (months)	10	—	—	15.6	12.3–18.9	95
Alopecia (%)	26	100	429	13.9	5.8–29.6	64
Fever (%)	28	83	465	12.5	6.9–21.6	69
Vomiting (%)	28	76	465	9.9	4.0–22.6	75
Nausea (%)	28	75	465	9.8	4.0–22.1	76
Gastric tract symptoms (%)	28	56	445	7.7	3.8–15.1	59
Leukopenia (%)	28	41	465	6.7	3.1–13.9	64
Rash (%)	28	29	465	5.5	2.7–10.7	56
Diarrhea (%)	28	24	465	5.3	3.0–9.0	27
Infection (%)	28	15	465	5.4	2.9–10.0	49
Allergic/Urticaria (%)	28	10	465	6.1	4.0–9.1	0
Oligospermia (%)	28	9	465	4.6	2.6–8.0	28
Anorexia (%)	28	8	465	5.0	3.0–8.3	19
Severe complications*	28	4	465	3.9	2.4–6.2	0
Mortality (%)	28	2	465	3.4	2.1–5.6	0

PY, person-year; *Severe complications contains bilateral pneumonia and pseudomonas aeruginosa sepsis.

Adverse events assessment

The common adverse events contained alopecia, fever, vomiting, nausea, gastric tract symptoms, leucopenia, rash and diarrhea. The incident rate of severe adverse events, as bilateral pneumonia and pseudomonas aeruginosa sepsis, was 3.9% (95% CI: 2.4–6.2%). Two patients died due to the pseudomonas aeruginosa sepsis reported in two studies [31, 40]. The incidence rate of mortality was 3.4% (95% CI: 2.1% to 5.5%, I² = 0) (Table 5).

Discussion

The prevalence of T1DM has been increasing steadily over the past decades. Although the rapid developments of modern diabetes treatment have improved the life quality of T1DM patients, the long-term complications of T1DM cannot be completely avoided. More than a decade ago, a new research field named SCs therapy emerged in health science. Many medical research institutes began nearly a decade research on SCT in patients with T1DM.

To draw a more definitive conclusion of the efficacy and safety of SCT in T1DM, we performed a meta-analysis of 29 published articles. As indicated by the pooled effect in our meta-analysis, MSCs + HSCs and HSCs groups showed significant effects to reduce daily insulin requirement and HbA1c levels, as well as to improve C-peptide levels at 1 year follow-up compared to the baseline. Whereas the MSCs group only showed significant effect to decrease HbA1c levels but no significant effect to reduce daily insulin requirement or to improve C-peptide levels, which was different from the meta-analysis conducted by Gan J et al. [45]. As we know that, β cells function will decrease rapidly within years during the natural course of T1DM. The analysis suggest that MSCs, MSCs + HSCs and HSCs groups could be effective in maintaining β cells function at 1 year follow-up. However, further studies are needed to confirm the long-term effectiveness. Our data also showed that the older age was significantly associated with the decrease of daily insulin requirement, the decrease of HbA1c and the improvement of C-peptide levels in the HSCs subgroup. We further explored the detail data of these studies and found that the mean disease durations before treatment in these patients were less than 1 year, especiailly most of them were less than 6 months. Thus, if newly diagnosed older patients with T1DM could undergo HSCs transplantation within 1 year, the better outcomes would be achieved. In addition, we also found that the higher GADA positive rate before treatment showed significant association with the decrease of daily insulin requirement. The positive of GADA represented the definitely autoimmunity dierected towards the islet β cells. The immunomodulatory of SCT might play a role on the glycemic control and the decreasd insulin dosage.

Initially, the SCs therapeutic potential was considered as the ability of differentiation into insulin-producing cells. However, this dogma has changed when new results had been published regarding to the immunomodulatory properties of SCs [46]. Vanikar et al. reported that the potential therapeutic effects of SCs was attributed to their intrinsic regenerative capacity and immunomodulatory properties, which could arrest autoimmune β cells destruction, protect residual β cells, promote endogenous regeneration, improve innate/alloimmune graft rejection, and restore β-cell-specific unresponsiveness without chronic immunosuppression and to reverse hyperglycemia [12].

While the UCB group showed no significant effect to decrease HbA1c levels or to improve C-peptide levels, but even significantly increased daily insulin requirement after treatment. In addition to the small sample of the studies, some other factors might actually limit the capacity of UCB to preserve β cells function. First, an insufficient number of cells that carried regenerative or immunoregulatory capacity may have been transferred to patients with T1DM. Second, the ongoing autoimmune response in newly-onset T1DM may contain memory T cells, which are refractive to regulate by UCB [47], that facilitate the ongoing autoimmune destruction of endogenous or de novo β cells. Parker MJ et al. suggested that a therapeutic approach involving transient immune depletion and subsequent induction of immune regulation is optimal [48]. Because of containing an abundant number of immature unprimed highly functional regulatory T cells (Tregs) which may facilitate bystander suppression of effector T cells, UCB still owns therapeutic potential on T1DM. Now the clinical trials are currently underway in patients with T1DM using expanded Tregs isolated from UCB (NCT02932826). We are looking forward to see the results to explore the therapeutic potential of UCB.

In addition, we analyzed the insulin independence length after SCT. The mean length of patients who achieved transient insulin independence after SCT was 15.6 months. As we all know, newly diagnosed T1DM patients may enter the honeymoon phase after initiation of insulin therapy. During this period, patients can maintain a good glycemic control with a little or no insulin treatment. This period usually lasts no longer than 1.5 years. After that, they will need more insulin to control their blood glucose. To eliminate the honeymoon period, we analyzed RR from the 4 RCT studies, the result suggested the SCT showed significant effect to improve insulin independence rate compared to control group. Among the 4 RCT studies, only one patient achieved 7-months insulin independence in the insulin group [42]. Thus, we preliminary draw the conclusion that SCT might prolong the insulin independence in T1DM. And more RCT studies with long times follow-up and large sample size analyses are needed to further investigation.

Although SCT appears to be effective, the safety is an important issue. The common adverse events contained alopecia, fever, vomiting, nausea, gastric tract symptoms, leucopenia, rash and diarrhea. Gu et al. reported that most of the early adverse events were caused by the mobilization and conditioning drugs used during HSCs transplantation, whereas later adverse events were unremarkable compared with patients using insulin therapy [42]. D’Addio et al. found that, most of the adverse events due to the administration of a high-dose immunosuppressive regimen. This observation indicated the need for reducing the dosage of immunosuppressive or for eventually changing the immunosuppressive agents and for a stronger prophylaxis for infections to obtain a safer approach [31]. Our meta-analysis showed that the incidence rate of severe adverse events and the mortality rate were 3.9% and 3.4% respectively. Two patients died due to pseudomonas sepsis after autologous HSCs transplantation reported by two studies respectively. Thus, further studies and more clinical evidence are needed to ensure the safety of autologous HSCs transplantation before it can be more widely used in clinical.

In conclusion, our systematic review and meta-analysis suggested that MSCs + HSCs and HSCs transplantation have potential to significantly improve the β cells function of patients with T1DM. However, a complete cure of T1DM is yet to be achieved. Since the number of patients with T1DM is increasing, there needs more controlled and randomized trials with larger number of patients and long-term observations to draw more credible results of SCs therapy. In order to find the best approach, it is worthwhile exploring the precise mechanisms, dosage, timing and frequency to consummate the therapy.

Authors’ Contributions

W.Q. was involved in conception and design, analysis and interpretation of data, drafting of the article. Z.S. worked on collection and interpretation of data. Q.Y., Z.XQ., C.H. and Y.T. were involved in the analysis and interpretation of data and critical revision of the manuscript for important intellectual content. Z.M. was involved in conception and design, and critical revision of the manuscript for important intellectual content. All the co-authors gave final approval of the version to be published.

Acknowledgements

The authors thank the National Nature Science Foundation of China (81670756; 81974103), Provincial Key Research and Development Foundation of Jiangsu (BE2018748), 333 High-level Talent Training Project of the Jiangsu Province (LGY2016007) for financial support.

Author Disclosure Statement

No competing financial interests exist.

References

• 1   Pathak  V,  Pathak  NM,  O’Neill  CL,  Guduric-Fuchs  J,  Medina  RJ (2019) Therapies for type 1 diabetes: current scenario and future perspectives. Clin Med Insights Endocrinol Diabetes 12: 1179551419844521.
• 2   Bretzel  RG,  Jahr  H,  Eckhard  M,  Martin  I,  Winter  D, et al. (2007) Islet cell transplantation today. Langenbecks Arch Surg 392: 239–253.
• 3   Qi  X,  Guo  X,  Su  C (2015) Clinical outcomes of the transplantation of stem cells from various human tissue sources in the management of liver cirrhosis: a systematic review and meta-analysis. Curr Stem Cell Res Ther 10: 166–180.
• 4   Tao  H,  Li  Y,  Wang  T,  Zhou  C (2018) Umbilical cord blood stem cells transplantation as an adjunctive treatment strategy for liver cirrhosis in Chinese population: a meta-analysis of effectiveness and safety. Ther Clin Risk Manag 14: 417–440.
• 5   Ma  XR,  Tang  YL,  Xuan  M,  Chang  Z,  Wang  XY, et al. (2015) Transplantation of autologous mesenchymal stem cells for end-stage liver cirrhosis: a meta-analysis based on seven controlled trials. Gastroenterol Res Pract 2015: 908275.
• 6   Kim  G,  Eom  YW,  Baik  SK,  Shin  Y,  Lim  YL, et al. (2015) Therapeutic effects of mesenchymal stem cells for patients with chronic liver diseases: systematic review and meta-analysis. J Korean Med Sci 30: 1405–1415.
• 7   Dave  M,  Jaiswal  P,  Cominelli  F (2017) Mesenchymal stem/stromal cell therapy for inflammatory bowel disease: an updated review with maintenance of remission. Curr Opin Gastroenterol 33: 59–68.
• 8   Jeong  H,  Yim  HW,  Park  HJ,  Cho  Y,  Hong  H, et al. (2018) Mesenchymal stem cell therapy for ischemic heart disease: systematic review and meta-analysis. Int J Stem Cells 11: 1–12.
• 9   Chhabra  P,  Brayman  KL (2013) Stem cell therapy to cure type 1 diabetes: from hype to hope. Stem Cells Transl Med 2: 328–336.
• 10   Kakkar  A,  Sorout  A,  Tiwari  M,  Shrivastava  P,  Meena  P, et al. (2018) Current status of stem cell treatment for type i diabetes mellitus. Tissue Eng Regen Med 15: 699–709.
• 11   Sordi  V,  Pellegrini  S,  Krampera  M,  Marchetti  P,  Pessina  A, et al. (2017) Stem cells to restore insulin production and cure diabetes. Nutr Metab Cardiovasc Dis 27: 583–600.
• 12   Vanikar  AV,  Trivedi  HL,  Thakkar  UG (2016) Stem cell therapy emerging as the key player in treating type 1 diabetes mellitus. Cytotherapy 18: 1077–1086.
• 13   Farooq  T,  Rehman  K,  Hameed  A,  Akash  MSH (2019) Stem cell therapy and type 1 diabetes mellitus: treatment strategies and future perspectives. Adv Exp Med Biol 1084: 95–107.
• 14   Higgins  JP,  Thompson  SG,  Deeks  JJ,  Altman  DG (2003) Measuring inconsistency in meta-analyses. BMJ 327: 557–560.
• 15   DerSimonian  R,  Laird  N (1986) Meta-analysis in clinical trials. Control Clin Trials 7: 177–188.
• 16   Couri  CE,  Oliveira  MC,  Stracieri  AB,  Moraes  DA,  Pieroni  F, et al. (2009) C-peptide levels and insulin independence following autologous nonmyeloablative hematopoietic stem cell transplantation in newly diagnosed type 1 diabetes mellitus. JAMA 301: 1573–1579.
• 17   Haller  MJ,  Wasserfall  CH,  McGrail  KM,  Cintron  M,  Brusko  TM, et al. (2009) Autologous umbilical cord blood transfusion in very young children with type 1 diabetes. Diabetes Care 32: 2041–2046.
• 18   Vanikar  AV,  Dave  SD,  Thakkar  UG,  Trivedi  HL (2010) Cotransplantation of adipose tissue-derived insulin-secreting mesenchymal stem cells and hematopoietic stem cells: a novel therapy for insulin-dependent diabetes mellitus. Stem Cells Int 2010: 582382.
• 19   Gu  W,  Sun  S,  Hu  J (2010) Efficacy and safety of autologous hematopoietic stem cell transplantation in treating type 1 diabetes mellitus. Chinese Journal of Endocrinology and Metabolism 26: 1023–1026 (In Chinese).
• 20   Feng  K,  Xu  YW,  Ye  FG,  Xiao  L,  Ma  XH, et al. (2011) Autologous peripheral blood hematopoietic stem cell transplantation in the treatment of type 1 diabetic mellitus: a report of 16 cases. Zhonghua Yi Xue Za Zhi 91: 1966–1969 (In Chinese).
• 21   Haller  MJ,  Wasserfall  CH,  Hulme  MA,  Cintron  M,  Brusko  TM, et al. (2011) Autologous umbilical cord blood transfusion in young children with type 1 diabetes fails to preserve c-peptide. Diabetes Care 34: 2567–2569.
• 22   Yu  W,  Gao  H,  Yu  X,  Wang  L,  Yan  S, et al. (2011) Umbilical cord mesenchymal stem cells transplantation for newly-onset type 1 diabetes. Journal of Clinical Rehabilitative Tissue Engineering Research 15: 4363–4366 (In Chinese).
• 23   Gu  W,  Hu  J,  Wang  W,  Li  L,  Tang  W, et al. (2012) Diabetic ketoacidosis at diagnosis influences complete remission after treatment with hematopoietic stem cell transplantation in adolescents with type 1 diabetes. Diabetes Care 35: 1413–1419.
• 24   Li  L,  Shen  S,  Ouyang  J,  Hu  Y,  Hu  L, et al. (2012) Autologous hematopoietic stem cell transplantation modulates immunocompetent cells and improves beta-cell function in Chinese patients with new onset of type 1 diabetes. J Clin Endocrinol Metab 97: 1729–1736.
• 25   Zhang  X,  Ye  L,  Hu  J,  Tang  W,  Liu  R, et al. (2012) Acute response of peripheral blood cell to autologous hematopoietic stem cell transplantation in type 1 diabetic patient. PLoS One 7: e31887.
• 26   Zhao  Y,  Jiang  Z,  Zhao  T,  Ye  M,  Hu  C, et al. (2012) Reversal of type 1 diabetes via islet beta cell regeneration following immune modulation by cord blood-derived multipotent stem cells. BMC Med 10: 3.
• 27   Haller  MJ,  Wasserfall  CH,  Hulme  MA,  Cintron  M,  Brusko  TM, et al. (2013) Autologous umbilical cord blood infusion followed by oral docosahexaenoic acid and vitamin d supplementation for c-peptide preservation in children with type 1 diabetes. Biol Blood Marrow Transplant 19: 1126–1129.
• 28   Hu  J,  Yu  X,  Wang  Z,  Wang  F,  Wang  L, et al. (2013) Long term effects of the implantation of wharton’s jelly-derived mesenchymal stem cells from the umbilical cord for newly-onset type 1 diabetes mellitus. Endocr J 60: 347–357.
• 29   Mesples  A,  Majeed  N,  Zhang  Y,  Hu  X (2013) Early immunotherapy using autologous adult stem cells reversed the effect of anti-pancreatic islets in recently diagnosed type 1 diabetes mellitus: preliminary results. Med Sci Monit 19: 852–857.
• 30   Wang  B,  Gao  J,  Zhang  J,  Wang  L,  L  L, et al. (2013) Comparison of blood glucose after autologous hematopoietic stem cell transpalntation and insulin therapy in patients with type 1 diabetes by continuous glucose monitoring system. Shanghai Med J 36: 394–397 (In Chinese).
• 31   D’Addio  F,  Valderrama Vasquez  A,  Ben Nasr  M,  Franek  E,  Zhu  D, et al. (2014) Autologous nonmyeloablative hematopoietic stem cell transplantation in new-onset type 1 diabetes: a multicenter analysis. Diabetes 63: 3041–3046.
• 32   Giannopoulou  EZ,  Puff  R,  Beyerlein  A,  von Luettichau  I,  Boerschmann  H, et al. (2014) Effect of a single autologous cord blood infusion on beta-cell and immune function in children with new onset type 1 diabetes: a non-randomized, controlled trial. Pediatr Diabetes 15: 100–109.
• 33   Gu  Y,  Gong  C,  Peng  X,  Wei  L,  Su  C, et al. (2014) Autologous hematopoietic stem cell transplantation and conventional insulin therapy in the treatment of children with newly diagnosed type 1 diabetes: long term follow-up. Chin Med J (Engl) 127: 2618–2622.
• 34   Hou  T,  Shi  Y (2014) Control study of autologous stem cell transplantation action glycemic control in patients with type 1 diabetes. China Modern Doctor 52: 34–37 (In Chinese).
• 35   Carlsson  PO,  Schwarcz  E,  Korsgren  O,  Le Blanc  K (2015) Preserved beta-cell function in type 1 diabetes by mesenchymal stromal cells. Diabetes 64: 587–592.
• 36   Delgado  E,  Perez-Basterrechea  M,  Suarez-Alvarez  B,  Zhou  H,  Revuelta  EM, et al. (2015) Modulation of autoimmune T-cell memory by stem cell educator therapy: phase 1/2 clinical trial. EBioMedicine 2: 2024–2036.
• 37   Thakkar  UG,  Trivedi  HL,  Vanikar  AV,  Dave  SD (2015) Insulin-secreting adipose-derived mesenchymal stromal cells with bone marrow-derived hematopoietic stem cells from autologous and allogenic sources for type 1 diabetes mellitus. Cytotherapy 17: 940–947.
• 38   Jia  X,  Shen  S,  Li  L,  Hui  H,  XU  Q, et al. (2015) Clinical efficacy of allogenic mesenchymal stem cells transplantation in patients with ketosis-onset type 1 diabetes mellitus. Chinese Journal of Diabetes Mellitus 7: 425–430 (In Chinese).
• 39   Cai  J,  Wu  Z,  Xu  X,  Liao  L,  Chen  J, et al. (2016) Umbilical cord mesenchymal stromal cell with autologous bone marrow cell transplantation in established type 1 diabetes: a pilot randomized controlled open-label clinical study to assess safety and impact on insulin secretion. Diabetes Care 39: 149–157.
• 40   Snarski  E,  Milczarczyk  A,  Halaburda  K,  Torosian  T,  Paluszewska  M, et al. (2016) Immunoablation and autologous hematopoietic stem cell transplantation in the treatment of new-onset type 1 diabetes mellitus: long-term observations. Bone Marrow Transplant 51: 398–402.
• 41   Ye  L,  Li  L,  Wan  B,  Yang  M,  Hong  J, et al. (2017) Immune response after autologous hematopoietic stem cell transplantation in type 1 diabetes mellitus. Stem Cell Res Ther 8: 90.
• 42   Gu  B,  Miao  H,  Zhang  J,  Hu  J,  Zhou  W, et al. (2018) Clinical benefits of autologous haematopoietic stem cell transplantation in type 1 diabetes patients. Diabetes Metab 44: 341–345.
• 43   Mohamed  MT,  Embaby  EA,  Labib  A,  El-Husseiny  M,  Khamis  H, et al. (2018) Effects of exercise in combination with autologous bone marrow stem cell transplantation for patients with type 1 diabetes. Physiother Theory Pract 35: 1233–1242.
• 44   Ulyanova  O,  Askarov  M,  Kozina  L,  Karibekov  T,  Shaimardanova  G, et al. (2019) Autologous mesenchymal stem cell transplant in patients with type 1 diabetes mellitus. Exp Clin Transplant 17: 236–238.
• 45   Gan  J,  Wang  Y,  Zhou  X (2018) Stem cell transplantation for the treatment of patients with type 1 diabetes mellitus: a meta-analysis. Exp Ther Med 16: 4479–4492.
• 46   Borisov  MA,  Petrakova  OS,  Gvazava  IG,  Kalistratova  EN,  Vasiliev  AV (2016) Stem cells in the treatment of insulin-dependent diabetes mellitus. Acta Naturae 8: 31–43.
• 47   Schneider  A,  Rieck  M,  Sanda  S,  Pihoker  C,  Greenbaum  C, et al. (2008) The effector T cells of diabetic subjects are resistant to regulation via CD4+ foxp3+ regulatory T cells. J Immunol 181: 7350–7355.
• 48   Parker  MJ,  Xue  S,  Alexander  JJ,  Wasserfall  CH,  Campbell-Thompson  ML, et al. (2009) Immune depletion with cellular mobilization imparts immunoregulation and reverses autoimmune diabetes in nonobese diabetic mice. Diabetes 58: 2277–2284.