Prevention and Reversal of Type 1DiabetesdPast Challenges andFuture OpportunitiesDiabetes Care 2015;38:997–1007 | DOI: 10.2337/dc15-0349

Over the past three decades there have been a number of clinical trials directed atinterdicting the type 1 diabetes (T1D) disease process in an attempt to prevent thedevelopment of the disease in those at increased risk or to stabilizedpotentiallyeven reversedthe disease in people with T1D, usually of recent onset. Unfortu-nately, to date there has been no prevention trial that has resulted in delay orprevention of T1D. And, trials in people with T1D have had mixed results withsome showing promise with at least transient improvement in b-cell functioncompared with randomized control groups, while others have failed to slow thedecline in b-cell function when compared with placebo. This Perspective willassess the past and present challenges in this effort and provide an outline forpotential future opportunities.

THE BEGINNINGS

The first randomized, double-masked, controlled trials with sufficient statistical power togive confidence for the outcome were conducted in the mid-1980s with cyclosporine(1,2). Two large studies were conducteddthe French cyclosporine study that included122patientswith type1diabetes (T1D) aged 15–40 yearswhohadbeen symptomatic for6 months or less and were on insulin therapy for 2 months or less (1) and the Canadian-European cyclosporine study that included 188 subjects aged 9–35 years who had beensymptomatic for 14 weeks or less and were on insulin therapy for 6 weeks or less (2).Both trials used as their primary outcome the achievement of remission defined twoways. First, “complete remission”was defined as goodmetabolic control (fasting glucose,140 mg/dL [7.8 mmol/L], postprandial glucose ,200 mg/dL [11.1 mmol/L], HbA1c#7.5%) in the absence of insulin treatment. Second, “partial remission” was definedas goodmetabolic controlwith insulin dose,0.26 units/kg per day. Both studies found agreater proportion of cyclosporine patients than placebo patients achieving and main-taining remissions. In both studies, cyclosporine could be given for 1 year, with stoppingrules in place that led to blinded substitution of placebo for cyclosporine if remissionwaslost. Although both studies found that cyclosporine had superior efficacy than placebo,the magnitude and duration of benefit did not appear sufficient to justify cyclosporinetreatment in clinical practice, given the potential of cyclosporine-induced nephrotoxicity.The importance of the studies, however, was that they demonstrated the impact ofimmune intervention on the evolution of T1D, in a sense fulfilling Koch postulates, whichwere developed for infectious diseases, but the response to immune therapy can beconsidered as indicative that T1D is immune mediated.

CHALLENGES COMPLICATING PREVIOUS STUDIES

Since the completion of the early trials, particularly during the past decade, a num-ber of additional randomized, double-masked, adequately powered, controlled

Diabetes Research Institute, University of MiamiMiller School of Medicine, Miami, FL

Corresponding author: Jay S. Skyler, jskyler@miami.edu.

Received 16 February 2015 and accepted 17March 2015.

© 2015 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered.

See accompanying articles, pp. 968,971, 979, 989, 1008, 1016, 1030,and 1036.

Jay S. Skyler

Diabetes Care Volume 38, June 2015 997 TYPE1DIABETES

ATACROSSR

OADS

clinical trials have been conducted usingmany different immunological strate-gies. For the most part, these have beendisappointing, with none showing unam-biguous benefit in preserving b-cell func-tion. Some studies have shown transientbenefit with anti-CD3 monoclonal anti-bodies targeting T cells (Fig. 1A and B)(3–6), an anti-CD20 monoclonal antibody(i.e., rituximab) targeting B cells (Fig. 1C)(7,8), and with a costimulation blockingagent (i.e., abatacept) that prevents im-mune activation (Fig. 1D) (9,10). One pilotstudy combining antithymocyte globulin(ATG) and granulocyte colony-stimulatingfactor (GCSF) suggestedbenefit (Fig. 1E) butawaits confirmation in a larger trial (11).

A number of other strategies havebeen without benefit in preservingb-cell function (12–17). In addition,some studies had ambiguous effects,not meeting the primary outcome mea-sure (i.e., preservation ofb-cell function)but showing potential benefit either onsecondary outcome measures or mech-anistic measures or in only a subgroup ofsubjects (18–20).

Thus, most immune intervention trialsin T1D have either failed to achieve suc-cess in preserving b-cell function orhave met that hurdle but have nonethe-less shown only a transient effect. Thishas resulted in a flurry of editorial andcommentary articles in peer-reviewed

journals that take a bleak look at the field.Rather than looking negatively on the re-sults to date, an opportunity exists to ex-amine the details of previous studies toidentify what can be learned and whatcan be applied to future studies. To thatend, several problems are notable, andthese are detailed below.

MISLEADING PILOT STUDIES

The Diabetes Prevention Trial–Type 1Parenteral Insulin Trial was a fully pow-ered, randomized, controlled clinical tri-al that enrolled 339 relatives of patientswith T1Dwhowere estimated to have atleast a 50% risk of developing T1D inthe next 5 years (21). To enroll these

Figure 1—There has been a progressive decline in b-cell function, as measured by C-peptide, even in most studies that have been “successful,” thusshowing only a transient benefit, as depicted for the anti-CD3 monoclonal antibody teplizumab (A) (3,4), the anti-CD3 monoclonal antibodyotelixizumab (B) (5,6), the anti-CD20 monoclonal antibody rituximab (C) (7,8), and the costimulation blocker abatacept (D) (9,10). E: Also shownis the preservation of b-cell function in a pilot study (only 25 subjects randomized) with the combination of low-dose ATG plus GCSF (11), in whichthere is preservation of b-cell function at 1 year, something needing confirmation in a larger study.

998 Prevention and Reversal of T1D Diabetes Care Volume 38, June 2015

subjects, more than 93,000 relativeswere screened. The trial showed no dif-ference in the rate of development ofT1D (Fig. 2A) (21). This was a surprisingoutcome, as two pilot studies had sug-gested that the interventiondlow-doseinsulindcould delay the disease. One ofthose apparently promising pilot studieswas a nonrandomized study of 12 indi-viduals offered the intervention, 7 ofwhom declined and were analyzed as acomparison group, alongwith a group ofhistorical control subjects of undefinednumber (Fig. 2B) (22). Life-table analysissuggested a statistically significant (P =0.002) and dramatic difference in rate ofdevelopment of T1D. Interestingly, the

treated group was younger (mean age10 years) than the untreated group(mean age 20.7 years), and one wouldhave expected younger individuals toprogress faster, which did not happen.In the other pilot trial, 14 subjects wererandomized to insulin treatment or tothe control group, with the insulin grouphaving longer diabetes-free survival(P , 0.03) (23).

The first study in T1D using GAD in avaccine formulation with alum (GAD-alum) randomized 70 subjects, aged10–18 years, within 18 months of diag-nosis, to receive either two doses ofGAD-alum or two doses of alum alone(24). The primary outcome measure,

fasting C-peptide at 15 months,showed no difference between groups.However, mixed-meal tolerance test(MMTT)-stimulated C-peptide washigher in GAD subjects whowere treatedwithin 6 months of diagnosis, butthis included only 11 subjects treatedwith GAD and 14 treated with placebo(Fig. 2C). On the basis of this secondarysubgroup outcome, three additionalstudiesdall enrolling subjects within 3months of diagnosisdwere undertaken,including two phase 3 trials. A TrialNetstudy enrolled 145 subjects aged 3–45years (Fig. 2D), a European phase 3 trialenrolled 334 subjects aged 10–20 years(Fig. 2E), and a U.S. phase 3 trial enrolled

Figure 2—A: Life-table analysis showing lack of benefit in the fully powered DPT-1 Parenteral Insulin Trial (21), despite perceived benefit (by life-tableanalysis) in a small pilot study (B) (22). C: Putative benefit in a small subgroup (those treatedwithin 6months of diagnosis) in a GAD-alum vaccine trial(24) that was not confirmed in two larger trials (D and E) with GAD-vaccine (12,13).

care.diabetesjournals.org Skyler 999

328 subjects aged 10–20 years (12,13,25).In all three of these trials, there was noevidence of a treatment effect withGAD-alum. Yet, it should be appreciatedthat the success of GAD in mice wasapproached differently. For example,the GAD-alum studies perhaps used thewrong formulation (with adjuvant). Thestudies could have used GAD-alum bythe wrong route (subcutaneous) and atthe wrong time (after clinical diagnosisof T1D). Although GAD vaccine still maybe able to prevent or delay the develop-ment of T1D if used as a vaccine prior todisease onset, the results of these trialshave greatly diminished the enthusiasmfor GAD that once existed in our field.Another approach considered was

DiaPep277. DiaPep277 is a peptide de-rived from positions 437–460 of the hu-man heat shock protein 60 (Hsp60),named peptide 277, that is alleged to in-duce anti-inflammatory T cells. The firstreport of its use in T1D was a pilot studyin which 35 subjects, aged 16–55 years,with recent onset T1D, received eitherthree doses of DiaPep277 or three dosesof placebo (26). Primary outcomewas glucagon-stimulated C-peptide at10 months, which was higher in theDiaPep277groupthan in theplacebogroup,an effect that was sustained with follow-up to 18 months (27). Four additionalphase 2 studies showed results thatwere ambiguous at best, with no clearbenefit (28–30). Nonetheless, the spon-sormounted two full-scale phase 3 trials.The first enrolled 457 subjects, aged 16–45 years, with recent-onset T1D, who re-ceived injections of DiaPep277 or placeboquarterly for 2 years (14,15). The initial

report alleged that glucagon-stimulatedtest (GST) of C-peptide at 24 months wasimproved in the DiaPep277 group versusthe placebo group, although there was nodifference in MMTT-stimulated C-peptide.However, MMTT was the original primaryoutcomemeasure and therefore was mea-sured at randomization (month 0) and after6, 12, 18, and 24 monthsda total of 5measurements. As GST originally was a sec-ondary outcome measure, it was per-formed at month 1 (defined as “baseline”for theGSTbut 1monthafter thefirst treat-ment had been given) and at 12 and 24monthsda total of 3 measurements. Theauthors intended, initially, to have MMTTbe the primary outcome measuredtheyperformed the first MMTT before initiatingtreatment (a true baseline measurement)and conducted the test at more frequentintervals. However, the primary outcomemeasure was changed from the MMTT tothe GST. Specifically, it was stated “thestudy protocol was amended and the Sta-tistical Analysis Plan was planned and final-ized before the study was unblinded, withthe GST clearly defined as the primary end-point” (14). Had that really been the case, itraisedquestions ofwhydifferences in thesetwo outcome measures existed and led tothe question of whether both measureswould be needed in future trials (31). How-ever, subsequently thearticlewas retractedwith a statement that there had been evi-dence uncovered that some employees ofthe sponsorhadengaged in seriousmiscon-duct, including collusion with a third-partybiostatistics firm to improperly receive un-blinded trial data and to use such data inorder to manipulate the analyses toobtain a favorable result.

Another pilot study involved the use ofBacillus Calmette-Guerin (BCG) vaccine insix subjects with long-standing T1D whowere randomized to receive either twodoses of BCG or placebo (32). The authorsclaimed that the BCG subjects, and one ofthe three control subjects who devel-oped acute Epstein-Barr virus (EBV) in-fection, showed increases in deadinsulin-autoreactive T cells and inductionof regulatory T cells. They also claimedthat there was transient increase inC-peptide levelsnot seen inanonrandom-ized cohort of subjects with T1D in theirinstitution. They have garnered a lot ofpublicity for this very tiny study. Althoughthe authors reported an increase inC-peptide levels using an ultrasensitiveassay, there are no data that the trivialincreases in C-peptide have any biologicalimportance.Moreover, two larger previousrandomized clinical trials of BCG, involving26 and 47 subjects, showed no effect ofBCG on preservation of b-cell function,but in both trials there was a trend togreater decline of b-cell function in the BCGgroup than in the control group (33,34).

What can one then conclude aboutthe value of pilot studies in this area?Essentially, the bottom line is that pilotstudies must be viewed with great cau-tion and definitive answers can only beobtained by adequately powered ran-domized controlled trials.

FAILURE TO DISTINGUISHTRANSIENT SYMPTOMS FROMSERIOUS ADVERSE EVENTS

A number of interventions used inrecent-onset T1D have shown transientsymptoms related to cytokine release

Figure 3—A: Transient effect of the anti-CD3 monoclonal antibody teplizumab in all treated subjects in one study (62). B: In that same study,retention of b-cell function for 2 years in responders, whereas nonresponders were identical to the comparison control group.

1000 Prevention and Reversal of T1D Diabetes Care Volume 38, June 2015

Table 1—Immune intervention trials in T1D

Study name Intervention Outcome ResultYear

reported Reference

Primary preventionstudies

Finnish TRIGR pilot Casein hydrolysate formula Autoantibodies Apparentbenefit

2010 66

TRIGR Casein hydrolysate formula Autoantibodies No difference 2014 67FINDIA Insulin-free whey-based formula Autoantibodies Apparent

benefit2012 68

BABYDIET Gluten-free diet Autoantibodies No difference 2011 69TRIGR Casein hydrolysate formula Diagnosis of T1D * Ongoing 67

Secondary preventionstudies

DENIS Nicotinamide Diagnosis of T1D No difference 1998 70ENDIT Nicotinamide Diagnosis of T1D No difference 2004 71DPT-1 Parenteral

Insulin Injected insulin Diagnosis of T1D No difference 2002 21DPT-1 Oral Insulin Oral insulin Diagnosis of T1D No difference 2005 20Belgian parenteral

insulin Injected insulin Diagnosis of T1D No difference 2009 51DIPP birth cohort Nasal insulin Diagnosis of T1D No difference 2008 52DIPP sibling cohort Nasal insulin Diagnosis of T1D No difference 2008 52INIT II Nasal insulin Diagnosis of T1D * Ongoing 60DIAPREV-IT GAD Diagnosis of T1D * Ongoing 58TrialNet oral insulin Oral insulin Diagnosis of T1D * Ongoing 59TrialNet teplizumab Anti-CD3, teplizumab Diagnosis of T1D * Ongoing 72TrialNet abatacept Abatacept Diagnosis of T1D * Ongoing 73

Studies in recent-onsetT1D

French cyclosporine Cyclosporine Remission Benefit 1986 1Canadian-European

cyclosporine CyclosporineC-pep $0.6 nmol/L or noninsulin

treated Benefit 1988 2Azathioprine +

glucocorticoidsAzathioprine

and prednisone Peak C-pep/glucose ratio Benefit 1988 74Azathioprine, adults Azathioprine Remission Benefit 1985 75Azathioprine, children Azathioprine Partial remission No difference 1989 76Linomide French trial Linomide Glucagon-stimulated C-pep Benefit 1998 77BCG BCG vaccine Glucagon-stimulated C-pep No difference 1998 33BCG BCG vaccine Primary, remission; secondary,

MMTT C-pepNo difference 1999 34

French oral insulin Oral insulin Glucagon-stimulated C-pep No difference 2000 53Italian oral insulin Oral insulin Fasting C-pep No difference 2000 54U.S. oral insulin Oral insulin Loss of C-pep # 2004 55Herold anti-CD3 Teplizumab MMTT C-pep Benefit 2002, 2005 3,4Keymeulen anti-CD3 Otelixizumab C-pep after clamp Benefit 2005, 2010 5,6Protege Teplizumab Insulin ,0.5 unit/kg + HbA1c

,6.5%No difference 2011, 2013 35,36

Protege Encore Teplizumab Insulin ,0.5 unit/kg + HbA1c,6.5%

* 2011 78

DEFEND-1 Otelixizumab MMTT C-pep No difference 2014 37DEFEND-2 Otelixizumab MMTT C-pep No difference 2014 38AbATE (ITN study) Teplizumab MMTT C-pep Benefit 2013 62DELAY Teplizumab MMTT C-pep Benefit 2013 79GAD pilot GAD-alum vaccine Fasting C-pep Apparent

benefitin secondaryoutcome insubgroup

2008 24

GAD TrialNet GAD-alum vaccine MMTT C-pep No difference 2011 12GAD Europe GAD-alum vaccine MMTT C-pep No difference 2012 13GAD U.S. (DiaPrevent) GAD-alum vaccine MMTT C-pep N/A 2011 25DiaPep–Israeli adults DiaPep277 peptide Glucagon-stimulated C-pep Benefit 2001, 2007 26,27DiaPep–Israeli

pediatrics DiaPep277 peptide MMTT C-pep No difference 2007 28

Continued on p. 1002

care.diabetesjournals.org Skyler 1001

syndrome, including headache, fever,and hypotension, at the time of infusionof the treatment. These symptoms arenot intolerable and are both transientand fully reversible. Thus, they do notconstitute major adverse effects thatwould justify withdrawal of the sub-ject from the study. Cytokine releasesyndrome has been seen with infusionsof anti-CD3 monoclonal antibodies (3,5),

rituximab (7), and thymoglobulin(11,18). Yet, although the infusionshave been completed within the firstdays or within the first month afterrandomization, beneficial effects fromthese interventions have been seen12 months to 4 years after randomiza-tion (3–7,11). Thus, subjects may havesome discomfort, but there needs tobe a reality check as to the potential

benefit versus the discomfort. A fewdays of symptoms early on need to bebalanced with a sustained beneficialoutcome over a protracted time frame.

Some of these trials have used anti-histamine and analgesic prophylaxis toobviate symptoms related to cytokinerelease. At least one trial also usedlow doses of glucocorticoids (18) butonly in the experimental group, not in

Table 1—Continued

Study name Intervention Outcome ResultYear

reported Reference

DiaPep–Belgian adults DiaPep277 peptide Glucagon-stimulated C-pep Benefit at onedose

2007 29

DiaPep–Europe adults DiaPep277 peptide Glucagon-stimulated C-pep No difference 2007 30DiaPep–Europe

pediatrics DiaPep277 peptide Glucagon-stimulated C-pep No difference 2007 30DiaPep–phase III DiaPep277 peptide Glucagon-stimulated C-pep ♓ 2014 14,15MMF/DZB Mycophenolate mofetil

with/without daclizumabMMTT C-pep No difference 2010 16

Anti-CD20 TrialNet Anti-CD20 rituximab MMTT C-pep Benefit 2009, 2014 7,8Abatacept TrialNet Abatacept MMTT C-pep Benefit 2011, 2014 9,10Canakinumab TrialNet Anti-IL1b canakinumab MMTT C-pep No difference 2013 17START thymoglobulin

ITN Thymoglobulin MMTT C-pep No difference 2013 18T1DAL–alefacept ITN Alefacept MMTT C-pep No difference 2013 19IL-2 & rapamycin

safety ITN IL-2 and rapamycin MMTT C-pepTransientworsening 2012 45

AIDA anakinra trial Anakinra MMTT C-pep No difference 2013 17a1-Antitrypsin a1-Antitrypsin MMTT C-pep § 2014 80Altered peptide ligand B9–23 altered peptide ligand MMTT C-pep No difference 2009 56Plasmid-encoded

proinsulinPlasmid-encoded

proinsulin Safety MMTT C-pep ¶ 2013 57Proinsulin peptide Proinsulin peptide Safety study No safety issues 2009 61ATG–GCSF trial ATG and GCSF MMTT C-pep Benefit 2015 11DIATOR Atorvastatin MMTT C-pep No difference 2011 81Etanercept Etanercept MMTT C-pep Benefit 2009 82Low-dose IL-2 safety

trial IL-2 (3 doses) T-reg number Increased 2013 46REPAIR-T1D Sitagliptin and lansoprazole MMTT C-pep No difference 2014 48AHSCT + profound

immunosuppressionCyclophosphamide,GCSF, ATG, AHSCT MMTT C-pep Benefit

2007, 2009,2009 83–85

AHSCT + profoundimmunosuppression

Cyclophosphamide,GCSF, ATG, AHSCT MMTT C-pep Benefit 2014 86

ATG–GCSF trial ATG and GCSF MMTT C-pep * Ongoing 87EXTEND trial Tocilizumab MMTT C-pep * Ongoing 88Otelixizumab

dose-ranging trial Otelixizumab MMTT C-pep * Ongoing 89a1-Antitrypsin trial a1-Antitrypsin Basal C-pep * Ongoing 90a1-Antitrypsin trial a1-Antitrypsin MMTT C-pep * Ongoing 91Ustekinumab pilot Ustekinumab Safety * Ongoing 92Imatinib trial Imatinib MMTT C-pep * Ongoing 93Tauroursodeoxycholic

acid trial Tauroursodeoxycholic acid MMTT C-pep * Ongoing 94DIABGAD GAD-alum& vitamin Dwith/without

ibuprofenMMTT C-pep * Ongoing 95

Proinsulin peptide Proinsulin peptide Safety * Ongoing 96Methyldopa Methyldopa Inhibition of DQ8 Ag * Ongoing 97Low-dose IL-2 IL-2 T-reg number * Ongoing 98

AHSCT, autologous hematopoietic stem cell transplantation; C-pep, C-peptide; INT, Immune Tolerance Network; T-reg, regulatory T cell. *Data notyet available. #Data ambiguousdauthors claim benefit but only seen in one dose in post hoc subgroup. N/Adactual data not available; press releaseannounced negative result and study discontinuation. ♓Article retracted. §Data ambiguousdauthors claim benefit but single-arm trial and“benefit” unclear. ¶Data ambiguous (as discussed in text).

1002 Prevention and Reversal of T1D Diabetes Care Volume 38, June 2015

the placebo group. One might considergiving the same prophylaxis to both ex-perimental and placebo groups in orderto minimize the risk of unblinding.

ERRORS IN TRIAL DESIGN

There have been extensive studies withtwo anti-CD3 monoclonal antibodiesdteplizumab and otelixizumab. The firststudy with teplizumab (given over 14days) demonstrated a slower decline ofb-cell function (by MMTT) at 1 year (3),with sustained improvement on b-cellfunction at 2 years (4). Meanwhile, thefirst study with otelixizumab (given over6 days) showed a slower decline of b-cellfunction (measured using a hyperglyce-mic clamp followed by glucagon stimula-tion) at 18 months (5). After 4 years offollow-up, although b-cell function wasnot measured, the otelixizumab grouphad lower insulin requirements despitesimilar glycemic control as measured byHbA1c (6). Thus, effects of a 6-day treat-ment course appeared to be evident4 years later.The results from these early studies

with anti-CD3 led to the initiation of phase3 clinical trials with both agents. Unfor-tunately, the phase 3 studies did notmeet their primary outcome criteria. Forteplizumab, the primary outcome wasthe combination of HbA1c ,6.5% and in-sulin dose ,0.5 units/kg/day (35). Thisoutcome measure was arbitrarily selectedwithout sufficient data to justify its selec-tion. Moreover, by using a composite

outcome that requires a subject to meettwo criteria, the outcome becomes adichotomous measure that dilutes theeffect of two continuous variablesdHbA1cand insulin dose. More important, whenthe conventional outcome measure ofC-peptidewasassessed, therewasevidenceof efficacy both at 1 year (35) and at 2 years(36) following two 14-day courses of tepli-zumab (at entry and at 26 weeks into thestudy). This was especially evident in sub-jects enrolled in the U.S., in younger sub-jects (age 8–17 years), in subjects enrolledwithin 6 weeks of diagnosis, and in sub-jects with higher levels of C-peptide at en-try (35). In addition, the phase 3 study alsoenrolled subjects in South Asia. Althoughthese subjectsmet the clinical criteria usedfor enrollment, it is important to note thattypical immune-mediated T1D (also calledtype 1A diabetes) is a disease principally ofEuropoid Caucasians. Enrollment of Asiansubjectsmay have confounded the results.Another issue was that the control groupin the phase 3 study maintained residualC-peptide to a greater extent and longerthan expected, which was especially truefor adult subjects. This made it more diffi-cult to detect differences between groups.

For otelixizumab, the phase 3 studiesused a dose that was one-sixteenth (to-tal of 3.1 mg over 8 days) of that used inthe original phase 2 study (total of 48mg), in an effort to avoid any side effects(37,38). One has to question what theinvestigators were trying to avoid. Theone noninfusion-related side effect seen

in the first trial with otelixizumab wastransient EBV reactivation (39). Al-though the authors concluded thatsuch EBV reactivation was of no appar-ent clinical concern over the long term,others have asserted that this must beavoided at all costs (40). I was Chair ofthe Data Safety Monitoring Committeefor that study, and prior to the studythe committee had concluded that tran-sient EBV reactivation was possible andwould neither constitute a reason to haltthe study nor was a side effect thatneeded to be avoided. So, in the phase3 trials with use of the lower dose, sideeffects were obviated; however, benefi-cial effects were also completely obvi-ated. This unfortunate dose reductionreminds us that all effective therapiesare likely to have some side effects andthat if one lowers the dose to eliminateall side effects, the drug may no longerhave benefit.

DOSING ISSUES

Getting the dose right is important. Inaddition to the otelixizumab dosingissue, several other examples are worthnoting. Interleukin-2 (IL-2) stimulatesregulatory T cells at low doses but alsostimulates effector T cells at higherdoses (41). Thus, it has been notedthat the use of IL-2 is a double-edgedsword (42). In other conditions (e.g.,hepatitis-C virus–induced vasculitisand graft-versus-host disease), low-dose IL-2 has shown beneficial effects(43,44). The first human study of IL-2in T1D used a relatively high dose andcombined its use with rapamycin (45).An adverse effect on b-cell function wasobserved. A subsequent study usinglow-dose IL-2 appeared to be relativelysafe, without a decrease in b-cellfunction and with the expected in-crease in regulatory T cells (46). There-fore, studies are currently under wayto study low-dose IL-2 in T1D.

Another example of a dosing issue isthymoglobulin (ATG). In a study using arelatively high dose of ATG, no beneficialeffect was seen, and it was observed thatthere was suppression of both effectorand regulatory T cells (18). In anotherstudy in which a lower dose of ATG wasused, regulatory T cells were not sup-pressed (11). Interpretation of that studyis confounded, however, because low-dose ATG was used in combination withGCSF.

Figure 4—Potential scheme of combination therapy using agents with complementary effects.This scheme includes anti-inflammatory therapy targeting innate immunity, immunomodulatorytherapy targeting adaptive immunity, therapy driving regulatory immunity, antigen-based ther-apy directing regulation to b-cells, and an agent promoting b-cell health. T-reg, regulatory T cell.

care.diabetesjournals.org Skyler 1003

A related issue providing an addi-tional confounder is choosing surrogateagents. A study in NOD mice found thatcombination therapy with glucagon-likepeptide 1 (GLP-1) and gastrin restoresnormoglycemia (47). This combinationresulted in increases in pancreatic insu-lin content, b-cell mass, b-cell prolifera-tion, and b-cell neogenesis; a reductionin b-cell apoptosis; and a beneficial ef-fect on the immune response (47). Itseemed like a natural combination totest in human beings. Yet, rather thantesting the combination of GLP-1 and gas-trin, a study evaluated the combination ofthe dipeptidyl peptidase-4 inhibitor sita-gliptin and theproton-pump inhibitor lan-soprazole, as these agents, respectively,increase circulating levels of GLP-1 andgastrin (48). The primary end point wasnot achieved, but not all participantshad the expected increases in GLP-1and gastrin levels. Thus, they effectivelydid not get the doses that may havebeen needed to achieve the desiredeffects.

ANTIGEN-BASED THERAPIES HAVENOT WORKED YET

The desirability of antigen-basedimmunotherapy is grounded on the no-tion that such therapies are specific forT1D and are unlikely to have adverse off-target effects. Thus, they should have ahigh degree of safety. The two diabetes-specific antigens that have been usedare insulin and GAD, both of which hadgreat success in animal models of T1D(49,50). Yet, to date, there has been nounambiguous success with antigen-based therapies in human T1D. GAD-alum vaccine has failed in new-onsetT1D (12,13). Injected insulin has failedin prevention trials (21,51). Nasal insulinhas failed in prevention trials (52). Oralinsulin has failed in new-onset T1D (53–55) and did not meet its primaryoutcome in a prevention trial (20). Analtered peptide ligand of insulin failed innew-onset T1D (56). A plasmid-encodingproinsulin was claimed to have a benefiton b-cell function, but actually the ef-fect was seen in but one of four dosestested and only at one time point (57).Moreover, the number of subjects stud-ied at each dose was small, and there didnot appear to be statistical adjustmentfor multiple comparisons. The field is ea-gerly awaiting the conclusion of ongoingGAD-alum (58), oral insulin (59), and

nasal insulin (60) prevention trials andof the development of an approachthat uses a combination of peptidesderived from proinsulin (61). Yet, thedesirable concept of an antigen-basedtherapy sustains the efforts to findan effective one, perhaps as a compo-nent of a combination therapeuticapproach.

NOT ALL PEOPLE MAY RESPONDTO THERAPY

In one study using the anti-CD3 mono-clonal antibody teplizumab, a group of“responders” to treatment was identi-fied, who at 2 years maintainedC-peptide better than the randomizedbut untreated comparison group (Fig.3) (62). In that study, responders consti-tuted 48% of subjects treated with tepli-zumab. Interestingly, the respondersnot only did better than the comparisongroup but also, as a group, actuallymaintained b-cell function essentiallyat the level seen at randomization,whereas the nonresponders had lostb-cell function at a rate similar to thecomparison group. Had the analysisbeen confined to the total treatedgroupdincluding both responders andnonrespondersdthe full retention ofb-cell function in nearly half of the sub-jects would have been missed. Sometype of responder analysis should be ap-plied to all intervention studies. Indeed,it has been suggested that a responderanalysis be included in the statisticalplan for all T1D intervention studies,and for that purpose, the definitionof a responder should be the mainte-nance of 100% of baseline b-cell func-tion (63).

A fundamental question is whysome subjects fail to respond. It couldbe that the immunotherapy used wasineffective (at least at the dosetested), that the immunologicalprocessdperhaps a relapsing and re-mitting onedwas in a latent period atthe time of drug administration andthus not responsive to immunother-apy, that b-cell mass or b-cell functionhad already deteriorated to a point ofno return, that the immunological pro-cesses damaging b-cells are differentamong individuals, or for some otherreason. It is important to assess poten-tial biomarkers that might discrimi-nate responders from nonrespondersand thus might be used as enrollment

criteria for future use of a giventherapy.

FUTURE DIRECTIONS

As discussed above, there have beenmany attempts at immune interventionin T1D, with mixed results. Table 1provides a listing of all of themajor stud-ies to date, including randomized con-trolled trials and other studies mentionedin this Perspective; not included aresmall pilot studies not discussed in thisarticle.

It should be appreciated that in theevolution of T1D several immune path-ways are involved. This complicates thedesign of an ideal therapeutic strategyto control the immune system and pre-vent the loss ofb-cell function andb-cellmass. Indeed, if one pathway is con-trolled, another pathway may becomemore active. Thus, success may requirethat a combination approach be used.Such a combination might include(Fig. 4) one or more anti-inflammatoryagents targeting innate immunity,such as agents that target IL-1 (IL-1b) or tumor necrosis factor (TNF-a);one or more immunomodulatory agentstargeting adaptive immunity, such asanti-CD3, anti-CD20, or costimulationblockade (three agents that alreadyhave shown some beneficial effect onpreserving b-cell function) or ATG; per-haps some agent that would drive regu-latory T cells, such as low-dose IL-2 orGCSF or the infusion of regulatory T cellsthemselves; any of these perhaps cou-pled with a diabetes-related antigenthat might help target the regulatory Tcells to b-cells; and agents that help pre-serve b-cell health, such as GLP-1 (64).In such a strategy, the combination ofagents is selected for having potentialcomplementary effects. It may also benecessary to tailor the selection ofagents for different individuals, i.e., wemay need to move to a personalizedmedicine approach with different treat-ments for different subtypes, if thesebecome better defined (65).

Therefore as outlined, there are manypotential interventions that hold prom-ise, particularly if they are used as com-ponents of combination therapy. Severalnew strategies are approaching clinicalevaluation. To be successful, we mustbe patient, yet proceed with diligence.Moreover, it is important that trials becarefully designed, well controlled, and

1004 Prevention and Reversal of T1D Diabetes Care Volume 38, June 2015

have adequate sample size to assurevalid interpretation.

Duality of Interest. No potential conflicts ofinterest relevant to this article were reported.

References1. Feutren G, Papoz L, Assan R, et al.; Cyclo-sporin/Diabetes French Study Group. Cyclo-sporin increases the rate and length ofremissions in insulin-dependent diabetes of re-cent onset. Results of a multicentre double-blind trial. Lancet 1986;2:119–1242. The Canadian-European Randomized ControlTrial Group. Cyclosporin-induced remission ofIDDM after early intervention. Association of 1yr of cyclosporin treatment with enhanced insulinsecretion. Diabetes 1988;37:1574–15823. Herold KC, Hagopian W, Auger JA, et al. Anti-CD3 monoclonal antibody in new-onset type 1 di-abetesmellitus. N Engl JMed 2002;346:1692–16984. Herold KC, Gitelman SE, Masharani U, et al. Asingle course of anti-CD3 monoclonal antibodyhOKT3gamma1(Ala-Ala) results in improvementin C-peptide responses and clinical parametersfor at least 2 years after onset of type 1 diabe-tes. Diabetes 2005;54:1763–17695. Keymeulen B, Vandemeulebroucke E, ZieglerAG, et al. Insulin needs after CD3-antibody ther-apy in new-onset type 1 diabetes. N Engl J Med2005;352:2598–26086. Keymeulen B, Walter M, Mathieu C, et al.Four-year metabolic outcome of a randomisedcontrolled CD3-antibody trial in recent-onsettype 1 diabetic patients depends on their ageand baseline residual beta cell mass. Diabetolo-gia 2010;53:614–6237. Pescovitz MD, Greenbaum CJ, Krause-Steinrauf H, et al.; Type 1 Diabetes TrialNetAnti-CD20 StudyGroup. Rituximab, B-lymphocytedepletion, and preservation of beta-cell func-tion. N Engl J Med 2009;361:2143–21528. PescovitzMD, Greenbaum CJ, Bundy B, et al.;Type 1 Diabetes TrialNet Anti-CD20 StudyGroup. B-lymphocyte depletion with rituximaband b-cell function: two-year results. DiabetesCare 2014;37:453–4599. Orban T, Bundy B, Becker DJ, et al.; Type 1Diabetes TrialNet Abatacept Study Group. Co-stimulation modulation with abatacept inpatients with recent-onset type 1 diabetes:a randomised, double-blind, placebo-controlledtrial. Lancet 2011;378:412–41910. Orban T, Bundy B, Becker DJ, et al.; Type 1Diabetes TrialNet Abatacept Study Group.Costimulation modulation with abatacept in pa-tients with recent-onset type 1 diabetes: follow-up 1 year after cessation of treatment. DiabetesCare 2014;37:1069–107511. Haller MJ, Gitelman SE, Gottlieb PA, et al.Anti-thymocyte globulin/G-CSF treatment pre-serves b cell function in patients with estab-lished type 1 diabetes. J Clin Invest 2015;125:448–45512. Wherrett DK, Bundy B, Becker DJ, et al.;Type 1 Diabetes TrialNet GAD Study Group.Antigen-based therapy with glutamic aciddecarboxylase (GAD) vaccine in patients withrecent-onset type 1 diabetes: a randomiseddouble-blind trial. Lancet 2011;378:319–327

13. Ludvigsson J, Krisky D, Casas R, et al. GAD65antigen therapy in recently diagnosed type 1diabetes mellitus. N Engl J Med 2012;366:433–44214. Raz I, Ziegler AG, Linn T, et al.; DIA-AID 1Writing Group. Treatment of recent-onsettype 1 diabetic patients with DiaPep277: re-sults of a double-blind, placebo-controlled,randomized phase 3 trial [retracted in: Diabe-tes Care 2015;38:178]. Diabetes Care 2014;37:1392–140015. Pozzilli P, Raz I, Peled D, et al. Evaluation oflong-term treatment effect in a type 1 diabetesintervention trial: differences after stimulationwith glucagon or a mixed meal [retracted in:Diabetes Care 2015;38:179]. Diabetes Care2014;37:1384–139116. Gottlieb PA, Quinlan S, Krause-Steinrauf H,et al.; Type 1 Diabetes TrialNet MMF/DZB StudyGroup. Failure to preserve beta-cell functionwith mycophenolate mofetil and daclizumabcombined therapy in patients with new-onsettype 1 diabetes. Diabetes Care 2010;33:826–83217. Moran A, Bundy B, Becker DJ, et al.; Type 1Diabetes TrialNet Canakinumab Study Group;AIDA Study Group. Interleukin-1 antagonism intype 1 diabetes of recent onset: two multi-centre, randomised, double-blind, placebo-controlled trials. Lancet 2013;381:1905–191518. Gitelman SE, Gottlieb PA, Rigby MR, et al.;START Study Team. Antithymocyte globulin ther-apy for patients with recent-onset type 1 diabe-tes: a randomized double-blind phase 2 trial.Lancet Diabetes Endocrinol 2013;1:306–31619. Rigby MR, DiMeglio LA, Rendell MS, et al.;T1DAL Study Team. Targeting of memory T cellswith alefacept in new-onset type 1 diabetes(T1DAL study): 12 month results of a rando-mised, double-blind, placebo-controlled phase2 trial. Lancet Diabetes Endocrinol 2013;1:284–29420. Skyler JS, Krischer JP, Wolfsdorf J, et al.; Di-abetes Prevention Trial–Type 1 Diabetes StudyGroup. Effects of oral insulin in relatives of pa-tients with type 1 diabetes mellitus. DiabetesCare 2005;28:1068–107621. Diabetes Prevention Trial–Type 1 DiabetesStudy Group. Effects of insulin in relatives ofpatients with type 1 diabetes mellitus. N Engl JMed 2002;346:1685–169122. Keller RJ, Eisenbarth GS, Jackson RA. Insulinprophylaxis in individuals at high risk of type Idiabetes. Lancet 1993;341:927–92823. Fuchtenbusch M, Rabl W, Grassl B,Bachmann W, Standl E, Ziegler AG. Delay oftype I diabetes in high risk, first degree relativesby parenteral antigen administration: theSchwabing Insulin Prophylaxis Pilot Trial. Diabe-tologia 1998;41:536–54124. Ludvigsson J, Faresjo M, Hjorth M, et al.GAD treatment and insulin secretion in recent-onset type 1 diabetes. N Engl J Med 2008;359:1909–192025. Diamyd initiates closure of US phase IIIstudy [press release]. Stockholm, Sweden: Dia-myd Medical AB; June 23, 2011. Available fromhttp://www.diamyd.com/docs/pressClip.aspx?section=investor&ClipID=58443. Accessed 15February 201526. Raz I, Elias D, Avron A, Tamir M, Metzger M,Cohen IR. b-Cell function in new-onset type 1

diabetes and immunomodulation with a heat-shock protein peptide (DiaPep277): a rando-mised, double-blind, phase II trial. Lancet2001;358:1749–175327. Raz I, Avron A, Tamir M, et al. Treatmentof new-onset type 1 diabetes with peptideDiaPep277 is safe and associated with pre-served beta-cell function: extension of a ran-domized, double-blind, phase II trial. DiabetesMetab Res Rev 2007;23:292–29828. Lazar L, Ofan R, Weintrob N, et al. Heat-shock protein peptide DiaPep277 treatment inchildren with newly diagnosed type 1 diabetes:a randomised, double-blind phase II study. Di-abetes Metab Res Rev 2007;23:286–29129. Huurman VA, Decochez K, Mathieu C,Cohen IR, Roep BO. Therapy with the hsp60peptide DiaPep277 in C-peptide positive type1 diabetes patients. Diabetes Metab Res Rev2007;23:269–27530. Schloot NC, Meierhoff G, Lengyel C, et al.Effect of heat shock protein peptide DiaPep277on beta-cell function in paediatric and adult pa-tients with recent-onset diabetes mellitus type1: two prospective, randomized, double-blindphase II trials. Diabetes Metab Res Rev 2007;23:276–28531. Skyler JS. Struggles with clinical translationof immune intervention trials. Diabetes Care2014;37:1173–117532. Faustman DL, Wang L, Okubo Y, et al. Proof-of-concept, randomized, controlled clinical trialof Bacillus-Calmette-Guerin for treatment oflong-term type 1 diabetes. PLoS One 2012;7:e4175633. Elliott JF, Marlin KL, Couch RM. Effect ofbacille Calmette-Guerin vaccination on C-peptidesecretion in children newly diagnosed withIDDM. Diabetes Care 1998;21:1691–169334. Allen HF, Klingensmith GJ, Jensen P, SimoesE, HaywardA, ChaseHP. Effect of Bacillus Calmette-Guerin vaccination on new-onset type 1 diabe-tes. A randomized clinical study. Diabetes Care1999;22:1703–170735. Sherry N, Hagopian W, Ludvigsson J, et al.;Protege Trial Investigators. Teplizumab fortreatment of type 1 diabetes (Protege study):1-year results from a randomised, placebo-controlled trial. Lancet 2011;378:487–49736. Hagopian W, Ferry RJ Jr, Sherry N, et al.;Protege Trial Investigators. Teplizumab pre-serves C-peptide in recent-onset type 1 diabe-tes: two-year results from the randomized,placebo-controlled Protege trial. Diabetes2013;62:3901–390837. Aronson R, Gottlieb PA, Christiansen JS,et al.; DEFEND Investigator Group. Low-doseotelixizumab anti-CD3 monoclonal antibodyDEFEND-1 study: results of the randomizedphase III study in recent-onset human type1 diabetes. Diabetes Care 2014;37:2746–275438. Ambery P, Donner TW, Biswas N,Donaldson J, Parkin J, Dayan CM. Efficacy andsafety of low-dose otelixizumab anti-CD3monoclonal antibody in preserving C-peptidesecretion in adolescent type 1 diabetes:DEFEND-2, a randomized, placebo-controlled,double-blind, multi-centre study. Diabet Med2014;31:399–40239. Keymeulen B, Candon S, Fafi-Kremer S,et al. Transient Epstein-Barr virus reactivation

care.diabetesjournals.org Skyler 1005

in CD3 monoclonal antibody-treated patients.Blood 2010;115:1145–115540. Sprangers B, Van der Schueren B, Gillard P,Mathieu C. Otelixizumab in the treatment oftype 1 diabetes mellitus. Immunotherapy2011;3:1303–131641. Yu A, Snowhite I, Vendrame F, et al. Selec-tive IL-2 responsiveness of regulatory T cellsthroughmultiple intrinsic mechanisms supportsthe use of low-dose IL-2 therapy in type 1 di-abetes. Diabetes 2015;64:2172–218342. Bluestone JA. The yin and yang of interleukin-2-mediated immunotherapy. N Engl J Med 2011;365:2129–213143. Koreth J,MatsuokaK, KimHT, et al. Interleukin-2 and regulatory T cells in graft-versus-host disease.N Engl J Med 2011;365:2055–206644. Saadoun D, RosenzwajgM, Joly F, et al. Reg-ulatory T-cell responses to low-dose interleukin-2 in HCV-induced vasculitis. N Engl J Med 2011;365:2067–207745. Long SA, Rieck M, Sanda S, et al.; DiabetesTrialNet and the Immune Tolerance Network.Rapamycin/IL-2 combination therapy in pa-tients with type 1 diabetes augments Tregs yettransiently impairs b-cell function. Diabetes2012;61:2340–234846. Hartemann A, Bensimon G, Payan CA, et al.Low-dose interleukin 2 in patients with type 1diabetes: a phase 1/2 randomised, double-blind, placebo-controlled trial. Lancet DiabetesEndocrinol 2013;1:295–30547. Suarez-Pinzon WL, Power RF, Yan Y,Wasserfall C, Atkinson M, Rabinovitch A. Com-bination therapy with glucagon-like peptide-1and gastrin restores normoglycemia in diabeticNOD mice. Diabetes 2008;57:3281–328848. Griffin KJ, Thompson PA, Gottschalk M, KylloJH, Rabinovitch A. Combination therapy with si-tagliptin and lansoprazole in patientswith recent-onset type 1 diabetes (REPAIR-T1D): 12-monthresults of a multicentre, randomised, placebo-controlled, phase 2 trial. Lancet Diabetes Endo-crinol 2014;2:710–71849. Anderson MS, Bluestone JA. The NODmouse: a model of immune dysregulation.Annu Rev Immunol 2005;23:447–48550. Shoda LK, Young DL, Ramanujan S, et al. Acomprehensive review of interventions in theNOD mouse and implications for translation.Immunity 2005;23:115–12651. Vandemeulebroucke E, Gorus FK, DecochezK, et al.; Belgian Diabetes Registry. Insulin treat-ment in IA-2A-positive relatives of type 1 diabeticpatients. Diabetes Metab 2009;35:319–32752. Nanto-Salonen K, Kupila A, Simell S, et al.Nasal insulin to prevent type 1 diabetes in chil-dren with HLA genotypes and autoantibodiesconferring increased risk of disease: a double-blind, randomised controlled trial. Lancet 2008;372:1746–175553. Chaillous L, Lefevre H, Thivolet C, et al. Oralinsulin administration and residual beta-cellfunction in recent-onset type 1 diabetes: a mul-ticentre randomised controlled trial. Diabete In-suline Orale group. Lancet 2000;356:545–54954. Pozzilli P, Pitocco D, Visalli N, et al.; IMDIABGroup. No effect of oral insulin on residualbeta-cell function in recent-onset type I diabe-tes (the IMDIAB VII). Diabetologia 2000;43:1000–1004

55. Ergun-Longmire B, Marker J, Zeidler A, et al.Oral insulin therapy to prevent progression ofimmune-mediated (type 1) diabetes. Ann N YAcad Sci 2004;1029:260–27756. Walter M, Philotheou A, Bonnici F, ZieglerAG, Jimenez R.; NBI-6024 Study Group. No ef-fect of the altered peptide ligand NBI-6024 onbeta-cell residual function and insulin needs innew-onset type 1 diabetes. Diabetes Care 2009;32:2036–204057. Roep BO, Solvason N, Gottlieb PA, et al.Plasmid-encoded proinsulin preserves C-peptidewhile specifically reducing proinsulin-specificCD8+ T cells in type 1 diabetes. Sci Transl Med2013;5:191ra8258. Lund University. Diabetes Prevention–Immune Tolerance (DIAPREV-IT). In: ClinicalTrials.gov [Internet]. Bethesda, MD, National Libraryof Medicine, 2010. Available from https://clinicaltrials.gov/ct2/show/NCT01122446 NLMIdentifier: NCT01122446. Accessed 15 February201559. National Institute of Diabetes and Digestiveand Kidney Diseases. Oral insulin for preventionof diabetes in relatives at risk for type 1 diabetesmellitus. In: ClinicalTrials.gov [Internet]. Be-thesda, MD, National Library of Medicine,2007. Available from https://clinicaltrials.gov/ct2/show/NCT00419562 NLM Identifier:NCT00419562. Accessed 15 February 201560. Melbourne Health. Trial of intranasal insu-lin in children and young adults at risk of type 1diabetes (INIT II). In: ClinicalTrials.gov [Inter-net]. Bethesda, MD, National Library of Medi-cine, 2006. Available from https://clinicaltrials.gov/ct2/show/NCT00336674 NLM Identifier:NCT00336674. Accessed 15 February 201561. Thrower SL, James L, Hall W, et al. Proinsu-lin peptide immunotherapy in type 1 diabetes:report of a first-in-man phase I safety study. ClinExp Immunol 2009;155:156–16562. Herold KC, Gitelman SE, Ehlers MR, et al.;AbATE Study Team. Teplizumab (anti-CD3 mAb)treatment preserves C-peptide responses in pa-tientswithnew-onset type1diabetes in a random-ized controlled trial: metabolic and immunologicfeatures at baseline identify a subgroup ofresponders. Diabetes 2013;62:3766–377463. Beam CA, Gitelman SE, Palmer JP; Type 1Diabetes TrialNet Study Group. Recommenda-tions for the definition of clinical responder ininsulin preservation studies. Diabetes 2014;63:3120–312764. Skyler JS, Ricordi C. Stopping type 1 diabe-tes: attempts to prevent or cure type 1 diabetesin man. Diabetes 2011;60:1–865. Arif S, Leete P, Nguyen V, et al. Blood and isletphenotypes indicate immunological heterogeneityin type 1 diabetes. Diabetes 2014;63:3835–384566. KnipM, Virtanen SM, Seppa K, et al.; FinnishTRIGR Study Group. Dietary intervention in in-fancy and later signs of beta-cell autoimmunity.N Engl J Med 2010;363:1900–190867. Knip M, Akerblom HK, Becker D, et al.;TRIGR Study Group. Hydrolyzed infant formulaand early b-cell autoimmunity: a randomizedclinical trial. JAMA 2014;311:2279–228768. Vaarala O, Ilonen J, Ruohtula T, et al. Re-moval of bovine insulin from cow’s milk formulaand early initiation of beta-cell autoimmunity inthe FINDIA pilot study. Arch Pediatr AdolescMed 2012;166:608–614

69. Hummel S, PflugerM, HummelM, BonifacioE, Ziegler AG. Primary dietary intervention studyto reduce the risk of islet autoimmunity in chil-dren at increased risk for type 1 diabetes: theBABYDIET study. Diabetes Care 2011;34:1301–130570. Lampeter EF, Klinghammer A, ScherbaumWA, et al.; DENIS Group. The Deutsche Nicotin-amide Intervention Study: an attempt toprevent type 1 diabetes. Diabetes 1998;47:980–98471. Gale EA, Bingley PJ, Emmett CL, Collier T;European Nicotinamide Diabetes InterventionTrial (ENDIT) Group. European Nicotinamide Di-abetes Intervention Trial (ENDIT): a randomisedcontrolled trial of intervention before the onsetof type 1 diabetes. Lancet 2004;363:925–93172. National Institute of Diabetes and Digestiveand Kidney Diseases. Teplizumab for preventionof type 1 diabetes in relatives “at-risk.” In:ClinicalTrials.gov [Internet]. Bethesda, MD, Na-tional Library of Medicine, 2009. Available fromhttps://clinicaltrials.gov/ct2/show/NCT01030861NLM Identifier: NCT01030861. Accessed 15February 201573. National Institute of Diabetes and Digestiveand Kidney Diseases. CTLA4-Ig (abatacept) forprevention of abnormal glucose tolerance anddiabetes in relatives at-risk for type 1 diabetesmellitus. In: ClinicalTrials.gov [Internet]. Be-thesda, MD, National Library of Medicine,2010. Available from https://clinicaltrials.gov/ct2/show/NCT01773707 NLM Identifier:NCT01773707. Accessed 15 February 201574. Silverstein J, Maclaren N, Riley W, Spillar R,Radjenovic D, Johnson S. Immunosuppressionwith azathioprine and prednisone in recent-onset insulin-dependent diabetes mellitus. N EnglJ Med 1988;319:599–60475. Harrison LC, Colman PG, Dean B, Baxter R,Martin FI. Increase in remission rate in newlydiagnosed type I diabetic subjects treated withazathioprine. Diabetes 1985;34:1306–130876. Cook JJ, Hudson I, Harrison LC, et al. Double-blind controlled trial of azathioprine in childrenwith newly diagnosed type I diabetes. Diabetes1989;38:779–78377. Coutant R, Landais P, Rosilio M, et al. Lowdose linomide in type I juvenile diabetes of re-cent onset: a randomised placebo-controlleddouble blind trial. Diabetologia 1998;41:1040–104678. MacroGenics. Protege Encore study–clinicaltrial of teplizumab (MGA031) in children andadults with recent-onset type 1 diabetes melli-tus. In: ClinicalTrials.gov [Internet]. Bethesda,MD, National Library of Medicine, 2009. Avail-able from https://clinicaltrials.gov/ct2/show/NCT00920582 NLM Identifier: NCT00920582.Accessed 15 February 201579. Herold KC, Gitelman SE, Willi SM, et al. Te-plizumab treatment may improve C-peptide re-sponses in participants with type 1 diabetesafter the new-onset period: a randomised con-trolled trial. Diabetologia 2013;56:391–40080. Gottlieb PA, Alkanani AK, Michels AW, et al.a1-Antitrypsin therapy downregulates Toll-likereceptor-induced IL-1b responses in monocytesand myeloid dendritic cells and may improveislet function in recently diagnosed patientswith type 1 diabetes. J Clin Endocrinol Metab2014;99:E1418–E1426

1006 Prevention and Reversal of T1D Diabetes Care Volume 38, June 2015

81. Martin S, Herder C, Schloot NC, et al.;DIATOR Study Group. Residual beta cell func-tion in newly diagnosed type 1 diabetes aftertreatment with atorvastatin: the RandomizedDIATOR Trial. PLoS One 2011;6:e1755482. Mastrandrea L, Yu J, Behrens T, et al. Eta-nercept treatment in children with new-onsettype 1 diabetes: pilot randomized, placebo-controlled, double-blind study. Diabetes Care2009;32:1244–124983. Voltarelli JC, Couri CE, Stracieri AB, et al. Au-tologous nonmyeloablative hematopoietic stemcell transplantation in newly diagnosed type 1 di-abetes mellitus. JAMA 2007;297:1568–157684. Couri CE, Oliveira MC, Stracieri AB, et al.C-peptide levels and insulin independencefollowing autologous nonmyeloablative hema-topoietic stem cell transplantation in newlydiagnosed type 1 diabetes mellitus. JAMA2009;301:1573–157985. Voltarelli JC, Martinez ED, Burt RK. Autolo-gous nonmyeloablative hematopoietic stem celltransplantation in newly diagnosed type 1 dia-betes mellitus. JAMA 2009;302:624–62586. D’Addio F, Valderrama Vasquez A, Ben NasrM, et al. Autologous nonmyeloablative hemato-poietic stem cell transplantation in new-onsettype 1 diabetes: amulticenter analysis. Diabetes2014;63:3041–304687. National Institute of Diabetes and Digestiveand Kidney Diseases. ATG-GCSF in new onsettype 1 diabetes. In: ClinicalTrials.gov [Internet].Bethesda, MD, National Library of Medicine,2014. Available from https://clinicaltrials.gov/ct2/show/NCT02215200 NLM Identifier:NCT02215200. Accessed 15 February 201588. National Institute of Allergy and InfectiousDiseases. TCZ in new-onset type 1 diabetes. In:ClinicalTrials.gov [Internet]. Bethesda, MD, Na-tional Library of Medicine, 2014. Available fromhttps://clinicaltrials.gov/ct2/show/NCT02293837

NLM Identifier: NCT02293837. Accessed 15 Feb-ruary 201589. GlaxoSmithKline. Investigation of otelixizu-mab in new-onset, autoimmune type 1 diabetesmellitus patients. In: ClinicalTrials.gov [Inter-net]. Bethesda, MD, National Library of Medi-cine, 2013. Available from https://clinicaltrials.gov/ct2/show/NCT02000817 NLM Identifier:NCT02000817. Accessed 15 February 201590. Kamada, Ltd. Phase II-III study to evaluatethe efficacy and safety of alpha-1-anti-trypsin(Glassia) in type-1 diabetes. In: ClinicalTrials.gov [Internet]. Bethesda, MD, National Libraryof Medicine, 2013. Available from https://clinicaltrials.gov/ct2/show/NCT02005848 NLMIdentifier: NCT02005848. Accessed 15 February201591. Grifols Therapeutics Inc. Study of humanplasma-derived alpha1-proteinase inhibitor insubjects with new-onset type 1 diabetes melli-tus. In: ClinicalTrials.gov [Internet]. Bethesda,MD, National Library of Medicine, 2014. Avail-able from https://clinicaltrials.gov/ct2/show/NCT02093221 NLM Identifier: NCT02093221.Accessed 15 February 201592. University of British Columbia. Pilot clinicaltrial of ustekinumab in patients with new-onsetT1D. In: ClinicalTrials.gov [Internet]. Bethesda,MD, National Library of Medicine, 2014. Avail-able from https://clinicaltrials.gov/ct2/show/NCT02117765 NLM Identifier: NCT02117765.Accessed 15 February 201593. University of California, San Francisco. Im-atinib treatment in recent onset type 1 diabetesmellitus. In: ClinicalTrials.gov [Internet]. Be-thesda, MD, National Library of Medicine,2013. Available from https://clinicaltrials.gov/ct2/show/NCT01781975 NLM Identifier:NCT01781975 Accessed 15 February 201594. Goland, R. Tauroursodeoxycholic acid(TUDCA) in new-onset type 1 diabetes. In:

ClinicalTrials.gov [Internet]. Bethesda, MD,National Library of Medicine, 2014. Availablefrom https://clinicaltrials.gov/ct2/show/NCT02218619 NLM Identifier: NCT02218619.Accessed 15 February 201595. Ludvigsson, J. DIABGADdtrial to preserveinsulin secretion in type 1 diabetes using GAD-alum (Diamyd) in combination with vitamind and ibuprofen. In: ClinicalTrials.gov [Internet].Bethesda, MD, National Library of Medicine,2013. Available from https://clinicaltrials.gov/ct2/show/NCT01785108 NLM Identifier:NCT01785108. Accessed 15 February 201596. Cardiff University. Safety study to assesswhether proinsulin peptide injections can slowor stop the body damaging its own insulin-making cells in the pancreas in patients newlydiagnosed with type 1 diabetes (MonoPepT1De).In: ClinicalTrials.gov [Internet]. Bethesda, MD, Na-tional Library of Medicine, 2012. Available fromhttps://clinicaltrials.gov/ct2/show/NCT01536431NLM Identifier: NCT01536431. Accessed 15February 201597. University of Colorado Denver School ofMedicine Barbara Davis Center. Effect ofmethyldopa on MHC class II antigen presenta-tion in type 1 diabetes. In: ClinicalTrials.gov[Internet]. Bethesda, MD, National Libraryof Medicine, 2013. Available from https://clinicaltrials.gov/ct2/show/NCT01883804NLM Identifier: NCT01883804. Accessed 15February 201598. Assistance Publique-Hopitaux de Paris.dose finding study of Il-2 at ultra-low dosein children with recently diagnosed type 1diabetes (DFIL2-Child). In: ClinicalTrials.gov[Internet]. Bethesda, MD, National Libraryof Medicine, 2012. Available from https://clinicaltrials.gov/ct2/show/NCT01862120NLM Identifier: NCT01862120. Accessed 15February 2015

care.diabetesjournals.org Skyler 1007