Low-Dose Interleukin 2 for the Treatment of Moderate to Severe Ulcerative Colitis

Abstract

Ulcerative colitis (UC) represents an inflammatory condition of the human colon believed in part to be mediated by dysregulation of the host immune system including potentially inadequate mucosal regulatory T-cell (Treg) responses.1Chang J.T. N Engl J Med. 2020; 383: 2652-2664Crossref PubMed Scopus (445) Google Scholar,2Laukova M. Glatman Zaretsky A. Eur J Immunol. 2023; 53e2250007Crossref PubMed Scopus (8) Google Scholar Low-dose (LD) interleukin 2 (IL2) has been shown to expand peripheral Tregs and ameliorate various inflammatory diseases in mouse models3Humrich J.Y. et al.Proc Natl Acad Sci U S A. 2010; 107: 204-209Crossref PubMed Scopus (139) Google Scholar, 4Grinberg-Bleyer Y. et al.J Exp Med. 2010; 207: 1871-1878Crossref PubMed Scopus (336) Google Scholar, 5Hulme M.A. et al.Diabetes. 2012; 61: 14-22Crossref PubMed Scopus (106) Google Scholar, 6Bonnet B. et al.J Immunol. 2016; 197: 188-198Crossref PubMed Scopus (42) Google Scholar as well as in human trials.7Saadoun D. et al.N Engl J Med. 2011; 365: 2067-2077Crossref PubMed Scopus (620) Google Scholar, 8Koreth J. et al.N Engl J Med. 2011; 365: 2055-2066Crossref PubMed Scopus (887) Google Scholar, 9Rosenzwajg M. et al.Ann Rheum Dis. 2019; 78: 209-217Crossref PubMed Scopus (258) Google Scholar Additionally, we have previously reported amelioration of colitis in humanized mice after LD IL2 treatment.10Goettel J.A. et al.Cell Mol Gastroenterol Hepatol. 2019; 8: 193-195Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar We hypothesize that LD IL2-mediated expansion of Tregs may suppress disease in UC and provide a novel treatment approach. Here, we assess whether LD IL2 is safe in UC patients, expands Tregs, and affects underlying immune mechanisms. We conducted an open-label phase 1b/2a 8-week induction trial of daily subcutaneous LD IL2 (Proleukin, Prometheus) in patients with moderate to severe UC. A total of 26 adult patients with active disease (total Mayo score, 6–10) and failure of response to ≥1 UC therapy were enrolled (Supplementary Table 1). The primary objectives were assessment of safety and tolerability of LD IL2 and determination of the maximum tolerated dose (MTD) via a 3 + 3 design with 3 dose-escalation cohorts: 0.3 × 106 IU/m2/d (dose A), 1 × 106 IU/m2/d (dose B) (MTD in Graft vs Host Disease [GvHD]8Koreth J. et al.N Engl J Med. 2011; 365: 2055-2066Crossref PubMed Scopus (887) Google Scholar), or 1.5 × 106 IU/m2/d (dose C) with an additional 10 enrolled at the MTD, once determined. Secondary outcomes included determination of efficacy as measured by clinical response (decrease in total Mayo score of ≥3 or ≥30% at week 8) and clinical remission (a total Mayo score of ≤2, with no individual subscore exceeding 1, including the week 8 Mayo endoscopic score). Flexible sigmoidoscopy was performed before treatment and at week 8. Laboratory and clinical parameters were assessed every 1 to 2 weeks. Flow cytometry was performed on peripheral blood mucosal samples to assess changes in Treg and conventional T-cell (Tcon) populations as well as activation states. Overall, LD IL2 was well tolerated, with common (occurring in >10% of patients) mild adverse events including injection site reactions, fever, and malaise across all doses. There were no serious adverse events and no deaths. One protocol-defined dose-limiting toxicity (DLT) was observed, a maculopapular rash at the injection site of a dose B patient, who was withdrawn from the study. Six additional patients withdrew because of exacerbation of UC (n = 2), uveitis (n = 1), and low-grade systemic symptoms (n = 2). Ultimately, 4 participants received dose A, 7 received dose B, and 5 received dose C. Although no dose C participants experienced a DLT, none experienced a clinical response, and all reported malaise. Therefore, although dose C qualified as the MTD, dose B was deemed the maximum efficacious dose (MED), and after discussion with the study data safety monitoring board, and an additional 10 patients were recruited at dose B/MED. Among all participants across all doses who completed the 8-week induction, 52.6% (10/19) achieved clinical response, with 21.1% (4/19) achieving clinical remission (Figure 1A). Dose B/MED specifically yielded a clinical response rate of 69.2% (9/13) and a clinical remission rate of 30.8% (4/13) in those who completed induction and a clinical response rate of 52.9% (9/17) and remission rate of 23.5% (4/17) among all enrolled at this dose. LD IL2 treatment expanded peripheral Tregs (pTregs) as assessed by flow cytometry, defined as a >2-fold increase in pTregs (out of total CD4+ lymphocytes) compared to pretreatment (Figure 1B). Excluding participants who withdrew without having achieved pTreg expansion (given lack of data regarding whether they may have achieved this at later timepoints), 67% (2/3) of dose A, 100% (16/16) of dose B/MED, and 100% (4/4) of dose C participants achieved pTreg expansion (Figure 1B and Supplementary Table 2). pTregs expanded in 100% (10/10) of responders and 92% (12/13) of nonresponders. Most participants (86%, 19/22) achieved expansion before or at week 2 (and 100% by week 4), with maximum expansion most commonly occurring at week 4. LD IL2 treatment increased the absolute pTreg count (imputed using flow data and absolute lymphocyte count), but the absolute pTcon count did not significantly change (data not shown). Greater pTreg expansion was seen in dose B/MED and dose C than dose A (P < .001), but no difference was observed between dose B/MED and dose C (Figure 1C). No difference was seen in pTreg expansion between responders and nonresponders (Figure 1C). In participants who completed the 8-week induction, phosphorylated STAT5 (pSTAT5; induced upon IL2 receptor activation) expression increased with LD IL2 treatment in a dose-dependent manner in pTregs and pTcons (Figure 1D). In dose B/MED, pSTAT5 expression in pTcons was significantly greater in nonresponders than in responders (P = .02), with a trend of greater expression in nonresponder pTregs as well (P = .08) (Figure 1D). We interpret these findings to suggest that unwanted IL2-mediated activation of pTcons could explain nonresponse in these patients. In the mucosa, there was a strong trend of diminished mucosal Tregs (mTregs) in responders compared to nonresponders, both at scope 1 (P = .06) and at scope 2 (P = .08) (Figure 1E). Expansion of mTregs (defined as >1.2 fold change compared to scope 1) was neither required nor sufficient for response (Figure 1F and Supplementary Table 2). We demonstrate that LD IL2 is safe and well tolerated in patients with UC. We identified the MED as 1 × 106 IU/m2/d (dose B) after the highest dose (dose C), despite lack of DLTs, did not result in clinical efficacy and was associated with enhanced pSTAT5 in pTcons. LD IL2 treatment at the MED for 8 weeks resulted in improved total Mayo score in most participants, yielding a response rate on par with established biologic therapies. LD IL2 treatment expanded pTregs, but not pTcons, in nearly all patients regardless of response. We found a dose-dependent increase in pSTAT5 activation in both pTregs and pTcons, along with increased pTcon activation in nonresponders to LD IL2. We include assessment of Treg populations in disease target tissue and show that mTreg expansion is neither necessary nor sufficient for response to LD IL2. Trial limitations include the lack of a placebo arm, small sample size, and lack of central/blinded reading of endoscopic images. Our results support the safety and efficacy of LD IL2 as a treatment modality in UC and its evaluation in larger trials. ClinicalTrials.gov, Number NCT02200445 Low-Dose IL2 UC Study Group: Matthew Hamilton, Jared Barends, Madeline Carrellas, Katherine Freer, Jordan Gringauz, Julia Green, Noah Herwood, Jonathan Hurtado, Ryan Kelly, Jennifer Mitri, Caroline Rourke, Gwen Saccocia, Sydney Whitcomb, Enju Liu, David Klatzmann, Punyanganie de Silva, Francis A. Farraye, Joseph D. Feuerstein, Alan Moss, Samir A. Shah, Joshua R. Korzenik, Athos Bousvaros, John Koreth, Robert Soiffer, Jerome Ritz, Tanya Logvinenko, Ashwin Ananthakrishnan, and Hans Herfarth. The authors would also like to acknowledge Prometheus for supplying the drug for this trial. Jessica R. Allegretti, MD, MPH (Investigation: Lead; Supervision: Supporting; Writing – original draft: Equal; Writing – review & editing: Equal). Vanessa Mitsialis, MD (Formal analysis: Lead; Writing – review & editing: Lead). James B. Canavan, MD, PhD (Conceptualization: Supporting; Methodology: Supporting; Writing – review & editing: Supporting). Scott B. Snapper, MD, PhD (Conceptualization: Lead; Methodology: Lead; Supervision: Lead; Writing – review & editing: Lead). Supplementary Table 1Patient DemographicsPatient CharacteristicsDose ADose BDose CDose B/MEDn47510Male, n (%)2 (50)4 (57.1)4 (80)8 (80)White, n (%)4 (100)7 (100)5 (100)9 (90)Average age, y36.74346.638.5Average disease duration, y6.759.3212.610.4Average baseline total Mayo score8.58.78.89.2Average ± SD baseline CRP, mg/L22.7 ± 31.87.1 ± 6.53.0 ± 1.82.9 ± 2.8Average ± SD baseline fecal calprotectin, μg/g stool1422 ± 1066297.6 ± 150373.2 ± 185319.3 ± 327.1Prednisone use at baseline, n (%)2 (50)3 (43)1 (20)7 (70)Average prior biologics/small molecules used2.52.41.82.8Failure of 2 or more biologics prior to baseline, n (%)3 (75)6 (86)4 (80)9 (90)Disease extent: proctitis, n1100Disease extent: left-sided, n2537Disease extent: pancolitis, n1123CRP, C-reactive protein; SD, standard deviation. Open table in a new tab Supplementary Table 2Peripheral and Mucosal Treg Numbers Across All Recruited PatientsDosePTBaseline (D0) pTregs, % CD4+ lymphocytesMax pTreg expansion, fold increase over D0pTreg expansion achievedFirst timepoint of pTreg expansion, >2 fold increase over D0Timepoint of max pTregexpansionScope 1 mean Tregs, % CD4+ lymphocytesScope 2 mean Tregs, % CD4+ lymphocytesColonic Tregs, fold changeResponderParticipationPT25.02.8YesW4W830.631.71.0NoCompleted studyDose APT3PT4PT55.1 5.15.01.2 1.32.9No No YesNANAW2W4W6W4Not available for analysis17.219.9NA20.320.2NA1.21.0No No YesWithdrew at W6 Completed study Completed studyDose BPT6PT7PT8PT9PT10PT11PT124.2 4.8 4.6 4.73.411.07.21.9 8.2 2.9 4.5 7.1 2.23.2No Yes Yes Yes Yes Yes YesNAW2W1W2W2W2W4W2W2W1W4W8W4W49.03.820.8 21.0 13.7 19.715.4NA6.6NA26.0 45.1 24.211.0NA1.7NA1.2 3.3 1.20.7NoYes No NoYes No yesWithdrew at W6Completed study Withdrew at W2 Completed study Completed study Completed study Completed studyMEDPT18PT19PT20PT21PT22PT23PT24PT25PT26PT276.87.510.15.4 8.8 8.5 5.8 6.1 3.96.42.8 3.3 4.8 5.5 3.8 3.8 4.7 5.4 7.13.6Yes Yes Yes Yes Yes Yes Yes Yes Yes YesW2W2W1W2W4W1W1W2W1W1W2W6W8W6W4W2W4W4W2W413.16.27.028.319.8 8.933.2 6.511.818.0NA5.814.7 29.525.2 7.115.2 11.816.1NANA0.9 2.1 1.0 1.3 0.8 0.5 1.81.4NANo Yes Yes No Yes Yes Yes Yes NoNoWithdrew at W4 Completed study Completed studyWithdrew after W8 Completed study Completed study Completed study Completed study Completed study Withdrew at W6Dose CPT13PT14PT15PT16PT174.09.311.910.07.34.2 3.6 1.0 3.62.9Yes Yes No Yes YesW1W2NAW1W1W2W8W1W2W414.9 20.822.4 9.731.314.641.3NA8.2NA1.02.0NA0.8NANo No No No NoCompleted study Completed study Withdrew at W2Completed study Withdrew at W5D0, day 0; max, maximum; NA, not applicable; PT, patient; W, week. Open table in a new tab CRP, C-reactive protein; SD, standard deviation. D0, day 0; max, maximum; NA, not applicable; PT, patient; W, week.

Topics & Concepts