📚 Part of the TILA-278 Regulatory Dossier — Reader's Guide. This article shows the live document; edits to the source appear here automatically.

---

Human PD & QT Study Reports (TILA-278)

Document ID: CLINPHARM-003
Version: 1.0
Change History: 1.0 — Initial issue.
Standard(s): ICH E14/S7B

Human Pharmacodynamic & QT Study Reports — TILA-278

Clinical pharmacodynamics and cardiac-safety assessment for TILA-278. For a bispecific antibody, a dedicated thorough-QT study is generally not warranted; cardiac safety is addressed by concentration-QTc analysis and the clinical safety database. ICH E14/S7B; E14/S7B Q&A.

1. Scope and Regulatory Framework

Sponsor: Virtual Biopharma Inc. Investigational product: TILA-278 — humanized IgG1 bispecific monoclonal antibody (anti-TL1A [TNFSF15] antagonist arm / IL-22 receptor [IL-22RA1/IL-10RB] agonist arm), CHO-derived, sterile solution for subcutaneous (SC) injection. Protocol: TILA278-201 (Phase 2b, randomized, double-blind, placebo-controlled, parallel-group, 12-week induction) in moderate-to-severe ulcerative colitis (UC). Report scope (CTD Module 5.3.4): Reports of human pharmacodynamic (PD) studies and the cardiac-safety (QT/QTc) evaluation for TILA-278. This report characterizes mechanism-related pharmacodynamic biomarkers of both binding arms, the exposure–response relationship for those biomarkers, and the concentration–QTc evaluation supporting the cardiac-safety conclusion. It was authored in accordance with ICH E3 and ICH M4E; the cardiac-safety assessment follows ICH E14/S7B and the E14/S7B Q&A; nonclinical cardiovascular safety pharmacology follows ICH S6(R1). The program is submitted as a BLA under 21 CFR 601.

Because TILA-278 is a full-length monoclonal antibody with target-mediated drug disposition (TMDD) and clinically relevant immunogenicity, the PD strategy was designed to read out (i) engagement of the soluble/membrane TL1A antagonist target, (ii) downstream activation of the IL-22 receptor agonist arm, and (iii) the net effect on systemic and intestinal inflammation and on mucosal healing. Pharmacodynamic conclusions are cross-referenced to the pharmacokinetic and exposure–response analyses (CLINPHARM-002) and the immunogenicity report (IMMUNO-001).

2. Pharmacodynamic Rationale and Biomarker Strategy

TILA-278 pairs two complementary mechanisms on a single IgG1 scaffold:

• TL1A antagonism neutralizes TL1A (TNFSF15) signaling through death receptor 3 (DR3/TNFRSF25), dampening TH1/TH17-driven mucosal inflammation and attenuating the pro-fibrotic program associated with stricturing disease.
• IL-22 receptor agonism activates epithelial STAT3 signaling, driving intestinal epithelial regeneration, antimicrobial-peptide production, and mucus-barrier repair.

The biomarker panel was structured to interrogate each arm and their integrated effect:

Biomarker domain	Analyte(s)	Interpretation
TL1A target engagement	Free serum TL1A; total serum TL1A	Free-ligand suppression and drug-bound total-ligand accumulation confirm antagonist-arm engagement
IL-22R downstream activation	Serum REG3A; serum amyloid A (SAA)	Proximal readouts of epithelial/hepatic IL-22RA1–STAT3 signaling confirm agonist-arm activity
Systemic inflammation	High-sensitivity C-reactive protein (hs-CRP)	Net anti-inflammatory effect
Intestinal inflammation	Fecal calprotectin	Mucosal neutrophilic activity / disease burden
Mucosal healing	Central endoscopy; histology	Structural/functional recovery of the epithelium

Serum PD biomarkers were collected at Weeks 0, 2, 4, 8, and 12; fecal calprotectin at Weeks 0, 4, 8, and 12; and centrally read endoscopy with paired mucosal biopsies at Weeks 0 and 12. Analyses were pre-specified on the full analysis set (FAS, N=840: TILA-278 High n=284, TILA-278 Low n=283, placebo n=273), with biomarker changes summarized as median percentage change from baseline unless otherwise noted.

3. Target Engagement — TL1A Antagonist Arm

Consistent with an antibody that binds soluble ligand, TILA-278 produced dose-ordered suppression of free serum TL1A alongside an increase in total (largely drug-bound) serum TL1A. The accumulation of total TL1A reflects protection of the ligand from clearance while it is complexed to the antibody and is a pharmacokinetic-linked marker of engagement rather than an indicator of increased TL1A biology.

Free serum TL1A — median % change from baseline at Week 12	TILA-278 High	TILA-278 Low	Placebo
Free TL1A	−89%	−73%	+2%

Near-maximal free-TL1A suppression was reached across the High regimen and was substantial but incomplete at the Low regimen, mirroring the greater-than-dose-proportional exposure described in the PK report. Suppression of free TL1A was sustained through the dosing interval at the High regimen, consistent with trough concentrations remaining above the concentration associated with high-level target occupancy.

4. Receptor Agonism — IL-22R Agonist Arm

Engagement of the IL-22 receptor arm was demonstrated by dose-ordered induction of the epithelial IL-22 response gene product REG3A, a canonical STAT3-dependent antimicrobial-peptide readout of IL-22RA1 signaling.

IL-22R downstream marker — geometric mean fold-change from baseline (peak, Week 4)	TILA-278 High	TILA-278 Low	Placebo
Serum REG3A	2.6×	1.7×	1.05×

REG3A induction peaked early (Week 2–4) and partially normalized by Week 12 as mucosal inflammation resolved and epithelial regenerative demand declined — a trajectory consistent with a healing-then-homeostasis pattern rather than tachyphylaxis. Serum amyloid A, a proximal hepatic/epithelial IL-22RA1 readout, showed a modest, transient early rise on active treatment without an accompanying clinical inflammatory correlate; this reflects agonist-arm target engagement and did not track with worsening disease. No adverse metabolic pharmacodynamic signal attributable to systemic IL-22R agonism (e.g., clinically meaningful shifts in glucose or lipids) was identified over the induction period.

5. Systemic and Intestinal Inflammatory Biomarkers

The integrated effect of TL1A antagonism and mucosal IL-22R agonism was a dose-ordered reduction in both systemic and intestinal inflammatory markers.

Marker — median % change from baseline at Week 12	TILA-278 High	TILA-278 Low	Placebo
High-sensitivity CRP	−64%	−43%	−9%
Fecal calprotectin	−72%	−50%	−13%

Reductions in fecal calprotectin — the marker most proximate to mucosal neutrophilic activity — were the largest and most clearly dose-separated, aligning with the endoscopic and histologic findings in Section 6. The net decline in hs-CRP indicates that the overall anti-inflammatory effect of the molecule dominated over any transient acute-phase contribution from the IL-22R agonist arm.

6. Mucosal Healing and Endoscopic Pharmacodynamics

Structural pharmacodynamic effects were assessed by centrally read endoscopy and by the clinical composite (modified Mayo) at Week 12.

Endpoint at Week 12	TILA-278 High	TILA-278 Low	Placebo
Endoscopic improvement, %	48.9%	27.9%	6.2%
Clinical remission (modified Mayo ≤ 2, no subscore > 1), n/N (%)	106/284 (37.3%)	46/283 (16.2%)	2/273 (0.7%)
Modified Mayo, LS-mean change from baseline	−3.36	−2.76	−1.00
Difference vs placebo (LS-mean)	−2.36	−1.77	—

Endoscopic improvement and clinical remission were both strongly dose-ordered and separated from placebo, with the largest effect at the High regimen. The concordance of REG3A induction and fecal-calprotectin reduction with endoscopic improvement supports the mechanistic hypothesis that combined TL1A neutralization and epithelial IL-22R agonism translates into measurable mucosal healing, and it provides pharmacodynamic corroboration of the clinical remission result.

7. Exposure–Response for Pharmacodynamic Effects

Biomarker responses were related to serum TILA-278 exposure using the population-PK-derived individual exposures (CLINPHARM-002). Free-TL1A suppression followed a sigmoid maximum-effect (Emax) relationship with serum concentration in which the concentration producing half-maximal suppression was low relative to the steady-state trough concentrations achieved on the High regimen; accordingly, free-TL1A suppression was near-maximal and flat across the High regimen exposure range, while the Low regimen operated on the ascending portion of the curve. Downstream markers (REG3A induction, hs-CRP and fecal-calprotectin reduction) showed a similar but right-shifted exposure–response, consistent with signal amplification and integration downstream of target binding.

This PD exposure–response is coherent with the efficacy exposure–response: the probability of clinical remission rose with increasing exposure and began to plateau within the exposure range spanned by the High regimen, indicating that the High dose operates near the top of the exposure–response relationship for both biomarkers and clinical benefit. Treatment-emergent anti-drug antibodies (ADA) increased clearance modestly (approximately +28%, per CLINPHARM-002) but, given the flat top of the exposure–response relationship at the High regimen, did not translate into a clinically relevant reduction in pharmacodynamic effect at the doses studied (see IMMUNO-001).

8. Cardiac Safety Assessment — Overview and Waiver Rationale

A dedicated thorough-QT (TQT) study was not conducted and is not warranted for TILA-278. The rationale is grounded in the modality and in ICH E14/S7B and the E14/S7B Q&A, which recognize that, for products with a low likelihood of direct cardiac ion-channel interaction — such as monoclonal antibodies — QT risk can be adequately characterized by a concentration–QTc analysis of ECGs collected in clinical studies together with the clinical safety database, without a stand-alone TQT study.

The mechanistic basis for the low pro-arrhythmic likelihood of TILA-278 is as follows:

• TILA-278 is a large (150 kDa) humanized IgG1 bispecific antibody with no physicochemical access to intracellular or transmembrane cardiac ion channels (hERG/Kv11.1, Nav1.5, Cav~1.2).
• Its target epitopes are TL1A and the IL-22 receptor; neither is a cardiac ion channel, and neither target is implicated in cardiac repolarization.
• The antibody is eliminated by proteolytic catabolism to peptides and amino acids, yielding no small-molecule metabolites that could carry independent ion-channel liability.
• Distribution is confined to the vascular and interstitial space with no active cardiac partitioning.

Consistent with this biologic modality, an in vitro hERG assay and a nonclinical S7B core repolarization assay are not applicable, and a stand-alone TQT study is not warranted. In line with ICH S6(R1), genotoxicity (ICH S2) and carcinogenicity (ICH S1) studies are likewise not warranted for this antibody.

9. ECG Data from TILA278-201

Serial 12-lead ECGs were recorded in triplicate at screening, on Day 1, and at Weeks 2, 4, 8, and 12 (and at early termination), with Fridericia correction (QTcF) as the primary correction method and Bazett (QTcB) as a supportive secondary correction. Central-tendency and categorical outlier analyses were pre-specified.

Central tendency. Mean change from baseline in QTcF was small and comparable to placebo at every scheduled timepoint, with no dose-ordered trend.

Mean change from baseline in QTcF at Week 12 (ms)	TILA-278 High	TILA-278 Low	Placebo
ΔQTcF	+2.1	+1.4	+1.0

Categorical outliers. No subject in any arm had an absolute QTcF > 480 ms or a change from baseline in QTcF > 60 ms. The small numbers of subjects with QTcF in the 450–480 ms band or with ΔQTcF in the 30–60 ms band were low and balanced across arms with no dose relationship. There were no episodes of torsade de pointes, no treatment-emergent sustained ventricular arrhythmia, and no sudden cardiac deaths. No treatment-emergent, clinically meaningful changes in heart rate, PR interval, QRS duration, or ECG morphology (including T-wave/U-wave changes or new bundle-branch block) were observed.

10. Concentration–QTc Analysis

The relationship between serum TILA-278 concentration and QTc was characterized by a pre-specified linear mixed-effects concentration–QTc (C-QTc) model, with change from baseline in QTcF (ΔQTcF) as the dependent variable, time-matched serum concentration as the primary predictor, and fixed effects for treatment and nominal time plus subject-level random intercept and slope.

• The estimated slope of the concentration–ΔQTcF relationship was not statistically significant and was close to zero (on the order of 0.01 ms per µg/mL; two-sided 90% CI spanning zero), indicating no concentration-dependent effect on cardiac repolarization.
• The model-predicted placebo-corrected change from baseline in QTcF (ΔΔQTcF) at the geometric-mean steady-state peak concentration of the High regimen (approximately 47.5 µg/mL; CLINPHARM-002) was small, with the upper bound of the two-sided 90% confidence interval remaining below the 10 ms threshold of regulatory concern. The upper bound remained below 10 ms even under a conservative high-exposure scenario.
• Given the slow SC absorption and the antibody mechanism, no hysteresis between concentration and effect was evident, supporting a direct (non-delayed) C-QTc model.

These results, together with the categorical ECG findings, exclude a clinically relevant QTc effect of TILA-278 across the clinical exposure range.

11. Nonclinical Cardiovascular Safety Pharmacology

In accordance with ICH S6(R1), safety pharmacology endpoints were integrated into the general (repeat-dose) toxicology program rather than assessed in stand-alone small-molecule safety-pharmacology studies. The cynomolgus monkey was the sole pharmacologically relevant species (confirmed cross-reactivity for both the TL1A and IL-22R binding arms) and was used for the pivotal repeat-dose SC toxicology studies. Cardiovascular endpoints — including telemetered arterial blood pressure, heart rate, and ECG intervals (PR, QRS, QT, and rate-corrected QTc) — together with respiratory and central-nervous-system observations were incorporated into these studies. No adverse effects on cardiovascular, respiratory, or neurobehavioral parameters were identified at exposures in multiples of the anticipated clinical exposure. No hERG or other in vitro ion-channel assay was performed, as such assays are not applicable to a monoclonal antibody.

12. Immunogenicity Considerations for Pharmacodynamics

Treatment-emergent ADA were predominantly low-titer and transient and modestly increased clearance without a clinically relevant impact on pharmacodynamic biomarker responses or on cardiac-safety parameters. Because the pharmacodynamic and efficacy exposure–response relationships are flat across the High regimen exposure range, the observed immunogenicity did not meaningfully attenuate free-TL1A suppression, downstream biomarker responses, or mucosal healing at the doses studied. The immunogenicity assays, incidence, and PK/PD impact are detailed in IMMUNO-001 and CLINPHARM-002.

13. Conclusions

• TILA-278 demonstrated on-mechanism pharmacodynamics for both binding arms: dose-ordered suppression of free serum TL1A (High −89%, Low −73%) with drug-bound total-TL1A accumulation confirming antagonist-arm engagement, and dose-ordered induction of the IL-22R downstream marker REG3A (High 2.6×, Low 1.7×) confirming agonist-arm activity.
• The integrated anti-inflammatory and mucosal-healing effect was reflected in dose-ordered reductions in hs-CRP (High −64%, Low −43%) and fecal calprotectin (High −72%, Low −50%) and in endoscopic improvement (48.9% / 27.9% / 6.2%), concordant with clinical remission (37.3% / 16.2% / 0.7%) and the modified Mayo LS-mean changes (−3.36 / −2.76 / −1.00; differences vs placebo −2.36 / −1.77).
• Pharmacodynamic and efficacy exposure–response were coherent, with near-maximal target engagement and a plateau reached within the High regimen exposure range; treatment-emergent ADA did not meaningfully reduce pharmacodynamic effect at the doses studied.
• A dedicated thorough-QT study is not warranted for this monoclonal-antibody modality. Concentration–QTc analysis showed no concentration-dependent QTc effect, with the upper bound of the two-sided 90% CI for predicted ΔΔQTcF below 10 ms at the High regimen peak exposure; no QTcF > 480 ms, no ΔQTcF > 60 ms, and no pro-arrhythmic events were observed. Nonclinical cynomolgus cardiovascular safety pharmacology identified no cardiovascular liability. These findings support the cardiac-safety conclusion for TILA-278 consistent with ICH E14/S7B and the E14/S7B Q&A.

14. Abbreviations

ADA, anti-drug antibody; BLA, Biologics License Application; C-QTc, concentration–QTc; CHO, Chinese hamster ovary; CRP, C-reactive protein; CTD, Common Technical Document; DR3, death receptor 3 (TNFRSF25); ECG, electrocardiogram; Emax, maximum effect; FAS, full analysis set; hERG, human ether-à-go-go-related gene (Kv11.1); hs-CRP, high-sensitivity C-reactive protein; IgG1, immunoglobulin G subclass 1; IL-22R, interleukin-22 receptor (IL-22RA1/IL-10RB); LS-mean, least-squares mean; PD, pharmacodynamic(s); PK, pharmacokinetic(s); QTcB, Bazett-corrected QT interval; QTcF, Fridericia-corrected QT interval; REG3A, regenerating islet-derived protein 3 alpha; SAA, serum amyloid A; SC, subcutaneous; STAT3, signal transducer and activator of transcription 3; TdP, torsade de pointes; TL1A, TNF-like ligand 1A (TNFSF15); TMDD, target-mediated drug disposition; TQT, thorough QT; UC, ulcerative colitis; ΔQTcF, change from baseline in QTcF; ΔΔQTcF, placebo-corrected change from baseline in QTcF.

Values represent standard, ICH-conformant clinical-pharmacodynamic and cardiac-safety parameters consistent with a humanized IgG1 bispecific antibody exhibiting target-mediated disposition, where a specific value stands in for the corresponding measured result.