ACG 2025 - Ulcerative Colitis Guideline

Citation: Rubin DT, Ananthakrishnan AN, Siegel CA, Barnes EL, Long MD. “ACG Clinical Guideline Update: Ulcerative Colitis in Adults.” Am J Gastroenterol 2025;120(6):1187–1224. Published online June 3, 2025. Updates the 2019 ACG UC guideline. Source file: raw/Rubin 2025 - ACG Clinical Guideline Update Ulcerative Colitis in Adults.pdf.

Type: Society clinical practice guideline (GRADE-based recommendations + expert-consensus “key concept statements”) — read as primary-tier evidence synthesis, one level more authoritative than a single trial but a level below reading the individual trials directly. Where the guideline reports a specific trial’s numbers, those numbers are relayed here but not independently re-verified against the original trial publication.

Scope

Full-adult-UC-management guideline: diagnosis/assessment/monitoring, goals of care, induction and maintenance of remission in mild-to-moderately active UC, induction and maintenance in moderately-to-severely active UC, positioning of advanced therapies relative to one another, management of the hospitalized patient with acute severe UC (ASUC) including a management algorithm (Fig. 3), and brief notes on surgery/nutrition around colectomy. Explicitly does not cover endoscopic surveillance/dysplasia management or UC in pregnancy (deferred to future ACG guidance).

Each recommendation carries a GRADE strength (Strong/Conditional) and evidence quality (High/Moderate/Low/Very low) rating; “key concept statements” are expert-consensus points not put through formal GRADE (used when the evidence base doesn’t fit the GRADE framework well).

Key points

• Diagnosis: stool testing for C. difficile recommended in all suspected UC (strong, very low quality); serologic antibody testing recommended against for both diagnosis and prognosis (strong, very low quality — notable in that a very-low-quality evidence base still produced a strong recommendation, because the “no benefit + potential harm from misdirection” balance was judged decisive despite thin data). Colonoscopy with ileal intubation and biopsies of affected/unaffected mucosa is the diagnostic standard.
• Disease extent (Montreal-style, matches what’s already on Inflammatory Bowel Disease): proctitis (≤18 cm from anal verge), left-sided (to splenic flexure), extensive/pancolitis (beyond splenic flexure). New: a primary sclerosing cholangitis phenotype with relative/absolute rectal sparing, and a periappendiceal/cecal “red patch” phenotype in limited distal disease — both managed the same as their extent-based counterparts.
• Severity assessment: new ACG UC Disease Activity Index (Table 4) combines patient-reported outcomes (stool frequency, blood, urgency), labs (Hgb, ESR, CRP, fecal calprotectin), and endoscopy (Mayo endoscopic subscore or UCEIS) into remission/mild/moderate-severe/fulminant tiers — a more multidimensional framework than the older Truelove and Witts criteria used on Inflammatory Bowel Disease.
• Treatment goal: endoscopic improvement (Mayo endoscopic subscore 0–1), not just symptomatic remission — an endoscopically healed mucosa is associated with sustained steroid-free remission and lower colectomy risk. “Deep remission” = symptomatic remission + endoscopic healing. Histologic remission is prognostically associated with better outcomes but not yet validated as a prospective treatment target.
• Fecal calprotectin is now considered a validated, non-experimental biomarker for monitoring (not just diagnosis) — used to assess therapy response, evaluate suspected relapse, and during maintenance (strong recommendation, moderate quality). Intestinal ultrasound (IUS) is newly endorsed as a point-of-care monitoring tool (pooled sensitivity 85%, specificity 92% vs. endoscopy/biomarkers), though a credentialing pathway is still being established in the US.
• Mild-to-moderate UC: stepwise 5-ASA-first approach — topical (rectal) 5-ASA for proctitis/ left-sided disease preferred over topical steroids; oral 5-ASA ≥2.0 g/d for extensive colitis; combination oral+rectal 5-ASA outperforms oral alone for left-sided disease; budesonide MMX 9 mg/d for 5-ASA nonresponders/intolerant; oral corticosteroids if 5-ASA fails outright. No benefit shown for dose-escalating 5-ASA beyond ~2.4 g/d in low-activity disease. Probiotics and FMT: insufficient evidence to recommend as primary induction therapy at this time.
• Moderate-to-severe UC — full advanced-therapy roster (this is the single biggest addition relative to what the wiki previously had, which was a bare drug-class table with no UC-specific positioning or trial evidence):
  • Anti-TNF: infliximab (ACT 1/ACT 2 trials), adalimumab, golimumab (PURSUIT program) — all strong/moderate-to-high quality recommendations. Infliximab is explicitly named the preferred anti-TNF agent for moderate-severe UC, based on observational data suggesting superior long-term outcomes and lower postoperative-complication association versus adalimumab. Combination therapy with a thiopurine is recommended when infliximab is used (strong, moderate-quality — azathioprine specifically).
  • Vedolizumab (anti-α4β7 integrin): GEMINI 1 trial data; a head-to-head RCT vs. adalimumab (VARSITY) showed vedolizumab superior for clinical remission (31.3% vs 22.5%) and endoscopic improvement (39.7% vs 27.7%) at 52 weeks — the guideline uses this to recommend vedolizumab over adalimumab specifically (strong, moderate quality). Preferred in patients at higher infection risk given its gut-selective mechanism.
  • Ustekinumab (anti-IL-12/23p40): UNIFI trial.
  • IL-23p19 inhibitors (newer class): guselkumab (QUASAR), mirikizumab (LUCENT-1/2), risankizumab (INSPIRE/COMMAND) — all strong/moderate recommendations for induction and maintenance.
  • JAK inhibitors: tofacitinib (OCTAVE 1/2 induction, OCTAVE Sustain maintenance) and upadacitinib (U-ACHIEVE/U-ACCOMPLISH induction, maintenance trial) — both strong recommendations, upadacitinib induction rated high-quality evidence (the single highest evidence grade of any induction recommendation in the guideline). Boxed-warning caveat: FDA restricts tofacitinib (and by extension the class, out of caution) to use after anti-TNF failure in the US, based on the ORAL SURVEILLANCE rheumatoid-arthritis safety trial showing increased MACE/cancer/VTE with tofacitinib vs. a TNF inhibitor in a high-cardiovascular-risk RA population — a signal not directly replicated in the UC trials themselves, but driving the label restriction anyway. Filgotinib (JAK-1 selective) is approved in Europe but not available in the US.
  • S1P receptor modulators: ozanimod and etrasimod (ELEVATE UC 52/12 trials) — newer oral small-molecule class, strong recommendations for induction and maintenance.
  • No head-to-head trials exist comparing most of these classes directly; positioning recommendations are explicitly not based on indirect/network meta-analysis comparisons (the guideline states this as a methodologic choice) except where real head-to-head RCTs exist (vedolizumab vs. adalimumab).
• Choosing among therapies: guided by prior anti-TNF exposure status, infection risk (favor vedolizumab or anti-IL-23 if higher infection risk — more gut-selective/less systemically immunosuppressive), extraintestinal manifestations (choose an agent effective at both the gut and joint/skin site), and payer-access realities — the guideline explicitly advocates against step-therapy requirements interfering with the physician-patient choice.
• Therapeutic drug monitoring: reactive TDM (checking trough levels + antidrug antibodies) is recommended on secondary loss of anti-TNF response. Proactive TDM in all unselected patients is not currently supported by sufficient evidence.
• Acute severe UC (ASUC) — defined as ≥6 bowel movements/day plus ≥1 systemic toxicity sign (tachycardia, fever, Hgb <10.5 g/dL, or elevated inflammatory markers). Full management algorithm provided (Fig. 3, reproduced logic below):
  • Day 0: baseline stool studies (incl. C. diff), flexible sigmoidoscopy within 72h (preferably 24h) to assess severity and biopsy for CMV, toxic megacolon assessment, VTE prophylaxis, hold 5-ASA (paradoxical diarrhea risk). If CDI or CMV positive, treat that first (fidaxomicin/vancomycin for CDI; ganciclovir/valacyclovir for CMV depending on serology/biopsy) — do not withhold other UC-directed therapy while treating the infection. If both negative: start IV corticosteroids without delay (methylprednisolone ≤60 mg/d or hydrocortisone 100 mg 3–4×/day).
  • Day 3: reassess. <4 BM/day for 2 days + no rectal bleeding → responder, transition to oral prednisone 40 mg, manage as outpatient. No response → escalate to IV cyclosporine (2–4 mg/kg, target trough 150–250 ng/mL) or infliximab (5 mg/kg, or 10 mg/kg if low serum albumin) or (emerging, off-label) tofacitinib/upadacitinib — the guideline flags JAK-inhibitor use in inpatient ASUC as under investigation, not yet a formal recommendation.
  • Day 6: reassess again. Response (<4 BM/day for 2 days, no bleeding) → maintain on whichever rescue agent worked (thiopurine/vedolizumab/others after cyclosporine; continued infliximab; continued tofacitinib/upadacitinib). No response, or development of toxic megacolon/perforation/ severe refractory hemorrhage → surgery.
  • Rescue-agent choice (infliximab vs. cyclosporine) should be based on institutional/provider familiarity, prior immunomodulator/anti-TNF exposure, and serum albumin — thiopurine-naive patients starting cyclosporine have lower subsequent colectomy risk than thiopurine-exposed/ -failed patients. Neither infliximab nor cyclosporine rescue therapy increases postoperative complications of colectomy, and necessary surgery should not be delayed out of concern for this (an explicit, guideline-level correction of an older surgical teaching that biologic exposure raises perioperative risk enough to warrant a “washout” delay — see note in Ulcerative Colitis below).
  • Delayed surgery in ASUC is independently associated with worse postoperative outcomes — early surgical referral/consultation is recommended, not held in reserve as a last resort.
  • Preferred surgery for ASUC: subtotal/total colectomy with end ileostomy (not a one-stage restorative proctocolectomy) — matches existing wiki teaching. Pouch construction is deliberately deferred to a later stage to let nutritional status and immunosuppression exposure normalize first.
  • Nutrition: enteral feeding preferred over bowel rest/TPN, which showed no benefit in RCTs; malnutrition (weight loss >10–15% in 6 months, BMI <18.5, albumin <30 g/L) should be corrected before colectomy where feasible given its association with higher postop morbidity.

Limitations

• GRADE methodology explicitly downweights indirect evidence (network meta-analyses, post hoc subgroup analyses) — the guideline states these “provide hypothesis-generating data but are not sufficient to stratify therapies for individual patients.” Most within-class positioning in this wiki page (e.g., “infliximab preferred anti-TNF”) rests on such lower-tier evidence even where the guideline issues a directional recommendation — treat these as reasonable-but-not-definitive rather than head-to-head-proven, except where a true RCT exists (vedolizumab vs. adalimumab is the one clear exception).
  • Note the ORAL SURVEILLANCE safety signal driving the tofacitinib/JAK restriction was measured in a rheumatoid arthritis population with pre-existing cardiovascular risk factors, not in UC patients directly — the guideline itself flags this as “not seen in the pivotal trials of tofacitinib” in UC, yet the FDA label restriction still applies. Worth remembering this is a cross-disease extrapolation, not a UC-specific safety finding.
• Does not cover endoscopic surveillance/dysplasia or pregnancy management — Ulcerative Colitis’s existing dysplasia-surveillance content (from the Sabiston textbook chapter) is therefore still the only source backing that section; not superseded or corroborated by this guideline.
• Cross-checked against both the AGA Living Guideline on IBD Pharmacotherapy and the British IBD 2025 guideline (both ingested alongside this one) — see Ulcerative Colitis’s “Efficacy positioning” section for the ACG-vs-AGA methodological disagreement (NMA-based ranking vs. not) that emerged from that cross-check.

Relevance

Substantially expands Ulcerative Colitis with the medical-management and ASUC-management content it previously lacked entirely (that page was surgery-only, deferring all medical content to the generic textbook-level Inflammatory Bowel Disease table). Directly closes the gap flagged in this wiki’s own lint-pass “Open items” notes on both pages (“no primary trial data on any specific biologic … yet”). Also directly resolves several of the primary-literature gaps identified when Stephen asked about trial coverage — this guideline gives full-text-sourced coverage of the UC advanced-therapy landscape that was previously a complete blank. (NETTER-1/PROMID/CLARINET/RADIANT-4 were a separate, NET-trial gap, out of scope for UC — those were independently ingested in the 2026-08-19 lint pass; see Small Bowel Neuroendocrine Tumors (NETs).)