SONIC

From Wiki Journal Club
Jump to navigation Jump to search
Colombel JF, et al. "Infliximab, azathioprine, or combination therapy for Crohn's disease". The New England Journal of Medicine. 2010. 362(15):1383-1395.
PubMedFull textPDF

Clinical Question

Among patients with moderate to severe Crohn disease without prior treatment with immunosuppressive therapies or biologics, does azathioprine, infliximab, or a combination of both therapies improve rates of steroid-free remission?

Bottom Line

In patients with moderate to severe Crohn disease who had not previously been treated with immunosuppressives or biologics, a combination of azathioprine and infliximab resulted in a higher rate of steroid-free remission at week 26 than either drug alone. Infliximab monotherapy outperformed azathioprine monotherapy for this outcome.

Major Points

The efficacy of the chimeric anti-TNF-α monoclonal antibody infliximab in maintenance therapy for Crohn disease was established in ACCENT I (2002)[1] and ACCENT II (2004).[2] A trial demonstrating its efficacy as an induction agent was lacking.

The 2010 Study of Biologic and Immunomodulator Naive Patients in Crohn Disease (SONIC) trial randomized 508 treatment naive patients with moderate to severe disease to azathioprine, infliximab, or a combination of both medications. At week 26, combination therapy was associated with higher rate of corticosteroid-free clinical remission than infliximab monotherapy (56.8% vs. 44.4%; NNT 8) and azathioprine monotherapy (56.8% vs. 30.0%; NNT 4). Infliximab monotherapy outperformed azathioprine monotherapy for this outcome (44.4% vs. 30.0%; NNT 7). Combination therapy was better than either monotherapy for rates of mucosal healing, any clinical remission, disease severity rating, and QOL rating. The combination therapy was associated with lower rates of serious adverse events.

A 2014 AHRQ comparative effectiveness review[3] found moderate strength of evidence for benefit of infliximab over azathioprine for mucosal healing, benefit of the combination therapy over azathioprine for disease activity and mucosal healing, and benefit of combination therapy over azathioprine for disease activity and mucosal healing. However, the data for these conclusions were almost entirely drawn from SONIC.

Guidelines

AGA medications for induction and remission of Crohn disease (2013, adapted)[4]

  • For remission induction in moderately severe disease:
    • Suggestion against thiopurine monotherapy like 6-MP or azathioprine (weak recommendation, moderate evidence)
    • Recommend anti-TNF-α medications (strong recommendation, moderate evidence)
    • Recommend anti-TNF-α monotherapy over thiopurine monotherapy (strong recommendation, moderate evidence)
    • Recommend anti-TNF-α plus thiopurines over thiopurine monotherapy (strong recommendation, high-quality evidence)
    • Suggest anti-TNF-α plus thiopurines over anti-TNF-α monotherapy (weak recommendation, moderate evidence)

Design

  • Multicenter, randomized, double blind, comparative trial
  • N=508 patients with Crohn disease
    • Azathioprine (n=170)
    • Infliximab (n=169)
    • Combination therapy (N=169)
  • Setting: 92 centers in multiple continents
  • Enrollment: 2005-2008
  • Follow-up: 26 weeks
  • Analysis: Intention-to-treat
  • Primary outcome: Corticosteroid-free clinical remission at week 26

Population

Inclusion Criteria

  • Age ≥21 years
  • Crohn's disease for ≥6 weeks
  • Crohn's disease activity Index (CDAI) score 220-450 off corticosteroids
    • Out of a total score of 600, remission <150 and severe >450
  • One of the following:
    • Dependent on corticosteroids for disease control
    • Consideration of a second course of corticosteroids in a year
    • No response to ≥4 weeks of mesalamine ≥2.4 g/day or budesonide ≥6 mg/day

Exclusion Criteria

  • Treatment with 6MP, MTX, or anti-TNF biologic
  • Short gut
  • Ostomy
  • Symptomatic stricture
  • Abscess
  • Abdominal surgery in prior 6 months
  • Granulomatous infection including TB
  • OI in prior 6 months
  • Active infection with HIV or hepatitis B or C
  • MS
  • Cancer
  • Mutant thiopurine methyltransferase phenotype

Baseline Characteristics

From the azathioprine group.

  • Demographics: Age 35 years, male 53%, White race 91%
  • Baseline health data: Weight 70 kg
  • Baseline labs: Median CRP 1.0 mg/dL
  • Baseline medications: Budesonide 15%, 5-ASA 61%
    • Systemic corticosteroids:
      • None: 77%
      • Prednisone equivalent <20 mg/day: 8%
      • Prednisone equivalent ≥20mg/day: 15%
  • Crohn's data:
    • Duration of disease 2.4 years
    • CDAI score: 287 (see interventions above)
    • Inflammatory bowel disease questionnaire (IBDQ[5]) score: 112 (out of 224, higher indicates better QOL)
    • Mucosal lesions on baseline colonoscopy: 67.6%
    • Area of GI tract involved:
      • Ileum or colon: 100%
      • Ileum and colon: 41%
      • Proximal GI tract: 4%
  • Fistula: prior history in 16%, current in 13%
  • History of extraintestinal manifestations: 55%
  • Crohn's related surgery: 31%
  • Current or prior smoker: 52%

Interventions

  • Randomized to a group with stratification by center, disease duration, and baseline prednisone equivalent dosing:
    • Infliximab - Infliximab 5 mg/kg IV at weeks 0, 2, 6, then q8 weeks, plus placebo tablets
    • Azathioprine - Azathioprine 2.5 mg/kg qday, plus placebo infusions
    • Combination - Infliximab plus azathioprine as above
  • Oral mesalamine was continued at a stable dosing
  • Corticosteroids could be initiated and/or modified to a maximum prednisone equivalent of 40 mg PO qday through week 14, after which the dose was reduced at at a rate of 5 mg q7 days
  • Budesonide dosing could be modified to a maximum dose of 9 mg PO qday through week 14, after which it was reduced at a rate of 3 mg q14 days to a dose ≤6 mg/day
  • Colonoscopy at baseline, repeated at week 26 for those with mucosal ulcers at baseline
  • Groups were followed until week 30
  • Participants had the option of continuing on their assigned therapies in a blinded fashion until week 50 as part of an extension trial, though these results are out of the scope of this review

Outcomes

Presented as azathioprine vs. infliximab vs. combination (infliximab vs. azathioprine | combination vs. infliximab | combination vs. azathioprine).

Primary Outcome

Corticosteroid-free clinical remission at week 26
Defined as CDAI score <150, budesonide <6mg/day, and no systemic corticosteroids in prior 3 weeks.
30.0% vs 44.4% s. 56.8% (P=0.006 | P=0.02 | P<0.001)

Secondary Outcomes

Corticosteroid-free clinical remission
At week 6: 14.1% vs. 29.6% vs. 32.5% (P<0.001 | P=0.55 | P<0.001)
At week 10: 24.1% vs. 37.3% vs. 46.7% (P=0.006 | P=0.07 | P<0.001)
At week 18: 25.9% vs. 42.6% vs. 52.7% (P<0.001 | P=0.06 | P<0.001)
At week 26: 30.0% vs. 44.4% vs. 56.8% (P=0.006 | P=0.02 | P<0.001; the primary outcome)
Mucosal healing at week 26
Of patients with mucosal ulcerations on baseline colonoscopy, defined as resolution of ulcerations on repeat colonoscopy.
16.5% vs. 30.1% vs. 43.9% (P=0.02 | P=0.06 | P<0.001)
Any clinical remission at week 26
Defined as CDAI score <150.
31.8% vs. 47.9% vs. 60.4% (P=0.002 | P=0.02 | P<0.001)
Questionnaire scores, change from baseline to week 26
CDAI reduction >70 points: 41.8% vs. 56.2% vs. 66.9% (P=0.008 | P=0.04 | P<0.001)
CDAI reduction >100 points: 37.6% vs. 54.4% vs. 62.1% (P=0.002 | P=0.16 | P<0.001)
IBDQ: +31.4 vs. +39.9 vs. +45.2 (P=0.05 | P=0.13 | P<0.001)
Corticosteroids
At week 0
Patients on corticosteroids: 40 vs. 52. vs. 47 patients
Average dose: 23.8 vs. 24.8 vs. 24.9 mg/day
At week 2
Patients on corticosteroids: 48 vs. 50 vs. 49 patients
Average dose: 22.9 vs. 21.2 vs. 22.8 mg/day
At week 6
Patients on corticosteroids: 53 vs. 52 vs. 51 patients
Average dose: 18.6 vs. 17.7 vs. 18.3 mg/day
At week 10
Patients on corticosteroids: 56 vs. 56 vs. 52 patients
Average dose: 16.2 vs. 15.7 vs. 15.0 mg/day
At week 18
Patients on corticosteroids: 59 vs. 57 vs. 56 patients
Average dose: 13.5 vs. 13.2 vs. 11.6 mg/day
At week 26
Patients on corticosteroids: 60 vs. 60 vs. 60 patients
Average dose: 11.6 vs. 11.0 vs. 9.4 mg/day
Initiation of corticosteroids through week 26: 20 vs. 8 vs. 11 patients
Change in CRP from baseline to week 26
-1.1 vs. 1.3 vs. -1.4 (P=0.32 | P=0.44 | P=0.08)

Additional Analyses

Treatment duration
21.1 vs. 24.1 vs. 24.9 weeks
Infliximab antibodies at week 30
N/a vs. 14.6% vs. 0.9%
Infliximab trough at week 30
N/a vs. 1.6 vs. 3.5 ug/mL (P<0.001)

Subgroup Analysis

For the primary outcome. All are post-hoc.

Baseline CRP level
<0.08 mg/dL: 33.8% vs. 40.3% vs. 50.7% (P=0.41 | P=0.31 | P=0.09)
≥0.08 mg/dL: 27.6% vs. 47.5% vs. 63.5% (P=0.004 | P=0.03 | P<0.001)
Mucosal lesions
Present: 29.6% vs. 50.5% vs. 61.5% (P=0.002 | P=0.12 | P<0.001)
Absent: 40.7% vs. 33.3% vs. 40.0% (P=0.37 | P=0.69 | P=0.93)
Unknown: 21.4% vs. 38.2% vs. 57.1% (P=0.14 | P=0.07 | P=0.003)
Mucosal lesion present and baseline CRP ≥0.08 mg/dL
28.0% vs. 56.9% vs. 68.8% (P<0.001 | P=0.17 | P<0.001)

Adverse Events

Any
89.4% vs. 89.0% vs. 89.9% (P=1.00 | P=0.86 | P=1.00)
Resulting in discontinuation: 26.1% vs. 17.8% vs. 20.7% (P=0.08 | P=0.58 | P=0.25)
Serious: 26.7% vs. 23.9% vs. 15.1% (P=0.61 | P=0.04 | P=0.01)
Infection
45.3% vs. 46.0% vs. 41.9% (P=0.91 | P=0.45 | P=0.58)
Serious: 5.6% vs. 4.9% vs. 3.9% (P=0.81 | P=0.79 | P=0.61)
Sepsis: 0.6% vs. 0% vs. 0%
Infusion reaction
Adverse event ≤1 hour of receiving infusion.
5.6% vs. 16.6% vs. 5.0% (P=0.002 | P=<0.001 | P=1.00)
Adverse event occurring >10% in ≥1 group
Nausea: 32.3% vs. 22.1% vs. 25.1%
Abdominal pain: 16.1% vs. 25.2% vs. 17.3%
Crohn's disease worsening: 16.8% vs. 18.4% vs. 10.6%
Vomiting: 17.4% vs. 14.1% vs. 8.4%
Diarrhea: 8.1% vs. 12.3% vs. 7.8%
Fatigue: 15.5% vs. 14.7% vs. 14.5%
Pyrexia: 11.2% vs. 19.6% vs. 11.7%
Arthralgia: 11.2% vs. 19.6% vs. 11.7%
Headache: 12.4% vs. 16.6% vs. 12.8%
Nasopharyngitis: 12.4% vs. 9.2% vs. 11.7%

Criticisms

  • Relatively low rate of mucosal ulceration suggests symptom burden may have been partially attributable to IBS[6]
  • Induction therapy was only given to the groups receiving infliximab[6]
  • Azathioprine may take up to six months to become fully effective so this trial may not have been a fair long-term comparison between infliximab and azathioprine[6]
  • Combination therapy in the treatment naive is in opposition to guidelines recommending a stepwise approach for initial therapy[6]
  • Unclear efficacy in those with longer duration of disease[6]
  • Lack of reporting of outcomes pertaining to extraintestinal manifestations of the disease[6]
  • No cost analysis[6]

Funding

  • Centocor Ortho Biotech (now Janssen Biotech, sellers of Remicade in the US, the brand name of infliximab)
  • Schering-Plough (now owned by Merck, sellers of Remicade outside of the US)

Further Reading