Diagnosis

IBD is ultimately diagnosed by endoscopy and histology, but laboratory tests help identify patients who should undergo these more invasive tests (Figure).^5–7 Diagnosis often begins by ruling out infections and alternative conditions (eg, irritable bowel syndrome [IBS], ischemic colitis, diverticulitis, and colon cancer). The initial workup may include a complete blood count (test code 6399) and comprehensive metabolic panel (test code 10231), which can reveal anemia, thrombocytosis, and hypoalbuminemia associated with IBD.^4,6 Stool tests for common pathogens, including Clostridium difficile (test code 91664), are recommended for patients with diarrhea to rule out infectious causes.^4–6

Assessing inflammation is important to rule out noninflammatory conditions with similar symptoms (eg, IBS). Several biomarkers can be used to detect the inflammation associated with IBD; these include 2 serum markers, C-reactive protein (CRP, test code 4420) and erythrocyte sedimentation rate (ESR, test code 809), and 2 stool markers, calprotectin (test code 16796) and lactoferrin (test codes 10156 and 17321).

Serum markers: CRP and ESR

CRP is an acute-phase protein released from the liver in response to various inflammatory conditions. Although not specific for any one condition, it is highly sensitive for inflammation; CRP levels increase rapidly, up to 1,000-fold, at the onset of inflammation and decrease rapidly when it resolves.^8,9 CRP has moderate sensitivity and specificity for identifying IBD and differentiating IBD from IBS (Table 3).^10–12 A meta-analysis of studies including patients with IBD or IBS and healthy controls found that a patient with a CRP level >2.7 mg/L had a ≥90% likelihood of having IBD as opposed to having IBS or being healthy,¹³ though CRP levels at IBD diagnosis are usually much higher than that (medians: 20 mg/L in UC, 40 mg/L in CD).⁹

Table 3. Sensitivity and Specificity of Laboratory Tests for IBD [return to contents]

Clinical context, test 10–12	Sensitivity, % (95% CI)	Specificity, % (95% CI)
Diagnosis: IBD vs non-IBD
CRP	63 (51-73)	88 (80-93)
ESR	66 (58-73)	84 (80-88)
Stool calprotectin	88 (83-92)	80 (69-88)
Stool lactoferrin	82 (72-89)	95 (88-98)
Diagnosis: IBD vs IBS
CRP	75 (66-80)	60 (41-68)
ESR	55 (44-66)	47 (33-65)
Stool calprotectin	97 (91-99)	76 (66-84)
Stool lactoferrin	78 (75-82)	94 (91-96)
Diagnosis: UC vs CDa
ANCA (for UC)	55 (53-58)	89 (87-90)
ASCA (for CD)	53 (51-56)	89 (87-91)
Detect active IBD
CRP	49 (34-64)	92 (72-96)
Stool calprotectin	85 (82-87)	75 (71-79)
UC	87 (85-89)	77 (74-80)
CD	82 (80-84)	72 (69-75)
Stool lactoferrin	82 (73-88)	79 (62-89)
UC	81 (64-92)	82 (61-93)
CD	82 (73-88)	71 (63-78)
Predict IBD relapse
Stool calprotectin
Within 3 months	100	70
Within 12 months	69	75
Stool lactoferrin
Within 3 months	100	62
Within 12 months	62	65

ANCA, anti-neutrophil cytoplasmic antibodies; ASCA, anti-Saccharomyces cerevisiae antibodies; CD, Crohn disease; CI, confidence interval; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; IBD, inflammatory bowel disease; IBS, irritable bowel syndrome; UC, ulcerative colitis.
a	Among patients with established diagnoses.

 

ESR is a measure of the speed at which red blood cells (RBCs) fall through a sample of blood. Inflammation causes RBCs to aggregate more, which increases ESR, but elevated ESR is not specific to inflammation and can be caused by other conditions (eg, pregnancy, anemia).^9,14 ESR has moderate sensitivity and specificity for identifying IBD, but poor sensitivity and specificity for differentiating IBD from IBS (Table 3).¹³

CRP and ESR may not be elevated in all patients with IBD, especially in those whose disease activity is mild.^5,6,15 For this reason, testing ESR and CRP in combination can be useful,^9,14 as one marker may be elevated and the other normal in up to 38% of patients.¹⁵ Thus, elevated levels of either marker are consistent with IBD, but normal levels do not rule it out.

Stool markers: calprotectin and lactoferrin

Calprotectin and lactoferrin are proteins released by neutrophils into the intestinal lumen when the intestines are inflamed. Thus, high levels of calprotectin and lactoferrin in stool indicate intestinal inflammation.

Calprotectin testing has high sensitivity and specificity for diagnosing IBD (Table 3) and is recommended by the American College of Gastroenterology (ACG) for differentiating between IBD and IBS (cutoff >50-100 μg/g).⁶ High levels of calprotectin are consistent with IBD whereas low levels virtually exclude it; a meta-analysis found that a patient with a calprotectin level <40 μg/g had a <1% probability of having IBD as opposed to having IBS or being healthy.¹³

Fewer studies have assessed the utility of lactoferrin, but the available data suggests that it has comparable accuracy to calprotectin for diagnosing IBD (Table 3).^12,16 Similar to calprotectin, a normal level of lactoferrin indicates absence of intestinal inflammation, whereas a high level (medians: 1,100 μg/g in UC, 44 μg/g in CD9) is consistent with IBD.

Although stool markers are more specific than serum markers for bowel inflammation, they cannot distinguish between the causes of such inflammation; GI infections and colorectal cancer, for example, are also associated with bowel inflammation and may cause elevated calprotectin and lactoferrin levels.¹⁷ Additionally, for reasons that are not completely understood, stool markers are generally more accurate for diagnosing UC than CD.¹⁶

Differentiation of UC and CD

Differentiating UC from CD is an important part of IBD diagnosis with implications for prognosis and treatment selection. In most cases, differentiation is made with findings from endoscopy and histology. However, in 10% to 15% of patients, these findings are equivocal, and a definitive diagnosis cannot be made. This condition is known as indeterminate colitis (IC) or IBD-unclassified (IBD-U).¹⁸ For these patients, testing for certain antibodies involved in IBD inflammation—anti-neutrophil cytoplasmic antibodies (ANCAs, test code 70171) and anti-Saccharomyces cerevisiae antibodies (ASCAs, test codes 10294, 10295, 17609)—may provide support for differentiation.

ANCAs are autoantibodies that react to antigens in neutrophils. ANCAs can be detected by indirect immunofluorescence, which can identify the ANCA pattern, and immunoassay, which can detect specific antibodies. Common patterns and their associated antibodies include¹⁹

• Cytoplasmic (C-ANCA): associated with proteinase-3 (PR3) antibodies
• Perinuclear (P-ANCA): associated with myeloperoxidase (MPO) antibodies
• Atypical P-ANCA: associated with various antibodies other than PR3 and MPO

The various ANCA patterns tend to be characteristic of certain diseases. Whereas the cytoplasmic and perinuclear patterns are associated with small-vessel vasculitides, the atypical P-ANCA pattern is associated with IBD and relatively specific for UC (Table 4).^19,20 ASCAs, on the other hand, are antibodies to brewer’s yeast and are relatively specific for CD.¹⁹

Table 4. Antibody Prevalence in IBD and Healthy People [return to contents]

Antibody20	UC	CD	Healthy people
Atypical P-ANCA	50%-67%	6%-15%	<11%
ASCA	4%-14%	40%-60%	<5%

ASCA, anti-Saccharomyces cerevisiae antibodies; CD, Crohn disease; IBD, inflammatory bowel disease; P-ANCA, perinuclear anti-neutrophil cytoplasmic antibodies; UC, ulcerative colitis.

 

Given their specificities for the 2 diseases, ANCA and ASCA testing may be useful for differentiating UC from CD when conventional criteria cannot.²¹ Used in this way, their diagnostic utility is greatest when both antibodies are tested (test code 16503)^8,19; the combination of atypical P-ANCA–positive and ASCA–negative (P-ANCA+/ASCA−) results suggests UC, while the opposite combination (P-ANCA−/ASCA+) suggests CD.^7,20

Although antibody testing is commonly used to help differentiate UC from CD,²² only a few prospective studies have assessed its diagnostic accuracy in patients with IBD-U, and their results varied.^22,23 Additionally, ANCA and ASCA testing may be less useful in Asian (including Chinese, Japanese, and South Korean) patients, as they are less likely than White patients to have these antibodies.²⁴

Management

Laboratory tests are useful throughout the management of IBD. The inflammatory biomarkers used in diagnosis are also used to monitor disease activity in established IBD. For some patients and treatment classes, laboratory tests can also aid in treatment selection and optimization.

Assessing disease activity

Disease activity is monitored closely in the management of IBD to guide treatment selection, assess response to treatment, and monitor for relapse.^3,5,6 The gold standard for assessing disease activity in IBD is endoscopy,^8,25 but repeated endoscopies are impractical, as they are invasive, expensive, and uncomfortable.²⁶ Other methods, including clinical assessment, radiology, and laboratory tests, are therefore important adjunctive tools.⁶ In particular, biomarkers of inflammation are routinely used to assess IBD disease activity and can be used as alternatives to endoscopy in certain contexts.^6,8

Levels of CRP, calprotectin, and lactoferrin correlate with endoscopic disease activity in IBD,^5,6,9,17 and tests for these markers have moderate to high diagnostic accuracy for detecting active IBD (Table 3). CRP testing has high specificity but low sensitivity for IBD disease activity,²⁷ as patients with mild disease activity may have false-negative results,^5,15 but CRP levels ≥5 mg/dL are consistent with moderate to severe disease activity.²⁷ As in diagnosis, calprotectin and lactoferrin testing are more accurate than CRP testing for assessing disease activity, and elevated levels of these markers are consistent with active disease.²⁷

Guidelines from the ACG and American Gastroenterological Association (AGA) state that CRP, calprotectin, and lactoferrin testing have roles in assessing disease activity in UC and CD.^5,6,26 Additionally, recent guidance from the AGA allows using these biomarkers to obviate routine endoscopies in certain patients with UC.26 These guidelines suggest the following:

• For patients with UC in symptomatic remission, normal levels of CRP, calprotectin (suggested cutoff: <150 μg/g), or lactoferrin are sufficient to rule out active inflammation and obviate endoscopy.
• For patients with UC and moderate to severe symptoms, elevated levels of CRP, calprotectin, or lactoferrin are sufficient to rule in active inflammation and likewise obviate endoscopy.
• For patients with UC and mild symptoms, endoscopy is still suggested in most cases, regardless of biomarker status.

Because biomarkers are not perfectly accurate, their suitability to assess disease activity—especially in place of endoscopy—depends largely on a patient’s pretest probability of active disease and the clinical implications of a false-negative or false-positive result.^26,27 Additionally, as in diagnosis, stool biomarkers tend to be less accurate at assessing disease activity in CD than UC.¹⁴

Assessing response to therapy

As markers of disease activity, inflammatory biomarkers can be used to assess response to IBD treatment when tested serially. CRP and calprotectin levels tend to decrease or normalize after the start of treatment in patients who respond positively to treatment.^9,14 For example, in a study of patients with UC and CD starting biologic therapy with infliximab, a decrease in calprotectin level by 80% or to 50 μg/g by week 2 was associated with endoscopic remission at week 10.²⁸

Similarly, biomarker testing during the course of treatment can indicate long-term response, such that patients with lower levels of CRP and calprotectin during treatment are more likely to maintain their response.^14,25 In a study of 87 patients with UC on maintenance therapy with infliximab, all those in sustained deep remission after 1 year (n=30) had maintained calprotectin levels <40 μg/g throughout the year.²⁸

In a treat-to-target management approach, inflammatory biomarker levels may also serve as treatment goals themselves. The Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE-II) Initiative by the International Organization for the Study of IBD recommends using normalization of CRP and calprotectin (suggested cutoff: <100-250 μg/g) as an intermediate treatment target in IBD.²⁹

Monitoring for relapse

For patients with IBD in remission, calprotectin and lactoferrin testing can be used to monitor for disease relapse and may be able to detect it before symptoms appear (Table 3). Prospective studies of patients with IBD in remission have found that calprotectin levels increase about 3 months before relapse becomes otherwise apparent.^28,30

Serial measurements may predict relapse with better accuracy than isolated measurements. In one prospective study, calprotectin >300 μg/g across 2 measurements, 1 month apart, had 100% specificity for disease relapse within 12 months.²⁸ In another study, patients whose calprotectin levels doubled in 3 months had a doubled risk of relapse within the next 3 months.³⁰

Therapeutic drug monitoring

Laboratory tests are used before and during IBD treatment to guide treatment selection and dosing and to monitor for adverse effects. Routine testing includes complete blood count and comprehensive metabolic panel,⁴ but certain drug classes may require additional testing. For patients starting treatment with biologics and/or biosimilars, screening for viral hepatitis and tuberculosis (test codes 37616, 37620) is recommended.^4,6,31 For patients treated with biologics and thiopurines, drug analyte testing (therapeutic drug monitoring) may be used to optimize safety and effectiveness.

TNF blockers

Tumor necrosis factor (TNF) blockers, such as adalimumab (Humira^®), infliximab (Remicade^®), and the infliximab biosimilar infliximab-dyyb (Inflectra^®), can be used to treat UC and CD.^32–34  While some patients respond positively to these drugs, up to 80% may be refractory to treatment,³⁵ showing either nonresponse during induction (primary failure) or response during induction followed by loss of effectiveness (secondary failure).

When TNF blocker treatment fails, the AGA and ACG suggest measuring drug and/or anti-drug antibody (ADA) levels to guide therapy changes,^5,6,36 which may include adjusting the dose or dosing intervals, switching to a different TNF blocker, or switching to a non-TNF blocker.

Measuring TNF blocker drug levels can help differentiate treatment failure–related issues that are pharmacokinetic (PK, absence of detectable drug) from those that are pharmacodynamic (PD, presence of drug but lack of effectiveness).

Drug levels are typically assessed just before administration of the next dose to determine whether trough levels are therapeutic or subtherapeutic.³⁵ The AGA suggests target trough concentrations of ≥7.5 μg/mL for adalimumab and ≥5 μg/mL for infliximab in patients with active IBD on maintenance therapy.³⁶ Therapeutic trough levels generally indicate PD issues related to TNF-independent disease. Subtherapeutic trough levels, on the other hand, can indicate different types of issues, depending on whether ADAs have formed.^35,37

ADAs, which are reported to form in up to 83% of patients treated with TNF blockers,38 can cause subtherapeutic trough levels and reduce treatment effectiveness by (1) forming ADA-drug complexes that accelerate drug clearance and (2) directly preventing the drug from binding TNF.^35,37 Adalimumab or infliximab ADA levels ≥10 AU indicate detectable serum levels, whereas levels <10 AU are considered “not detected” and suggest that treatment failure is not caused by ADAs.

Quest assays measure total (ie, free and bound) ADA. Some tests for adalimumab or infliximab ADAs are susceptible to inaccurate results caused by cross-reactivity with rheumatoid factor (RF), but the ADA tests developed by Quest are not affected by the presence of RF.

Testing for drug levels can indicate bioavailability, whereas testing for ADAs can help differentiate causes of insufficient bioavailability. Therefore, appropriate test selection varies as follows:

• Measuring only drug levels (test codes 36298, 36303) may be appropriate if sequential testing is preferred to concurrent testing.
• Measuring only ADA levels (test codes 36294, 36301) may be appropriate if insufficient bioavailability has already been established.
• Measuring both drug and ADA levels (test codes 36296, 36311) may help identify the cause of treatment failure more quickly.

Patients with subtherapeutic drug trough levels who test negative for ADAs may have nonimmune PK issues, such as poor adherence to treatment or accelerated drug clearance caused by nonimmune mechanisms.^35,37,38 Nonimmune PK issues can therefore be managed by addressing adherence issues, increasing the dose, or shortening the dosing interval.^35,38

However, for patients with subtherapeutic trough levels who test positive for ADAs, the cause of treatment failure is likely to be immune-mediated, and switching to a different TNF blocker may be more effective than increasing the dose.^35,38 Therefore, testing for ADAs in addition to drug levels can help determine which changes in treatment approach are most appropriate (Table 5).

Table 5. Interpretation of Results in Patients With TNF Blocker Treatment Failurea [return to contents]

	ADAs not detected35,37,38	ADAs detected
Drug levels subtherapeutic	Suggests insufficient bioavailability caused by nonimmune PK or adherence issues Consider increasing the dose or addressing adherence issues	Suggests insufficient bioavailability caused by immunogenicity Consider switching to a different TNF blocker
Drug levels therapeutic	Suggests PD issue caused by TNF-independent disease Consider switching to a non-TNF treatment	Rare situation that may be caused by a false-positive result or nonfunctional ADAs Consider retesting or testing for neutralizing antibody by cell-based assay

ADA, anti-drug antibody; PD, pharmacodynamic; PK, pharmacokinetic; TNF, tumor necrosis factor.
a	Test interpretation for infliximab assays applies to both infliximab and infliximab-dyyb.

 

Anti-integrin and anti-interleukin agents

The anti-integrin agent vedolizumab (VDZ) and the anti-interleukin agent ustekinumab (UST) are also approved to treat UC and CD,^39,40 but compared with TNF blockers, less data are available to support the utility of TDM for these drugs.^41–43 Most studies have shown that trough levels of VDZ and UST negatively correlate with favorable clinical outcomes,^41,42,44 but the optimal trough levels identified by these studies have varied (Table 6).^41,45,46

Table 6. Drug Trough Levels Associated With Improved Therapeutic Outcomes [return to contents]

Drug	Time point	Drug trough level,45 μg/mL
VDZ	Induction (week 6)	>33-37
	End of induction (week 14)	>15-20
	Maintenance	>10-15
UST	Induction (week 8)	>3-7
	Maintenance	>1-3

UST, ustekinumab; VDZ, vedolizumab.

Accordingly, expert consensus panels, including the Building Research in IBD Globally (BRIDGe) group, have endorsed TDM for patients with primary or secondary failure of VDZ and UST but have not recommended specific drug trough levels to target or specific adjustments for optimizing treatment.^46,47 Target trough levels may depend on the indication, phase of treatment, desired therapeutic outcome, and other factors.^45,46

Quest offers tests for VDZ and UST drug levels (test codes 18244, 18218), ADA levels (test codes 18231, 18219), and both drug and ADA levels (test codes 18220, 18217). The formation of ADAs against VDZ and UST is relatively rare (<5% of patients41,44) but is associated with lower drug trough levels and treatment failure.^41,42 With Quest’s tests, ADA test results >15 ng/mL for VDZ and >2 AU/mL for UST indicate the presence of ADAs.

Thiopurines

Thiopurines, including 6-mercaptopurine (Purinethol^®) and its prodrug, azathioprine (Imuran^®), can be used to treat UC and CD.^48,49 However, thiopurines are not appropriate for all patients and may be ineffective or have adverse effects. To optimize safety and effectiveness, pretreatment testing and therapeutic drug monitoring may be appropriate for patients starting or receiving thiopurine treatment.

Although most patients can safely use thiopurines, a small number of people are unable to safely metabolize them and may experience serious harm.³⁶ Testing for thiopurine S-methyltransferase (TPMT) status can identify these patients and is suggested by the ACG and AGA before starting thiopurine treatment.^6,36

TPMT is the primary enzyme in thiopurine metabolism, and its level of expression affects a patient’s risk for adverse effects to thiopurines.⁵⁰ Most people have a normal level of TPMT activity, but 1 in 10 have low TPMT activity, which increases the risk of thiopurine toxicity, and 1 in 300 have very low or no TPMT activity, which greatly increases this risk.⁵⁰

TPMT status can be tested by phenotype (a blood test for TPMT activity level, test code 18831) or genotype (a genetic test for variants in TPMT, the gene that encodes TPMT, test code 37742). The AGA suggests using TPMT status to determine appropriate use of thiopurines, as follows^51,52:

• Patients with normal enzyme activity level (>12 nmol 6-MMP/hr/mL RBC) or a normal genotype (0 non-functional variants) can start on a standard thiopurine dose.
• Patients with low activity (4-12 nmol 6-MMP/hr/mL RBC) or a heterozygous genotype (1 non-functional variant) should start on a low dose with gradual, closely monitored escalation to a normal dose.
• Patients with very low enzyme activity (<4 nmol 6-MMP/hr/mL RBC) or a homozygous genotype (2 non-functional alleles) should not use thiopurines.

For patients already receiving thiopurine treatment, thiopurine metabolite testing can guide dose adjustments and identify thiopurine toxicity. The AGA therefore suggests metabolite testing (test code 91745) for thiopurine-treated patients with active disease or adverse effects that may be related to thiopurine toxicity.³⁶

Thiopurine metabolites include 6-thioguanine nucleotide (6-TGN) and 6-methylmercaptopurine (6-MMP), but individuals differ in how much of each they produce, and these differences can affect their therapeutic response and risk for adverse effects.⁵² Levels of 6-TGN, the therapeutic metabolite, must be high enough to be effective, but excessive levels of either metabolite can lead to various toxicities, including myelosuppression (excessive 6-TGN) and hepatoxicity (excessive 6-MMP).⁵²

Widely used target levels for metabolite monitoring are 230 to 450 pmol/8x108 RBCs for 6-TGN and <5,700 pmol/8x108 RBCs for 6-MMP.³⁶ Test interpretation and management strategies may depend on the indication for testing (ie, active disease or suspected toxicity) (Table 7).

Table 7. Interpretation of Results in Thiopurine-Treated Patients With Active IBD or Possible Thiopurine Toxicity [return to contents]

6-TGN level (pmol/8x108 RBCs)	6-MMP level (pmol/8x108 RBCs)	Interpretation51–53
Undetectable	Undetectable	Suggests poor adherence Consider addressing adherence issues
Low (<230)	Normal (<5,700)	Suggests a subtherapeutic dose Consider increasing the dose
Low (<230)	High (≥5,700)	Suggests hypermethylation (risk of hepatotoxicity) Consider using an alternative dosing strategya or switching to a non-thiopurine treatment
Normal (230-450)	Normal (<5,700)	Suggests a therapeutic dose For active IBD, consider switching to a non-thiopurine treatment For suspected toxicity, consider other causes
High (>450)	Any	Suggests a supratherapeutic dose (risk of myelosuppression and/or hepatotoxicity) For active IBD, consider switching to a non-thiopurine treatment For suspected toxicity, consider reducing the dose

6-TGN, 6-thioguanine nucleotide; 6-MMP, 6-methylmercaptopurine; IBD, inflammatory bowel disease.
a	Alternate dosing strategies for hypermethylation include (1) dose-splitting (administering half the daily dose twice daily) and (2) reducing the thiopurine dose and adding allopurinol.52,53