Appropriate Therapeutic Drug Monitoring of Biologic Agents for Patients With Inflammatory Bowel Diseases

Abstract

Background & Aims:

Therapeutic drug monitoring (TDM) is widely available for biologic therapies in patients with inflammatory bowel disease (IBD). We reviewed current data and provided expert opinion regarding the clinical utility of TDM for biologic therapies in IBD.

Methods:

We used a modified Delphi method to establish consensus. A comprehensive literature review was performed regarding the use of TDM of biologic therapy in IBD and presented to international IBD specialists. Subsequently, 28 statements on the application of TDM in clinical practice were rated on a scale of 1 to 10 (1=strongly disagree and 10=strongly agree) by each of the panellists. Statements were accepted if 80% or more of the participants agreed with a score ≥7. The remaining statements were discussed and revised based on the available evidence followed by a second round of voting.

Results:

The panel agreed on 24 (86%) statements. For anti-tumor necrosis factor (anti-TNF) therapies, proactive TDM was found to be appropriate after induction and at least once during maintenance therapy, but this was not the case for the other biologics. Reactive TDM was appropriate for all agents both for primary non-response and secondary loss of response. The panellists also agreed on several statements regarding TDM and appropriate drug and anti-drug antibody (ADA) concentration thresholds for biologics in specific clinical scenarios.

Conclusion:

Consensus was achieved towards the utility of TDM of biologics in IBD, particularly anti-TNF therapies. More data are needed especially on non-anti-TNF biologics to further define optimal drug concentration and ADA thresholds as these can vary depending on the therapeutic outcomes assessed.

INTRODUCTION

Biologic therapies, including the anti-tumor necrosis factor (anti-TNF) agents (infliximab, adalimumab, certolizumab pegol and golimumab), the adhesion molecule inhibitors (vedolizumab and natalizumab), and the p-40 interleukin-12/23 inhibitor ustekinumab, are effective treatments for patients with moderate to severe inflammatory bowel disease (IBD).^{1, 2} Nevertheless, up to 1/3 of patients with Crohn’s disease (CD) and ulcerative colitis (UC) show primary non-response (PNR) to biologic therapies and up to 50% of patients after an initial clinical response stop therapy either for secondary loss of response (SLR) or a serious adverse event.^{3, 4} Both PNR and SLR are due to either pharmacokinetic (PK) or pharmacodynamic (PD) problems. PK issues are associated with inadequate drug exposure, often due to the development of anti-drug antibodies (ADA), whereas PD issues are typically related to inflammatory process unrelated to the targeted immunoinflammatory pathway.^{5, 6}

Numerous studies have demonstrated a positive correlation between serum biologic drug concentrations and favorable therapeutic outcomes, while low or undetectable drug concentrations can lead to immunogenicity and treatment failure (Tables 1–3 and supplementary table 1).^7–95 Therapeutic drug monitoring (TDM), defined as the assessment of drug concentrations and ADA, is an important tool for optimizing biologic therapy. Reactive TDM has rationalized the management of PNR and SLR and has proven more cost-effective when compared to empiric dose escalation.^96–102 Preliminary data suggest that proactive TDM, with drug titration to a target trough concentration, performed in patients with clinical response/remission can also improve the efficacy of anti-TNFs.^{38, 39, 103, 104} Moreover, proactive TDM may also improve the cost-effectiveness and safety of biologic therapy via the implementation of a de-escalation strategy in patients with supra-therapeutic drug concentrations by reducing the dose, increasing the time interval and/or stopping the immunomodulator in patients on combination therapy (optimized monotherapy).^{39, 82, 105–107}

Table 1.

Serum adalimumab concentration thresholds associated with therapeutic outcomes in inflammatory bowel disease.

IBD type	Threshold (μg/ml)	Therapeutic outcome	TDM assay	Assay type	Ref.
Induction (week 2)
CD	>6.7	Clinical remission (w14)	ELISA	AHLC	23
Post-induction (week 4)
CD	>5	Drug retention	HMSA	Prometheus	29
CD	>12	Normal CRP (≤5mg/L)	ELISA	LFA/ELISA (R-Biopharm AG)	31
UC	≥7.5	Mucosal healing (w10–14)	ELISA	Leuven assay	30
UC	>4.6	Clinical response (w12)	ELISA	Leuven assay	26
UC	>7	Clinical response (w52)	ELISA	Leuven assay	26
Maintenance
CD	>5.9	Normal CRP (≤5mg/L)	ELISA	AHLC	15
CD	>5.9	Normal CRP (≤3mg/L)	ELISA	Sumitomo Bakelite Co Ltd	16
CD	>8.1	Mucosal healing	HMSA	Prometheus	18
CD	>5.6	Normal CRP (≤3mg/L)	ELISA	In-house	19
CD	>7.9	Mucosal healing	ELISA	In-house	19
CD	>10.3	Mucosal healing	ELISA	In-house	20
CD	>5 (w26)	Clinical remission (w52)	ELISA	Sanquin Diagnostics	21
CD	≥12	Endoscopic remission	HMSA	Prometheus	22
CD	≥12.2	Histologic remission	HMSA	Prometheus	22
CD	≥3.7 (w14)	CRP normalization (w14)	ELISA	AHLC	23
CD/UC	>6.6	Normal CRP (≤5mg/L)	ELISA	AHLC	13
CD/UC	≥6.9	No SLR	RIA	Biomonitor A/S	14
CD/UC	>7.1	Mucosal healing	ELISA	AHLC	13
CD/UC	>4.9	Mucosal healing	ELISA	Theradiag	9
CD/UC	>7.8	Histologic remission	HMSA	Prometheus	12
CD/UC	>7.5	Mucosal healing	HMSA	Prometheus	12
CD/UC	>12.2	Successful dose reduction	ELISA	Promonitor Grifols	11
CD/UC	>9	Clinical response	ELISA	Promonitor Grifols	11
CD/UC	>6.6	Normal CRP (≤5mg/L)	ELISA	Promonitor Grifols	11
CD/UC	>4.5	When SLR, better long-term outcome when change to a biological with a different mechanism of action compare to anti-TNF dosage increase or a switch within class	ELISA	AHLC	10
CD/UC	≥3	No active inflammationa	ELISA	AHLC	10
CD/UC	>4.9	When SLR, high risk of failure who subsequently after changing to infliximab	ELISA	Theradiag	8
CD/UC	>7.3	Clinical remission	ELISA	New Zealand assay	7

^adefined as increased CRP level and/or endoscopic/imaging documentation of inflammation.

ELISA: enzyme-linked immunosorbent assay; HMSA: homogeneous mobility shift assay; CRP: C-reactive protein, TDM: therapeutic drug monitoring; RIA: Radioimmunoassay; SLR: secondary loss of response; TNF: tumor necrosis factor; CD: Crohn’s disease; UC: ulcerative colitis; LFA: lateral flow-based assay; Ref.: references, AHLC: antihuman lambda chain.

Table 3.

Association of anti-drug antibodies with therapeutic outcomes in inflammatory bowel disease.

Drug	IBD type		ADA	Therapeutic outcome	TDM assay	Assay type	Ref.
IFX	CD		≥282 ng/mL-eq	Lower success rate of treatment optimization	ELISA	Leuven drug-tolerant assay	75
IFX	CD		>8 μg/mL-eq	Shorter clinical response	ELISA	Prometheus	28
IFX	CD		Detectable	Lack of mucosal healing	ELISA	MP Biomedicals	17
IFX	CD		Detectable	Elevated CRP (>5 mg/L)	HMSA	Prometheus	56
IFX	CD		Detectable	Elevated CPP (>5 mg/L)	HMSA	Prometheus	60
IFX	CD		Detectable	Lack of fistula healing	HMSA	Prometheus	12
IFX	CD		Detectable	SLR	ELISA	Prometheus	88
IFX	CD		Detectable	SLR	RIA	Biomonitor A/S	87
IFX	UC		Detectable	Lack of endoscopic response	HMSA	Prometheus	33
IFX	UC		Detectable	Lack of mucosal healing	ELISA	Leuven drug-tolerant assay	67
IFX	CD/UC		≥8.8 U/ml	Drug discontinuation	HMSA	Prometheus	86
IFX	CD/UC		Detectable	PNR	ELISA	AHLC	73
IFX	CD/UC		Detectable	Drug discontinuation	HMSA	Prometheus	63
IFX	CD/UC		>9.1 U/ml	Failure of dose intensification after SLR	HMSA	Prometheus	63
IFX	CD/UC		>12 U/mL	Surgery	HMSA	Prometheus	85
IFX	CD/UC		Undetectable	Mucosal healing	ELISA	AHLC	13
IFX	CD/UC		Undetectable	Short-term clinical response	HMSA	Prometheus	27
IFX	CD/UC	Detectable		SLR	ELISA	AHLC	32
IFX	CD/UC	Detectable		SLR	ELISA	AHLC	84
IFX	CD/UC	>9 μg/mL-eq		When SLR, longer duration of response when anti-TNF agents are switched than when dosage is increased	ELISA	AHLC	10
IFX	CD/UC	≥3.3 U/mL		Lack of post-adjustment endoscopic remission	HMSA	Prometheus	37
IFX	CD/UC	Detectable		Treatment related adverse events	ELISA	Promonitor Menarini / ImmunDiagnostik	83
IFX	CD/UC	Detectablea		PNR (w14)	ELISA	AHLC	73
IFX	CD/UC	>4.3 μg/mL-eqb		PNR (w14)	ELISA	AHLC	73
IFX	CD/UC	>9.1 U/mL		IFX discontinuation	HMSA	Prometheus	82
IFX	CD/UC	>9.1 U/mL		Infusion reactions	HMSA	Prometheus	82
IFX	CD/UC	>200 ng/mL-eq		No response to treatment optimization	ELISA	Theradiag	81
ADM	CD	Detectable		PNR	ELISA	AHLC	23
ADM	CD	Detectable		Drug discontinuation	HMSA	Prometheus	29
ADM	CD	Detectable		Drug discontinuation	ELISA	In-house	57
ADM	CD	>12 U/mL		Lack of clinical response	RIA	Biomonitor A/S	58
ADM	CD	Detectable		Active disease	ELISA	AHLC	15
ADM	CD	Detectable		Higher CRP and ESR	ELISA	Sumitomo Bakelite Co., Ltd	16
ADM	CD	Detectabled		No clinical remission (w52)	RIA	Sanquin	21
ADM	CD	Detectable (w12)		Higher needs for dose escalation less frequently sustained clinical benefit due to PNR or SLR	ELISA	R-Biopharm AG	31
ADM	CD/UC	Detectable		Drug discontinuation	RIA	Biomonitor A/S	80
ADM	CD/UC	>4 μg/mL-eq		When SLR, longer duration of response when anti-TNF agents are switched than when dosage is increased	ELISA	AHLC	10
ADM	CD/UC	Detectable		SLR	RIA	Biomonitor A/S	14

^aeither week 2 or 6;

^bweek 2;

^cUniversité François-Rabelais, Immuno-Pharmaco-Genetics of Therapeutic Antibodies, Tours, France;

^dweek 26.

ADA: anti-drug antibody; IFX: infliximab; ADM: adalimumab; ELISA: enzyme-linked immunosorbent assay; CD: Crohn’s disease; UC: ulcerative colitis; CRP: C-reactive protein; RIA: Radio-immunoassay; eq: equivalent; SLR: secondary loss of response; U: units; HMSA: homogeneous mobility shift assay; ESR: erythrocyte sedimentation rate; AHLC: antihuman lambda chain antibody; TDM: therapeutic drug monitoring; TNF: tumor necrosis factor; w: week; PNR: primary non-response; Ref.: references.

However, there are still some limitations when applying TDM into clinical practice, such as when to utilize TDM, proper interpretation and application of the results, and the identification of the optimal window/thresholds to target. These therapeutic windows or thresholds appear to vary based on the outcome of interest and the IBD phenotype (Tables 1 and 2 and supplementary table 1). Moreover, most of the data on implementation of TDM refer to anti-TNF therapies and the maintenance phase of treatment.

Table 2.

Association of serum certolizumab pegol, golimumab, vedolizumab and ustekimumab concentration thresholds with therapeutic outcomes in inflammatory bowel disease.

IBD type	Time point	Threshold (μg/ml)		Therapeutic outcome	TDM assay	Assay type	Ref.
A. Certolizumab pegol
CD	Post-induction (w6)	>31.8		Clinical response/remission (w6)	ELISA	UCB Pharma	94
CD	Post-induction (w6)	>31.9		Normal CRP (≤5mg/L) (w6)	ELISA	UCB Pharma	94
CD	Post-induction (w6)	>32.7		Normal FC (<250mg/g) (w6)	ELISA	UCB Pharma	94
CD	Post-induction (w6)	>34.5		Normal FC (<250mg/g) and CDAI (≤150) (w6)	ELISA	UCB Pharma	94
CD	Post-induction (w6)	>36.1		Normal FC (<250mg/g) and CDAI (≤150) (w26)	ELISA	UCB Pharma	94
CD	Post-induction (w8)	>23.3		Endoscopic remission (w10)	ELISA	UCB Pharma	95
CD	Maintenance (w12)	>13.8		Normal FC (<250mg/g) (w26)	ELISA	UCB Pharma	94
CD	Maintenance (w12)	>14.8		Normal FC (<250mg/g) and CDAI (≤150) (w26)	ELISA	UCB Pharma	94
B. Golimumab
UC	Induction (w2)	>8.9		Clinical response (w6)	ECLIA	Janssen Biotech Inc	48
UC	Post-induction (w4)	>7.4		Clinical response (w6)	ECLIA	Janssen Biotech Inc	48
UC	Post-induction (w6)	>2.5	Clinical response (w6)		ECLIA	Janssen Biotech Inc	48
UC	Post-induction (w6)	>2.6	Partial clinical response (w14)		ELISA	In house Leuven	93
UC	Maintenance (w28)	>0.9	Clinical remission (w30 and 54)		ECLIA	Janssen Biotech Inc	48
UC	Maintenance (w44)	>1.4	Clinical remission (w30 and 54)		ECLIA	Janssen Biotech Inc	48
C. Vedolizumab
CD	Induction (w2)	>35.2	Biological remission (w6)		ELISA	Leuven assay	90
UC	Induction (w2)	>28.9	Clinical response (w14)		ELISA	Leuven assay	90
UC	Induction (w2)	>23.7	Mucosal healing (w14)		ELISA	Leuven assay	90
CD/UC	Induction (w2)	≥24.5	No drug optimization (within w24)		ELISA	Theradiag	92
UC	Induction (w6)	>20.8	Clinical response (w14)		ELISA	Leuven assay	90
CD/UC	Induction (w6)	≥18.5	No need for extended therapy		ELISA	Theradiag	92
CD/UC	Induction (w6)	>27.5	Sustained clinical response		ELISA	Theradiag	92
CD/UC	Induction (w6)	>18	Mucosal healing (within w54)		ELISA	Theradiag	91
UC	Post-induction (w14)	>12.6	Clinical response (w14)		ELISA	Leuven assay	90
UC	Post-induction (w14)	>17	Mucosal healing (w14)		ELISA	Leuven assay	90
CD	Maintenance (w22)	>13.6	Mucosal healing (w22)		ELISA	Leuven assay	90
CD	Maintenance (w22)	>12	Biological remission (w22)		ELISA	Leuven assay	90
D. Ustekinumab
CD	Post-induction (w8)	>3.3	Clinical remission (w8)		ECLIA	Janssen Biotech Inc	49
CD	Maintenance	>4.5	Endoscopic response		HMSA	Prometheus	89
CD	Maintenance (w24)a	>0.8	Clinical remission (w24)		ECLIA	Janssen Biotech Inc	49
CD	Maintenance (w40)b	>1.4	Clinical remission (w44)		ECLIA	Janssen Biotech Inc	49

^aCombined q8w and q12w;

^bq8w only.

ELISA: enzyme-linked immunosorbent assay; HMSA: homogeneous mobility shift assay; CRP: C-reactive protein, FC: fecal calprotectin; ECLIA: electrochemiluminescence immunoassay; w: week; TDM: therapeutic drug monitoring; CD: Crohn’s disease; UC: ulcerative colitis; CDAI: Crohn’s disease activity index; Ref.: reference.

We aimed to reach a consensus on when and how to utilize TDM of biologic therapies during different phases of the treatment (i.e. induction, post-induction, and maintenance therapy) and sought to identify clinically relevant drug concentrations and ADA thresholds to help physicians apply TDM in clinical practice.

METHODS

We applied a modified Delphi method to establish consensus similar to that described in the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) program.¹⁰⁸ A comprehensive literature review was performed regarding the use of TDM of biologic therapies in IBD using PubMed and Medline databases. We utilized the search terms: ‘inflammatory bowel disease’; ‘Crohn’s disease’; ‘ulcerative colitis’; ‘anti-drug antibodies’; ‘therapeutic drug monitoring’ AND ‘infliximab’ OR ‘adalimumab’ OR ‘certolizumab pegol’ OR ‘golimumab’ OR ‘vedolizumab’ OR ‘ustekinumab’. The literature was then presented to a panel of 13 international IBD specialists. Subsequently, based on this review, 28 statements were formulated (K.P., A.S.C, C.A.S.) describing when and how to apply TDM in clinical practice. An Expert Consensus Development Meeting consisting of members of the BRIDGe group (www.BRIDGeIBD.com) and TDM specialists was held in New Orleans, on December 9, 2017, to refine and vote anonymously on the statements. Each statement was rated on a scale of 1 to 10 (1=strongly disagree, 10=strongly agree). Statements were accepted if 80% or more of the participants agreed with a score ≥7. If less than 80% of the panelists agreed with a score ≥7, statements were discussed and revised based on the available evidence followed by a second round of voting. The word ‘appropriate’ was used for each statement to suggest that application of TDM for treatment optimization in a particular clinical scenario is a good option. However, these are not recommendations applicable to every patient.

RESULTS

The panel reached consensus on 24 out of 28 (86%) statements (Tables 4 and 5).

Table 4.

Scenarios of applying therapeutic drug monitoring of biological therapy in patients with inflammatory bowel disease.

Statement	Vote agreement, %
A. Anti-TNFs
1. It is appropriate to order drug/antibody concentration testing, in responders at the end of induction for all anti-TNFs.	92 (12/13)
2. It is appropriate to order drug/antibody concentration testing at least once during maintenance for patients on all anti-TNFs.	100 (13/13)
3. It is appropriate to order drug/antibody concentration testing of anti-TNFs at the end of induction in primary non-responders.	100 (13/13)
4. It is appropriate to order drug/antibody concentration testing for all anti-TNFs, in patients with confirmed secondary loss of response.	100 (13/13)
B. Vedolizumab
5. It is appropriate to order drug/antibody concentration testing for vedolizumab, in responders at the end of induction.	15 (2/13)a
6. It is appropriate to order drug/antibody concentration testing at least once during maintenance for patients on vedolizumab.	46 (6/13)a
7. It is appropriate to order drug/antibody concentration testing for vedolizumab in non-responders at the end of induction.	92 (12/13)
8. It is appropriate to order drug/antibody concentration testing for vedolizumab, in patients with confirmed secondary loss of response.	83 (10/12)a
C. Ustekinumab
9. It is appropriate to order drug/antibody concentration testing for ustekinumab, in responders at the end of induction.	39 (5/13)a
10. It is appropriate to order drug/antibody concentration testing at least once during maintenance for patients on ustekinumab.	31 (4/13)a
11. It is appropriate to order drug/antibody concentration testing for ustekinumab in non-responders at the end of induction (at 8 weeks).	92 (12/13)
12. It is appropriate to order drug/antibody concentration testing for ustekinumab, in patients with confirmed secondary loss of response.	83 (10/12)a

^aAfter a second round of voting.

TNF: tumor necrosis factor

Table 5.

Biological drug concentrations and anti-drug antibodies when applying therapeutic drug monitoring in inflammatory bowel disease.

Statement	Vote agreement, %
A. General
13. There is no difference in indication for ordering drug/antibody concentrations or interpretation of results for biosimilars or the originator drug.	100 (13/13)
14. The threshold drug concentration may vary depending on disease phenotype and desired therapeutic outcome.	100 (13/13)
15. In the presence of adequate trough drug concentrations, anti-drug antibodies are unlikely to be clinically relevant.	100 (12/12)
16. Other than for anti-infliximab antibodies, there are not enough data to recommend a threshold for high anti-drug antibody titers for the biologic drugs.	100 (12/12)
B. Infliximab
17. The current evidence suggests that the variability of infliximab concentrations between the different assays is unlikely to be clinically significant.	100 (13/13)a
18. There is insufficient evidence that inter-assay drug concentration results are comparable for biologic drugs other than for infliximab.	100 (13/13)
19. The minimal trough concentration for infliximab post-induction at week 14 should be greater than 3 μg/ml, and concentrations greater than 7 μg/ml are associated with an increased likelihood of mucosal healing.	100 (13/13)
20. During maintenance the minimal trough concentration for infliximab for patients in remission should be greater than 3 μg/ml. For patients with active disease infliximab should generally not be abandoned unless drug concentrations are greater than 10 μg/ml.	92 (12/13)
21. In the absence of detectable infliximab, high titer anti-infliximab antibodies require a change of therapy. Low level antibodies can sometimes be overcome. For the ANSER assay, a high titer anti-infliximab antibody at trough is defined as 10 U/ml, for RIDAscreen the cut-off is 200 ng/ml, for InformTx/Lisa Tracker the cut-off is 200 ng/ml. For other assays, there is insufficient data to define an adequate cut-off for a high titer anti-infliximab antibody.	100 (13/13)
C. Adalimumab
22. The minimum drug concentration at week 4 for adalimumab should at least be 5 μg/ml. Drug concentrations greater than 7 μg/ml are associated with an increased likelihood of mucosal healing.	83 (10/12)a
23. During maintenance the minimum trough concentration for adalimumab for patients in remission should be greater than 5 μg/ml. For patients with active disease adalimumab should generally not be abandoned unless drug concentrations are greater than 10 μg/ml.	100 (12/12)
D. Certolizumab pegol
24. The minimum concentrations for certolizumab pegol at week 6 should be greater than 32 μg/ml.	100 (12/12)
25. During maintenance the minimum trough concentration for certolizumab pegol for patients in remission should be 15 μg/ml.	92 (11/12)
E. Golimumab
26. The minimum drug concentration at week 6 for golimumab should at least be 2.5 μg/ml.	92 (11/12)
27. During maintenance the minimum trough concentration for golimumab for patients in remission should be greater than 1 μg/ml.	92 (11/12)
F. Vedolizumab / Ustekinumab
28. Although there are emerging data that may show an association between drug concentrations and outcomes, they are not sufficient to guide specific induction and maintenance drug concentrations for vedolizumab and ustekinumab other than confirming that there is detectable drug.	100 (12/12)

^aAfter a second round of voting.

HMSA: homogeneous mobility shift assay; TNF: tumor necrosis factor.

Scenarios when TDM of biologic therapies should be performed

Anti-TNF therapy

Based on the literature review, consensus was reached on all 4 statements regarding anti-TNFs (Table 4A).

• 1It is appropriate to order drug/antibody concentration testing in responders at the end of induction for all anti-TNFs.
• 2It is appropriate to order drug/antibody concentration testing at least once during maintenance for patients on all anti-TNFs.
• 3It is appropriate to order drug/antibody concentration testing of anti-TNFs at the end of induction in primary non-responders.
• 4It is appropriate to order drug/antibody concentration testing for all anti-TNFs in patients with confirmed secondary loss of response.

Numerous studies have demonstrated a positive correlation between anti-TNF drug concentrations and favorable therapeutic outcomes (Table 1, Table 2A and 2B, supplementary table 1). However, the great majority of TDM studies refer to infliximab. A large retrospective study showed that at least one TDM, either proactive and/or reactive of infliximab compared to lack of any TDM was associated with less treatment failure.¹⁰⁹ Several studies have shown that reactive TDM can better identify the cause and consequently manage SLR to anti-TNF therapy, although the data for PNR are more scarce.^{4, 8, 10, 110, 111} Reactive TDM to guide infliximab dose adjustment compared to clinical decision making alone is associated with higher post-adjustment clinical response and endoscopic remission and fewer hospitalizations.³⁷ Moreover, reactive TDM of infliximab was found more cost-effective than utilizing clinical symptoms alone to guide therapeutic decisions.^{99, 101, 102, 112}

Proactive TDM of infliximab compared to empiric dose escalation and/or reactive TDM was found to be associated with increased drug retention.³⁹ The landmark randomized controlled trial (RCT), Trough Concentration Adapted Infliximab Treatment (TAXIT), despite failing to meet its primary endpoint, showed that proactive TDM of infliximab compared to clinically-based dosing was associated with lower frequency of undetectable drug concentrations and lower risk of relapse.¹⁰⁴ Additionally, in patients with CD and subtherapeutic drug concentrations a one-time dose optimization improved clinical remission rates and C-reactive protein.¹⁰⁴ Furthermore, proactive compared to reactive TDM of infliximab was associated with greater drug durability, less need for IBD-related surgery or hospitalization, and lower risk of antibodies to infliximab or serious infusion reactions.³⁸ Recently, proactive following reactive TDM of infliximab was found to be associated with greater drug persistence and fewer IBD-related hospitalizations than reactive TDM alone.¹⁰³ Proactive TDM can also efficiently guide immunomodulator withdrawal in patients on combination therapy. This concept of ‘optimized monotherapy’ was introduced in a retrospective study showing that patients with infliximab concentrations ≥5 μg/mL had similar drug persistence when treated with infliximab monotherapy or combination therapy with an immunomdulator⁵ and is further supported by a recent post-hoc analysis of the RCT Study of Biologic and Immunomodulator Naïve Patients in Crohn’s Disease (SONIC) which demonstrated that patients stratified by infliximab trough quartiles had comparable outcomes regardless of concomitant azathioprine.¹¹³

Vedolizumab

Consensus was reached on only 2 out of 4 statements regarding vedolizumab (Table 4B).

• 5It is appropriate to order drug/antibody concentration testing for vedolizumab in non-responders at the end of induction.
• 6It is appropriate to order drug/antibody concentration testing for vedolizumab in patients with confirmed secondary loss of response.

The current evidence supporting the role of TDM regarding vedolizumab derives only from exposure-response relationship studies showing that higher vedolizumab concentrations are associated with better therapeutic outcomes (Table 2C).^{90–92, 114} In particular, a large single-center retrospective cohort study of 179 patients (66 with UC and 113 with CD) showed that higher vedolizumab trough concentrations at week 2 and 6 were associated with a higher probability of attaining endoscopic healing, clinical response and biologic response or remission assessed at week 14 for UC and week 22 for CD.⁹⁰ A multi-center prospective observational study identified a vedolizumab trough concentration cut-off of 18 μg/mL at week 6 as the only independent variable associated with mucosal healing within the first year of treatment.⁹¹ Currently, there are no studies comparing either proactive or reactive TDM with symptom-based vedolizumab optimization.

Ustekinumab

Consensus was reached on only 2 out of 4 statements regarding ustekinumab (Table 4C).

• 7It is appropriate to order drug/antibody concentration testing for ustekinumab in non-responders at the end of induction (at 8 weeks).
• 8It is appropriate to order drug/antibody concentration testing for ustekinumab in patients with confirmed secondary loss of response.

The current evidence supporting the role of TDM regarding ustekinumab is based on two exposure-response relationship studies showing that higher ustekinumab concentrations correlate to better therapeutic outcomes (Table 2D).^{49, 89} At this time, there are still no studies comparing either proactive or reactive TDM with empiric ustekinumab optimization.

Assays, drug concentrations and anti-drug antibodies

General

Consensus was reached on all 4 statements regarding the use of biologic drug concentrations and anti-drug antibodies (Table 5A).

• 9There is no difference in indication for ordering drug/antibody concentrations or interpretation of results for biosimilars or originator drug.

Current data suggest that infliximab enzyme-linked immunosorbent assay (ELISA)s for evaluating either drug concentrations or ATI are suitable for monitoring the infliximab biosimilars SB2 and CT-P13.^115–118

• 10The threshold drug concentration may vary depending on disease phenotype and desired therapeutic outcome.

Numerous studies have shown an association between higher induction or maintenance biologic drug concentrations and favorable therapeutic outcomes in IBD (Tables 1 and 2, supplementary table 1). Current exposure-response relationship studies suggest that biologic drug concentration thresholds and ranges appear to differ depending on treatment goals and/or disease phenotypes. In general, higher drug concentrations tend to be associated with more stringent outcomes and higher drug concentrations appear to be needed for phenotypes with a higher inflammatory burden, such as fistulising CD (Tables 1 and 2, supplementary table 1, Figure 1).

Figure legend 1.

Infliximab (A)^{13, 17, 20, 40–43, 45, 46, 53, 55, 59–61, 64, 67} and adalimumab (B)^{9, 11–13, 15, 16, 18–23, 30, 31} concentration thresholds associated with biological (based on CRP), biochemical (based on FC), endoscopic or histologic remission in inflammatory bowel disease. Box whisker plots show the median (solid line within box), interquartile range (upper and lower box boundaries) and 5–95% lower and upper extreme (whiskers).

IFX: infliximab; ADM: adalimumab; CRP: C - reactive protein; FC: fecal calprotectin.

• 11In the presence of adequate trough drug concentrations, anti-drug antibodies are unlikely to be clinically relevant.

A study from Steenholdt et. al. showed that most antibodies to infliximab (ATI) detected via the drug tolerant homogeneous mobility-shift assay (HMSA) lack neutralising potential when tested via a functional cell-based reporter-gene assay, suggesting that they may not be clinically significant.¹¹⁹ A post-hoc analysis of the TAXIT study, which investigated the additional benefit of a drug-tolerant assay, concluded that although it allowed closer follow-up of ATI concentrations and identification of true transient versus persistent antibodies, it offered no clinical benefit over a drug-sensitive assay.¹²⁰ Nevertheless, other studies have suggested that ‘double positive’ patients (with positive ATI and drug on board) may be prone to SLR or lack of mucosal healing.^{60, 67, 121}

• 12Other than for anti-infliximab antibodies, there are not enough data to recommend a threshold for high anti-drug antibody titers for the biologic drugs.

Numerous studies have shown that ADA are associated with sub-therapeutic drug trough concentrations, loss of response and lack of recapture of response following dose escalation (Table 3).^{10, 12–17, 21, 23, 27–29, 31–33, 37, 56–58, 60, 63, 67, 73, 75, 80–88} However, the great majority of them and specifically the ones suggesting a threshold of high-titer ADA refer to ATI (Table 3).

Infliximab

Consensus was reached on all statements regarding infliximab concentrations and ATI (Table 5B).

• 13The current evidence suggests that the variability of infliximab concentrations between the different assays is unlikely to be clinically significant.
• 14There is insufficient evidence that inter-assay drug concentration results are comparable for biologic drugs other than for infliximab.

Current evidence suggests that although absolute drug concentrations can differ between different assays, including the commonly used ELISA, radio-immunoassay, HMSA and the recently developed electrochemiluminescence immunoassay, they correlate well and generally lead to the same therapeutic decision.^{83, 119, 122–124} However, these data refer mostly to infliximab, while there are only scarce data for adalimumab and none for non-anti-TNF agents.

• 15The minimal trough concentration for infliximab post-induction at week 14 should be greater than 3 μg/ml, and concentrations greater than 7 μg/ml are associated with an increased likelihood of mucosal healing.
• 16During maintenance the minimal trough concentration for infliximab for patients in remission should be greater than 3 μg/ml. For patients with active disease infliximab should generally not be abandoned unless drug concentrations are greater than 10 μg/ml.

These drug concentration thresholds were mainly based on infliximab exposure-response relationship studies depicted in supplementary table 1.

• 17In the absence of detectable infliximab, high titer anti-infliximab antibodies require a change of therapy. Low level antibodies can sometimes be overcome. For the ANSER assay, a high titer anti-infliximab antibody at trough is defined as 10 U/ml, for RIDAscreen the cut-off is 200 ng/ml, for InformTx/Lisa Tracker the cut-off is 200 ng/ml. For other assays, there is insufficient data to define an adequate cut-off for a high titer anti-infliximab antibody.

Differences in assay methodology result in varying sensitivity to detect ADA and discrepancies when reporting ADA titers.¹²³ Therefore, clinically relevant ADA cut-offs are assay specific, referring mostly to ELISAs and the HMSA (Table 3). Vande Casteele et al, showed that ATI >9.1 U/ml (measured with the HMSA) at time of loss of response resulted in a likelihood ratio of 3.6 for an unsuccessful intervention, suggesting these ATI are sustained and probably very hard to overcome.⁶³ Moreover, Yanai et al. showed ATI >9 μg/mL-eq can identify patients who do not respond to an increased drug dosage with 90% specificity.¹⁰Additionally, a small retrospective study of IBD patients in whom infliximab was optimized, either proactively or reactively, to overcome immunogenicity showed that an ATI titer < 8.8 U/mL (measured with the HMSA) was associated with drug retention, suggesting that lower titer ATI can often be overcome with dose intensification.⁸⁶ A post-hoc analysis of the TAXIT trial showed that ATI> 222 ng/mL eq (measured with an in-house developed drug tolerant ELISA) was not possible to be overcome following infliximab optimization.¹²⁰

Adalimumab

Consensus was reached on all 2 statements regarding adalimumab concentrations and antibodies to adalimumab (Table 5C).

• 18The minimum drug concentration at week 4 for adalimumab should at least be 5 μg/ml. Drug concentrations greater than 7 μg/ml are associated with an increased likelihood of mucosal healing.
• 19During maintenance the minimum trough concentration for adalimumab for patients in remission should be greater than 5 μg/ml. For patients with active disease adalimumab should generally not be abandoned unless drug concentrations are greater than 10 μg/ml.

These drug concentration thresholds were based mainly on adalimumab exposure-response relationship studies depicted in Table 1.

Certolizumab pegol

Consensus was reached on all 2 statements regarding certolizumab pegol concentrations and antibodies to certolizumab pegol (Table 5D).

• 20The minimum concentrations for certolizumab pegol at week 6 should be greater than 32 μg/ml.
• 21During maintenance the minimum trough concentration for certolizumab pegol for patients in remission should be 15 μg/ml.

These drug concentration thresholds were based on certolizumab pegol exposure-response relationship studies depicted in Table 2A.

Golimumab

Consensus was reached on all 2 statements regarding golimumab concentrations and antibodies to golimumab (Table 5E).

• 22The minimum drug concentration at week 6 for golimumab should at least be 2.5 μg/ml.
• 23During maintenance the minimum trough concentration for golimumab for patients in remission should be greater than 1 μg/ml.

These drug concentration thresholds were based on exposure-response relationship studies depicted in Table 2B.

Vedolizumab and ustekinumab

Consensus was reached on the statement regarding vedolizumab and ustekinumab concentrations and antibodies to vedolizumab or ustekinumab (Table 5F).

• 24Although there are emerging data that may show an association between drug concentrations and outcomes, they are not sufficient to guide specific induction and maintenance drug concentrations for vedolizumab and ustekinumab other than confirming that there is detectable drug.

At the time of the consensus meeting there were only limited data available from exposure-response relationship studies to suggest a clinically relevant vedolizumab (Table 2C) or ustekinumab (Table 2D) threshold or range associated with favorable therapeutic outcomes.

DISCUSSION

Unlike for rheumatoid arthritis and psoriasis, there are only a limited number of biologic agents approved for the treatment of IBD. Additionally, current data demonstrate that patients who fail anti-TNF therapies do no respond as well to subsequent agents.^{125, 126} Thus, optimizing the use of biologic therapies is of the utmost importance. TDM is one strategy to optimize biologics and maximise their effectiveness. Reactive TDM can better explain and manage SLR, and there is emerging evidence that proactive TDM further improves outcomes and is being used more frequently.^{127, 128}

In the recent American Gastroenterological Association guidelines, no recommendation was made regarding proactive TDM of anti-TNFs for patients who have quiescent disease due to a ‘knowledge gap’.⁹⁶ However, the IBD Sydney Organisation and the Australian Inflammatory Bowel Diseases Consensus Working Group recommended that in patients in clinical remission following anti-TNF therapy induction, TDM should be considered to guide management and additionally TDM should be considered periodically in patients in clinical remission if the results are likely to impact management.⁹⁷ Although well designed large prospective studies are lacking there are preliminary data mainly from retrospective studies which demonstrate that proactive TDM is associated with better therapeutic outcomes compared to empiric dose optimization and/or reactive TDM.^{38, 39, 103, 104, 129} Furthermore, numerous retrospective studies^{23, 24, 26, 29, 31–33, 67, 73, 74, 77–79, 130, 131} and some post-hoc analyses of RCT^{47–49, 71, 76, 94, 132, 133} have shown that higher biologic drug concentrations are associated with favourable short- and long-term therapeutic outcomes in IBD (supplementary Table 1, Table 2 and Table 3). There do appear to be certain clinical scenarios that proactive TDM of anti-TNF therapy can efficiently guide therapeutic decisions, such as treatment de-escalation,¹³⁴ the application of ‘optimized monotherapy’ instead of combo therapy with IMM,⁸² re-starting therapy after a long ‘drug holiday’¹³⁵ and treatment cessation upon deep remission.^{50, 51}

Nevertheless, before TDM can be widely applied in clinical practice there are several obstacles to their regular use including when to utilize TDM, how to accurately interpret and apply the results of such testing, and in defining the optimal drug concentration thresholds and ranges to target.¹³⁶ We feel these consensus statements help address these issues and hope they will aid physicians in better understanding and utilizing TDM.

Major limitations of the evidence and consequently these consensus statements relate to the lack of large prospective studies and RCT on TDM of biologic therapy applied on different IBD phenotypes, and sparse data on induction therapy and on biologic agents other than infliximab and adalimumab. Moreover, it is unclear if trough concentrations are the best predictor of initial response to biologics, compared to peak drug concentrations or total drug exposure. However, in the absence of RCT, consensus guidelines synthesizing the literature and extrapolating from available data serve to support clinicians in clinical decision making.

Further RCT to establish the utility of proactive TDM, particularly during the induction phase, should be performed. Additional future directions should include the development of accurate, easily accessible and affordable rapid assays and dashboards to allow fast dosing adaptation and incorporation of predictive PK models based on patient and disease characteristics.^{137, 138}

In conclusion, there is a growing body of evidence that demonstrates the clinical utility of TDM of biologic therapy in IBD. This is a big step towards personalised medicine and optimizing the care of patients with IBD. Although more prospective data are needed especially for proactive TDM, induction therapy, and non-anti-TNF biologics, these consensus statements provide a practical guide to apply TDM for optimizing biologic therapy in patients with IBD.

ABBREVIATIONS:

ADA

anti-drug antibodies

ATI

antibodies to infliximab

CD

Crohn’s disease

CI

confidence interval

ELISA

enzyme-linked immunosorbent assay

HMSA

homogeneous mobility shift assay

IBD

inflammatory bowel disease

IMM

immunomodulator

TDM

therapeutic drug monitoring

TNF

tumor necrosis factor

UC

ulcerative colitis

PNR

primary non response

SLR

secondary loss of response

PK

pharmacokinetic

PD

pharmacodynamic

RCT

randomized controlled trial