Persistence of Second and Third-Line Biologics in Inflammatory Bowel Disease: A Multi-Centre Cohort Study
, Robbie Chan 1, Daniel Tassone 2, Nik S. Ding 2,3, Chamara Basnayake 2,3, Julien Schulberg 2, Abhinav Vasudevan 1, Michael Kamm 2,3, Michael De Gregorio 2,3, Daniel R. van Langenberg 1,4 and Ola Niewiadomski 1,4,*
Bénédicte Caron
Round 1
Reviewer 1 Report
The present study is a retrospective study of patients receiving second and/or third-line biologics for inflammatory bowel disease from 2005-2021. The authors identified early predictors of second and/or third-line biologic persistence in IBD.
One hundred and seventy-nine patients were included. There was a significantly increased likelihood of longer treatment persistence in recipients who received an anti-TNF first, versus those that received a non-anti-TNF agent first (p<0.01). A diagnosis of CD (p<0.01), and endoscopic remission achieved on the first biologic (p = 0.03) were positive predictors of longer biologic persistence, whilst advancing age at IBD diagnosis (p= 0.04) and primary non-response to initial biologic (p<0.01) were inversely associated with biologic persistence.
The authors conclude that a diagnosis of CD opposed to UC, and the achievement of endoscopic remission during first-line biologic therapy are independently associated with increased likelihood of persistence of two or more biologics over time.
It is an original well written comprehensive manuscript.
There are some aspects that need to be addresses.
- Abstract: The conclusion could be modified, it does not highlight the interesting and main results of this study.
- Methods: Could we have more details on the inclusion criteria? Did the patients have to receive a minimum of 2 lines or a minimum of 3 lines of treatment? Could they be included if they had received more than 3 lines of treatment?
- Table 1 : What point is considered the baseline? Is it the induction of the 1st line of biotherapy or the 2nd line?
- Figure 1 : Would it be possible to have also the data separately for CD and UC?
- Results : What was the median follow-up of patients?
- Figure 2 : It would be interesting to add to this figure a subgroup analysis according to the first line of biotherapy, anti TNF versus other biologics.
- Results : The authors evaluated variables potentially associated with longer treatment persistence to biologic threrapy. What is the precise definition of persistence? The duration of all treatment with biologics? Only the 2nd line? or the 3rd line?
- Did the authors perform a subgroup analysis by treatment period, such as before 2015 versus after 2015?
Author Response
Please see attached document
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In this study, Timothy P Hanrahan, et al. described early predictors of second and third line biologic persistence in Inflammatory Bowel Disease. They conclude that diagnosis of CD and endoscopic remission during first biologic therapy increased likelihood of persistence of two or more biologics over time.
Strengths of the study:
- Study question is valid
- Adequate literature review was performed.
The manuscript can be improved by addressing following concerns.
- Table 2 is difficult to follow. Authors may want to add a paragraph to discuss Table 2 findings clearly.
- Define what is objectively demonstrated biologic failure and how is it different from pharmacodynamic failure/primary non-response.
- Authors may want to mention what were the non-disease related reasons for treatment cessation
- Univariate analysis indicates that those with Crohn’s Disease had significantly reduced persistence compared to those with Ulcerative colitis (Figure 5). However, multivariate analysis findings suggest a diagnosis of CD is associated with longer persistence. Authors may want to explain this disparity.
- Did authors study dietary patterns (eg., enteral nutrition, SCD diet) of their study population?
Author Response
Please see attached document
Author Response File: Author Response.pdf
Reviewer 3 Report
This study by Timothy P Hanrahan and colleagues analizes the response to second and third line biologic in a population with Inflammatory Bowel Disease and factors asociated with persistence treatment. The study, although retrospective, give information about the real clinical practice. The strenght of the paper is the relative lack of information published about second and third line biologics in real world. Nevertheless, there are some considerations the authors need to review prior to accept the paper.
Major comments:
1) Introduction:
¨from a patient´s perspective, it is the collective…. Robust quality of life¨à consider to indicate that, although this factor is important in relation to quality of life, patients also consider aspects related to safety when using more than one biologics (information about the expected response could be useful in this sense).
2) Methods:
· Indicate if dose of biologic therapy (ex: infliximab induction) was extracted. If no information regarding dose, consider this as a limitation of the study in the discussion.
· It would be useful to explain what is considered a biochemically failure of response. Example: CRP or calprotectin and levels of no response.
3) Results:
· 179 patients are incuded because of failure of biologic, but only 162 received two or more biological therapies. If possible explain the treatment of the 17 patients that failure biologic but no received another biologic therapy.
· Table 1: 75% of patients of the study have Crohn´s disease. As the study is a retrospective review please explain why the percentage of Crohs´disease (not Ulcerative Colitis) is so high. It would be useful to compare this frecuency with the frecuency of Crohn vs Ulcerative Colitis in Australia.
· Baseline characteristic of the cohort (Table 1) showed that 40% have a prior surgical intervention. Please, explain if it is a digestive intervention related to Inflammatory Bowel disease or another surgical intervention with no relation with the disease.
· Discussion:
Page 11, line 94: regarding polymorphisms related with non response to biological therapy consider another reference, not only about IgG Fc receptor. As an example: Genetic Predictors of Long-term Response to Antitumor Necrosis Factor Agents in Pediatric Inflammatory Bowel Disease. J Pediatr Gastroenterol Nutr. 2020 Oct;71(4):508-515. doi: 10.1097/MPG.0000000000002840.
Minor comments:
- Introductionà title is underlined, but no another titles (Methods, results, etc).
- Table 2 à line 3: results of line 2 (columns 2-3-4-5: biologic 1, biologic 2, biologic 3 and p value) are not alined.
- Titles of figure 1, 2 and 5 have a dot at the end. Titles of figure 2 and 5 don´t have it. Please unify.
- Title of figure 4, table 3 and figure 5: there is a double space between lines.
- Figure 3: letters A, B and C for each image is missing
- Line 50: ¨consecutive biologics: diagnosis of CD…¨ (double space after: ).
- Line 76: double space between biologic class y Multivariate.
- Line 142: doublee space between clínical practice. Until…
Author Response
Please see attached document
Author Response File: Author Response.pdf
Reviewer 4 Report
TITLE
OK.
ABSTRACT
Aims: The authors state that the aim of the study was to identify early predictors of second- and/or third-line biologic persistence in IBD. However, the objective of “assessing the relative persistence of sequential biologics in a real-world cohort”, which is included as the first objective in the Introduction section, is not mentioned in the abstract.
Conclusion: The conclusion of the abstract is unclear, please rephrase it. In particular, the term “(non)persistence” is misleading.
More keyword closely related with this article content could be added.
INTRODUCTION
OK.
METHODS
Exclusion criteria (if any) should be clearly defined.
Please, explain “NHMRC” abbreviation.
The definition of “Pharmacodynamic failure” is not according to the standard definition. Several definitions are mixed here, and included under the same term. Please clarify.
Statistical analysis: for percentages, 95% confidence intervals should be provided (at least for the most relevant percentages).
RESULTS
The patients were included in two tertiary hospitals in Melbourne; please indicate the number of patients included in each center; and how many physicians were responsible to include patients in each center.
As previously stated, please provide 95% confidence intervals for the most relevant percentages.
In Results, is stated: “Moreover, in this same subgroup where biologic cessation(s) occurred within the study follow-up period, there was a significant trend of shorter duration with each successive biologic (Figure 2)”. Please clarify if this was just a trend or if this difference achieved statistical significance (as it seems to be the case, Figure 2).
3.2 Factors associated with treatment persistence: The following sentence could be deleted, as it belongs to the Methods section: “To determine variables potentially associated with longer treatment persistence to biologic therapy in this cohort, Cox regression analyses were performed after exploratory Kaplan Meier univariate analysis to inform factor inclusion”.
3.2 Factors associated with treatment persistence: It is difficult to understand how some variables (such as CRP or albumin levels) at baseline can be significantly associated with treatment persistence of the 2nd or 3rd biological therapy.
DISCUSSION
The authors point out that “A large systematic review of 8 randomised controlled trials, inclusive of non-anti-TNF second-line biologic data (ustekinumab), found a 24 percent reduction in response to second-line biologic therapies when compared to those with SLOR”. Please consider adding here a reference of a meta-analysis supporting precisely these findings: Systematic review with meta-analysis: the efficacy of a second anti-TNF in patients with inflammatory bowel disease whose previous anti-TNF treatment has failed. Aliment Pharmacol Ther. 2015 Apr;41(7):613-23. doi: 10.1111/apt.13083. Epub 2015 Feb 4. PMID: 25652884.
In the present study, treatment persistence was longest if the first biologic was an anti-TNF. Please elaborate in the Discussion on the possible reasons for this association.
The limitations of the present study (not only the retrospective design but also additional shortcomings) should be recognized in the Discussion section.
REFERENCES
OK.
TABLES
OK.
FIGURES
All the abbreviations of the figures should be explained in a foot-note of respective figures.
In Figure 3, in all Kaplan-Meier curves, the number of patients at risk al each time point should be included in the horizontal axis.
Author Response
Please see attached document
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
No additional comment
Reviewer 2 Report
Authors addressed my concerns. I have no issues publishing this manuscript if other reviewers concerns are addressed.
Reviewer 4 Report
.
