Embed Size (px)
Citation preview
February 21
NIZO-REPORT E 2019 / MF
EFFECT OF LACTOBACILLUS LB PRODUCT ON GUT MICROBIOTA
Authors: Kardinaal, A., Kortman, G., Boekhorst, J., Floris-Vollenbroek, E., Lucas – van den Bos, E., van Schalkwijk, S., Schloesser, J.
February 21
NIZO-REPORT E 2019 / MF
EFFECT OF LACTOBACILLUS LB PRODUCT ON GUT MICROBIOTA IN HEALTHY VOLUNTEERS
AUTHORS: Kardinaal, A., Kortman, G., Boekhorst, J., Floris-Vollenbroek, E., Lucas – van den Bos, E., van Schalkwijk, S., Schloesser, J.
NIZO-REPORT E 2019 / MF TITLE
CONFIDENTIAL Effect of Lactobacillus LB product on gut microbiota In healthy volunteers
ON BEHALF OF: Adare Pharmaceuticals Inc.
CONTACT PERSON: Dr. Peter C. Richardson
PERIOD OF RESEARCH:
AUTHORS: Kardinaal, A., Kortman, G., Boekhorst, J., Floris-Vollenbroek, E., Lucas – van den Bos, E., van Schalkwijk, S., Schloesser, J.
SIGNED AS CORRECT: February 21
PROJECTNR. 111310 © NIZO food research B.V. (2016). All rights reserved. Without permission in writing from NIZO food research B.V., it is not permitted to publish parts of this report or this report in total any way.
Page 04 of 37
EFFECT OF LACTOBACILLUS LB PRODUCT ON GUT MICROBIOTA IN HEALTHY VOLUNTEERS AUTHORS: Kardinaal, A., Kortman, G., Boekhorst, J., Floris-Vollenbroek, E., Lucas – van den Bos, E., van Schalkwijk, S., Schloesser, J.
KEYWORDS
1. SYNOPSIS
Name of company: Name of finished product: Name of active substance(s):
Adare Prarmaceuticals Inc. Lactobacillus LB product The product is a combination of LB strains (Lactobacillus fermentum and Lactobacillus delbrueckii) and fermented culture medium (neutralized).
Title of the trial:
SOLVE Study: Effect of Lactobacillus LB product on gut microbiota in healthy volunteers
Investigators: Alwine Kardinaal, Guus Kortman Study center: NIZO Study period: February – December 2019 Objectives: Primary: To explore the effect of 5 weeks daily consumption of
Lactobacillus LB product (a spray-dried, heat-killed Lactobacillus LB with its spent culture supernatant), compared to a placebo supplement, on fecal microbiota composition and diversity in healthy individuals Exploratory: To explore the effect of 5 weeks daily consumption of Lactobacillus LB product on plasma CRP as a marker of infectious or inflammatory events
Methodology/Design: Randomised, double-blind, placebo-controlled 2-period cross-over trial
Number of subjects: 30 Main criteria for inclusion: Age ≥18 and ≤65 years.
BMI ≥18.5 and ≤30.0 kg/m2. Healthy as assessed by the NIZO health questionnaire.
Test product, dose, mode of administration:
4 capsules (size 2, color: orange/green) per day, containing 170 mg (± 10%) per capsule. At least 80% of the product existed of heat-killed Lactobacillus LB strains and fermentate; lactose was used as carrier. Intake: 2 capsules in the morning, 2 in the evening.
Reference therapy, dose, mode of administration:
4 capsules with 170mg sucrose per capsule. Administration schedule same as test product.
Duration of treatment: Treatment periods of 5 weeks and a wash-out period of 2 weeks
Criteria for evaluation: No pre-defined hypothesis was tested in this explorative study. The following results were evaluated:
Page 05 of 37
Fecal microbiota composition and diversity at the end of both treatment periods Change in fecal microbiota composition and diversity during both treatment periods Change in plasma CRP concentration during both treatment periods
Statistical methods: Microbiota: Multivariate: Principal Component Analysis (PCA) and Redundancy Analysis (RDA). The significance of the separation was determined by Monte Carlo permutation test. Univariate: Non-parametric Mann-Whitney (unpaired) and Kruskal-Wallis (paired) tests were applied. P-values are two-tailed and in case of taxa comparisons, corrected for multiple testing using FDR. CRP: ANOVA model including treatment, period, treatment × period interaction, gender and age. Subject was considered as a random effect.
Results:
No consistent significant effects of Lactobacillus LB product on the gut microbiota composition or diversity of healthy volunteers were detected. In period 2, the microbiota change within the Lactobacillus LB group was statistically significant (p=0.026), but not in period 1. The variation in microbiota composition at the end of both treatment periods was explained for 0.66% (p=0.074) by treatment, in the PP-population, but not in the ITT-population.
Date of report Draft 20 December 2019
Page 06 of 37
2. TABLE OF CONTENTS
1. SYNOPSIS ......................................................................................... 4
2. TABLE OF CONTENTS ........................................................................ 6
3. LIST OF ABBREVIATIONS AND DEFINITIONS OF TERMS .................... 9
4. ETHICS ............................................................................................ 10
4.1 INDEPENDENT ETHICS COMMITTEE ......................................................................... 10
4.2 ETHICAL CONDUCT OF THE STUDY ........................................................................... 10
4.3 SUBJECT INFORMATION AND CONSENT ................................................................... 10
5. INVESTIGATORS AND STUDY ADMINISTRATIVE STRUCTURE .......... 10
5.1 SPONSOR ................................................................................................................... 10
5.2 TESTING FACILITIES AND RESPONSIBLE PERSONNEL ................................................ 10
5.3 RETENTION OF SAMPLES AND RECORDS .................................................................. 11
6. INTRODUCTION .............................................................................. 12
7. STUDY OBJECTIVES ......................................................................... 13
8. INVESTIGATIONAL PLAN ................................................................ 13
8.1 OVERALL STUDY DESIGN AND PLAN DESCRIPTION .................................................. 13
8.2 DISCUSSION OF STUDY DESIGN, INCLUDING THE CHOICE OF CONTROL GROUPS .. 14
8.3 STUDY POPULATION ................................................................................................. 14
8.3.1 INCLUSION CRITERIA .............................................................................................. 14
8.3.2 EXCLUSION CRITERIA ............................................................................................. 15
8.3.3 REMOVAL OF SUBJECTS FROM ASSESSMENT ........................................................ 16
8.4 TREATMENTS ............................................................................................................ 16
8.4.1 TREATMENTS ADMINISTERED ............................................................................... 16
8.4.2 INVESTIGATIONAL PRODUCT ................................................................................. 16
Page 07 of 37
8.4.3 NON-INVESTIGATIONAL PRODUCTS ...................................................................... 17
8.4.4 METHOD OF ASSIGNING SUBJECTS TO TREATMENT GROUPS .............................. 17
8.4.5 SELECTION OF DOSES IN THE STUDY ..................................................................... 17
8.4.6 SELECTION AND TIMING OF DOSE FOR EACH PATIENT ......................................... 17
8.4.7 BLINDING ............................................................................................................... 17
8.4.8 PRIOR AND CONCOMITANT THERAPY/TREATMENT ............................................. 18
8.4.9 TREATMENT COMPLIANCE .................................................................................... 18
8.5 EFFICACY AND SAFETY VARIABLES ............................................................................ 19
8.5.1 EFFICACY AND SAFETY MEASUREMENTS ASSESSED AND FLOW CHART ............... 19
8.5.2 PRIMARY EFFICACY PARAMETERS ......................................................................... 19
8.5.3 DRUG CONCENTRATION MEASUREMENTS ........................................................... 19
8.5.4 SECONDARY AND EXPLORATORY PARAMETERS .................................................... 20
8.5.5 SAFETY VARIABLES ................................................................................................. 20
8.6 DATA QUALITY ASSURANCE ...................................................................................... 21
8.7 STATISTICAL METHODS ............................................................................................. 21
8.7.1 STATISTICAL AND ANALYTICAL PLANS ................................................................... 21
8.7.2 DETERMINATION OF SAMPLE SIZE ........................................................................ 23
8.7.3 INTERIM ANALYSIS BY DATA AND SAFETY MONITORING BOARD ......................... 23
8.8 CHANGES IN THE CONDUCT OF THE STUDY OR PLANNED ANALYSES ..................... 23
9. STUDY SUBJECTS ............................................................................ 24
9.1 DISPOSITION OF SUBJECTS ....................................................................................... 24
9.2 PROTOCOL DEVIATIONS ............................................................................................ 25
10. EFFICACY EVALUATION .................................................................. 26
10.1 DATA SETS ANALYSED ............................................................................................. 26
10.2 DEMOGRAPHIC AND OTHER BASELINE CHARACTERISTICS .................................... 26
10.3 MEASUREMENT OF TREATMENT COMPLIANCE ..................................................... 26
10.4 EFFICACY RESULTS .................................................................................................. 27
10.4.1 STATISTICAL/ANALYTICAL ISSUES ........................................................................ 27
10.4.2 PRIMARY OUTCOME: MICROBIOTA RESULTS ...................................................... 27
10.4.3 EXPLORATORY OUTCOME: CRP RESULTS ............................................................ 30
10.4.4 TABULATION OF INDIVIDUAL RESPONSE DATA ................................................... 31
10.4.5 EFFICACY CONCLUSIONS ...................................................................................... 31
Page 08 of 37
11. SAFETY EVALUATION ..................................................................... 32
11.1 ADVERSE EVENTS .................................................................................................... 32
11.1.1 BRIEF SUMMARY OF ADVERSE EVENTS ............................................................... 32
11.1.2 ANALYSIS OF ADVERSE EVENTS ........................................................................... 33
11.1.3 LISTING OF ADVERSE EVENTS BY PATIENT .......................................................... 33
11.2 DEATHS, OTHER SERIOUS ADVERSE EVENTS AND OTHER SIGNIFICANT ADVERSE
EVENTS ............................................................................................................................ 33
11.3 CLINICAL LABORATORY EVALUATION ..................................................................... 33
11.4 VITAL SIGNS, PHYSICAL FINDINGS, AND OTHER OBSERVATIONS RELATED TO
SAFETY ............................................................................................................................. 33
11.5 SAFETY SUMMARY AND CONCLUSIONS ................................................................. 33
12. DISCUSSION AND OVERALL CONCLUSIONS .................................... 33
13. REFERENCES ................................................................................... 35
14. APPENDICES ................................................................................... 36
14.1 MICROBIOTA RESULTS ............................................................................................ 36
14.2 DECLARATION AND SIGNATURE PAGE ................................................................... 36
Page 09 of 37
3. LIST OF ABBREVIATIONS AND DEFINITIONS OF TERMS °C : Degrees Centigrade AE : Adverse Event ANOVA : Analysis of Variance BMI : Body Mass Index CCMO : Centraal Comité voor Mensgebonden Onderzoek (Central Committee on Research
involving Human Subjects) CFU : Colony Forming Units CRF : Case Report Form CRO : Contract Research Organisation EPRO : Electronic Patient Reported Outcome FSI : First subject in g : Gram GCP : Good Clinical Practice ICH : International Conference on Harmonisation of Technical Requirements for
Registration of Pharmaceuticals for Human Use ITT : Intention-to-treat IRB : Institutional Review Board or Medical Ethics Committee (also Medical Ethics
Approval Committee / Institutional Review Board) LSO : Last subject out MEC : Medical Ethics Committee ml : Milli liter PCR : Polymerase Chain Reaction PP : Per protocol SAE : Serious Adverse Event SD : Standard deviation SEM : Standard Error of the Mean SOP : Standard Operating Procedure WMO : Wet Medisch Onderzoek met Mensen (Dutch Medical Research involving Human
Subjects Act)
Page 10 of 37
4. ETHICS
4.1 INDEPENDENT ETHICS COMMITTEE The protocol was submitted to the Independent Review Board Nijmegen, the Netherlands and registered as Protocol NL68859.072.19. Approval by the MEC was given on 21 March 2019.
4.2 ETHICAL CONDUCT OF THE STUDY The study was conducted according to:
• The current assembly of the Declaration of Helsinki (Fortaleza, Brazil, October 2013); • The requirements as described in the EU Clinical Trials Directive 2001/20/EC transposed in the
Revision of the Dutch Medical Research involving Human Subjects Act (“Wet medisch-wetenschappelijk onderzoek met mensen”, WMO, effective as of March 1, 2006).
• Guideline for Good Clinical Practice (ICH E6 (R2), 2017)
4.3 SUBJECT INFORMATION AND CONSENT The subjects in the study were recruited initially via recruitment websites and Facebook, by Link2Trials B.V., with recruiting texts that had been approved by the MEC. Candidates were pre-screened by telephone interviews. If they were interested and eligible, based on the pre-screening items, they received written information and were invited for an information session at NIZO, in the presence of the Principal Investigator, during which they were informed about the aim, procedures, constraints, the insurance cover and the financial compensation for the study. After the information session, interested candidates were asked to sign an Informed Consent form.
5. INVESTIGATORS AND STUDY ADMINISTRATIVE STRUCTURE
5.1 SPONSOR The study team of the sponsor exists of:
• Alessandro Martini (Associate director, project management) • Peter Richardson (Chief Medical Officer and Vice President R&D) • Karol Knoop (Senior Manager Clinical Operations) • Martin Waga (Senior Manager Clinical Quality Assurance) • Stephen Perrett (Senior Director Research and Development) • Gilles Chauvière (General Manager Responsable Pharmacist) • Massimo Latino (Regulatory Consultant Adare Pharmaceuticals Inc.)
5.2 TESTING FACILITIES AND RESPONSIBLE PERSONNEL The study team of the testing facilities exists of:
• Alwine Kardinaal (Principal Investigator) • Joyce Schloesser (Project manager) • Esther Floris (Study coordinator) • Guus Kortman (Microbiota expert) • Ben Witteman (Medical investigator) • Maartje van den Belt (Blinded person) • Petra Scholtens (Study execution)
Page 11 of 37
• Anita Hartog (Study execution) • Elly Lucas (Lab/Sample coordination), • Saskia van Schalkwijk (DNA isolation/16S sequencing), • Simon Klaassen (Recruitment) • Karin Nijssen (Monitoring) • Marieette Nelissen (IP handling) • Miep Troost & Lize van Heerder (Study nurse blood sampling)
5.3 RETENTION OF SAMPLES AND RECORDS All essential documents, in hard copy and as an e-file will be retained in the archives of NIZO during 15 years after reporting of the study.
Page 12 of 37
6. INTRODUCTION Adare Pharmaceuticals (Adare) aims to investigate the changes in gut microbiome composition that may occur in healthy individuals following administration of the Lactobacillus LB product. The product is a combination of LB strain (Lactobacillus fermentum and Lactobacillus delbrueckii) and fermented culture medium (neutralized), is used in addition to rehydration and/or dietary measures, as a symptomatic supplemental treatment for diarrhea in adults. A pharmaceutical grade of this product, is approved and sold in European, Latin American, African, and Asian markets as an Rx and OTC drug. Rotavirus is the most common pathogen of infectious watery diarrhea, followed by enterotoxigenic Escherichia coli and other bacteria. The use of a heat-killed Lactobacillus LB with its spent culture supernatant in the treatment of acute diarrhea has been reported effective in promoting a more rapid recovery from acute, watery diarrhea. Liévin-Le Moal (2016) recently reviewed the evidence from clinical and preclinical studies. Non-RCTs and RCTs have shown the therapeutic efficacy of heat-treated and lyophilized Lactobacillus LB cells and culture medium (Drug) in conjunction with ORS for the treatment of rotavirus-induced, well-established acute watery diarrhea in children. Consistent with the in vitro and in vivo antibacterial activities of Lactobacillus LB against the major diarrhoea-associated bacterial pathogens, two RCTs have shown that heat-treated Lactobacillus LB cells and culture medium treatment was therapeutically effective in children with established enterovirulent bacteria-induced diarrhea (Salazar-Lindo et al, 2007; Liévin-Le Moal et al, 2007). The mechanisms of action that are assumed to play a role in this effect are the following: Inhibition of adhesion to and invasion of enterocytes by the microorganisms responsible for diarrhea (Ducrotté, 2009; Chauvière et al, 1992 ; Coconnier et al, 1993) Bacteriostatic activity (Ducrotté, 2009; Coconnier et al, 1997) Stimulation of the growth of the acidogenic defense flora (Ducrotté, 2009) Increase of IgA synthesis (Ducrotté, 2009) Lactobacillus LB has shown high affinity in attaching to enterocytes, thus preventing colonization by Gram-positive and Gram-negative organisms. It also generates a heat-stable antimicrobial substance with anti-invasive and antiadhesive activity against intracellular pathogens. This substance released by Lactobacillus LB is effective on a great number of Gram-positive and Gram-negative enterovirulent bacteria, but it is believed to have no inhibiting activity regarding the microflora bacteria such as Bifidobacterium or Lactobacillus. However, this has been studied in in vitro and animal models only, no data are available for the effects that the product has on the gut microbiome in vivo, that may help explain its efficacy in promoting recovery from diarrhea. Studies performed with other Lactobacillus strains (Lahtinen et al, 2012) and fermentates (Berni Canani et al, 2017) have shown that these are able to affect the presence or abundance of specific commensal species in the gut. Berni Canani et al (2017) reported that the use of a fermented milk product containing the heat-killed probiotic strain Lactobacillus paracasei CBAL74 in young, healthy children, induces changes in the gut microbiota, promoting the development of butyrate producers. These changes in the gut microbiota composition correlated with increased levels of innate and acquired immunity biomarkers.
Page 13 of 37
The present study is a first step in exploring the manner in which Lactobacillus LB product affects the gut microbiome in healthy subjects. Gut microbiota composition of the Lactobacillus LB product and placebo groups were compared over time. Next to overall compositional differences, the focus was on the relative abundance of bacteria that are generally considered beneficial (Bifidobacterium and Lactobacillus) and that are potentially pathogenic (Enterobacteriaceae). This research project aims to characterize gut bacterial composition to find additional explanations how Lactobacillus LB product, a suspension containing the bacterial bodies and fermentation products of heat-killed Lactobacillus fermentum and Lactobacillus delbrueckii, may be effective in helping to maintain normal gastrointestinal conditions.
7. STUDY OBJECTIVES This study was explorative in nature. Primary Objective: To explore the effect of 5 weeks daily consumption of Lactobacillus LB product (a spray dried, heat-killed Lactobacillus LB with its spent culture supernatant), compared to a placebo supplement, on fecal microbiota composition and diversity in healthy individuals Exploratory Objective: To explore the effect of 5 weeks daily consumption of Lactobacillus LB product on plasma CRP as a marker of infectious or inflammatory events
8. INVESTIGATIONAL PLAN
8.1 OVERALL STUDY DESIGN AND PLAN DESCRIPTION The study was set up as a randomised, double-blind, placebo-controlled cross-over trial. After a 1-week run-in period, in which all subjects received placebo treatment, subjects were randomised and assigned to a treatment order (placebo – Lactobacillus LB product or Lactobacillus LB product – placebo). The assigned first treatment was consumed on a daily basis during 5 weeks, after which followed a wash-out period of 2 weeks . After the wash-out, the other treatment was consumed, for another 5 weeks. See below a schematic presentation of the study:
Page 14 of 37
Table 1. Schedule of events
8.2 DISCUSSION OF STUDY DESIGN, INCLUDING THE CHOICE OF CONTROL GROUPS The study was designed as a crossover study in healthy volunteers, with a wash-out period of 2 weeks between the treatments. Because the variation in gut microbiota composition between individuals is large, a crossover design has the advantage of being able to compare changes within individuals. This reduces the overall variation, and consequently the number of subjects in the study population required to be able to detect relevant differences. A number of 30 subjects is very common for an exploratory study like this, aiming to look at shifts in microbiota composition and diversity. However, no actual power calculation was performed in advance.
8.3 STUDY POPULATION
8.3.1 INCLUSION CRITERIA In order to be eligible to participate in this study, a subject must meet all of the following criteria: Substantial
1. Age ≥18 and ≤65 years. 2. BMI ≥18.5 and ≤30.0 kg/m2. 3. Healthy as assessed by the NIZO health questionnaire.
Procedural:
4. Ability to follow Dutch verbal and written instructions. 5. Availability of internet connection. 6. Signed informed consent.
Run-inStudy week 1 2 3 4 5 6 7 8 9 10 11 12 13Visit number 1 2 3 4 5Informed consentIn- & exclusion criteriaDemographic dataHeight and body weightMedical/lifestyle quest.RandomizationDispensing of IP/Placebo POvernight fastReturn of remaining IPBlood sample Fecal sample
Intake IP/PlaceboCompliance monitoringAE/SAE monitoring
Treatment Phase 1 Treatment Phase 2Screening
Page 15 of 37
7. Willing to accept disclosure of the financial benefit of participation in the study to the authorities concerned.
8. Willing to accept use of all encoded data, including publication, and the confidential use and storage of all data for at least 15 years.
9. Willing to comply with study procedures and guidelines, including collection of stool and blood samples.
10. Willingness to abstain from (products containing) probiotics and prebiotics, starting from run-in and during the entire study.
11. Willingness to give up blood donation starting at run-in and during the entire study. 12. Willing to take precautions not to become pregnant during the study period. 13. Willingness to avoid use of dietary fiber supplements along the duration of the study
8.3.2 EXCLUSION CRITERIA A potential subject who meets any of the following criteria was excluded from participation in this study:
Substantial 1. Acute gastroenteritis in the 2 months prior to inclusion. 2. Known history or presence of clinically significant neurologic, hematologic, endocrine,
oncologic, pulmonary, immunologic, genitourinary, gastrointestinal (including bowel preparation for colonoscopy), psychiatric, or cardiovascular disease or any other condition which, in the opinion of the Investigator, would jeopardize the safety of the subject or impact the validity of the study results
3. Any active infections, potentially requiring antibiotic use, at time of screening 4. Excessive alcohol usage (men: >4 consumptions/day or >20 consumptions/week; women: >3
consumptions/day or >15 consumptions/week) or drug (ab)use, and not willing/able to stop this during the study.
5. Reported average stool frequency of >3 per day or <1 per 2 days. 6. Having used Lactobacillus LB product/s or (products containing) added pro- and/or prebiotics
within 2 weeks prior to inclusion. 7. Use of antibiotics or regular use of norit (or other activated carbons) or laxatives (in 6 months
prior to inclusion) 8. Reported special diets such as vegetarian, vegan, or macrobiotic. 9. A self-reported lactose intolerance. 10. Pregnancy or lactating
Procedural:
11. Not having a general practitioner, not allowing disclosure of participation to the general practitioner or not allow to inform the general practitioner about abnormal results.
12. Participation in any clinical trial including blood sampling and/or administration of substances starting 1 month prior to study start and during the entire study
Page 16 of 37
13. Personnel of NIZO food research and Adare Pharmaceuticals and their partner and their first and second degree relatives.
8.3.3 REMOVAL OF SUBJECTS FROM ASSESSMENT Subjects could leave the study at any time for any reason if they wished to do so without any consequences. The investigator could decide to withdraw a subject from the study for urgent medical reasons or after severe protocol violation.
8.4 TREATMENTS
8.4.1 TREATMENTS ADMINISTERED Lactobacillus LB product Lactobacillus LB product strain products are derived from co-fermentation of media with Lactobacillus fermentum and Lactobacillus delbrueckii. This “Lactobacillus LB strain” becomes part of the final product, which consists of heat-killed bacterial cells and fermentate. The LB strain in the product is not considered a probiotic because by definition, probiotics are alive. A pharmaceutical grade of this product is approved and sold in European, Latin American, African, and Asian markets as an Rx and OTC drug. For this study, the active product was given in capsules (size 2, color: orange/green), containing 170 mg (± 10%) per capsule. At least 80% of the product will exist of heat-killed Lactobacillus LB strains and fermentate; lactose was used as carrier. Control (Placebo) Capsules were filled with 170mg sucrose as placebo. Sucrose was used which is commercially available for human consumption.
8.4.2 INVESTIGATIONAL PRODUCT Lactobacillus strain LB (Lactobacillus Boucard) was isolated from the stool of a healthy human in 1907. The Lactobacillus LB culture has now been molecularly identified as a culture containing L. fermentum (L. fermentum LB-f) (CNCM I-2998) [Chauvière et al. 2005; Servin et al. 2006] and L. delbruekii (L. delbruekii LB-d) (CNCM I-4831) strains with a ratio of 95:5, respectively, when cultured in both laboratory conditions and industrial batches. Lactobacillus LB product from Adare is produced according to current GMP requirements set to comply with EU standards and the U.S. Code of Federal Regulations, 21 CFR part 117. Lactobacillus LB product strain products are derived from co-fermentation of media with Lactobacillus fermentum and Lactobacillus delbrueckii. This “Lactobacillus LB strain” becomes part of the final product, which consists of heat-killed bacterial cells and fermentate. The LB strain in Lactobacillus LB product strain is not considered a probiotic because by definition, probiotics are alive (AIBMR Life Sciences, 2016). Since 1978, the strain has been utilized to manufacture and market a pharmaceutical-grade strain ingredient
Page 17 of 37
sold as capsules or sachets containing a ratio of 30 billion Lactobacillus LB killed bacterial bodies to 480 mg of fermented culture medium.
8.4.3 NON-INVESTIGATIONAL PRODUCTS Not applicable.
8.4.4 METHOD OF ASSIGNING SUBJECTS TO TREATMENT GROUPS Subjects were stratified according to age and gender. Stratification was performed after enrolment of all 30 subjects, making use of the actual age distribution of the enrolled subjects. Stratification was performed manually using the individual data in Excel. Within strata, subjects were randomly assigned to treatment order. Randomization of subjects to treatment order was performed using the randomization module in ResearchManager. Stratification and randomization of study subjects was done by an unblinded NIZO-project manager outside the study team.
8.4.5 SELECTION OF DOSES IN THE STUDY Lactobacillus LB product and placebo were provided to the subjects as capsules. Subjects will receive 4 capsules per day, divided into two servings (2x2). Capsules contain 170 mg of Lactobacillus LB product or placebo, adding up to a daily intake of 680 mg product. The capsules were provided to the subjects in a pill organizer. Subjects will receive 6 pill organizers, for each intervention period (+ one reserve), each containing a weekly dose. Pill organizers were packaged and labelled in a food grade environment.
8.4.6 SELECTION AND TIMING OF DOSE FOR EACH PATIENT The subjects will consume Lactobacillus LB product (daily exposure of 40 billion killed bacteria bodies [kbb]) or Placebo daily for 5 weeks each. Oral administration. Capsules may be swallowed with half a glass of water in the morning after breakfast and in the evening after dinner. Subjects were asked to store the capsules at room temperature, away from heat and direct light.
8.4.7 BLINDING After arrival of the bulk package at NIZO, the unblinded person from NIZO, who was not part of the NIZO study team, was responsible for labelling of the weekly doses of the study products. For both placebo and IP, this person generated 3 unique number/letter codes, making it impossible for the study team to make inferences about the treatment group. The unblinded person from NIZO, who was not part of the NIZO study team, was responsible for labelling of the study products with subject codes. The NIZO team, laboratory personnel and participants were unaware of the coding until after the blind review meeting and locking of the database. The randomisation code linking the subject to the treatment order was kept by the unblinded person and a back-up person outside the study team. Only in a case of emergency possibly
Page 18 of 37
related to treatment, can the Principal investigator request the independent party to check the type of treatment of (a) particular subject(s) without disclosure of the other treatments.
8.4.8 PRIOR AND CONCOMITANT THERAPY/TREATMENT Probiotics and prebiotics Based on published literature, use of probiotics, probiotic based products or prebiotics may affect the gut microbiota composition. Hence, participating subjects were instructed to avoid probiotics or probiotic based products and prebiotic supplements, starting from the run-in period and during the whole study. Subjects were asked to omit below products in general:
• Probiotics (all kinds) • Yoghurts with active cultures of probiotics • Probiotic based supplements • Products with added prebiotics • Prebiotic supplements
Alcohol and drugs Throughout the trial alcohol was restricted to a maximum of four glasses a day and 20 glasses a week for men, and a maximum of three glasses a day and 15 glasses a week for women. The use of any kind of drugs was forbidden throughout the trial. Use of medication Use of any medication was recorded throughout the trial. General instructions Subjects were instructed to follow their usual lifestyle pattern during the whole study period and not to introduce significant changes in their diet or physical activity pattern. They were requested to report any changes in an online diary.
8.4.9 TREATMENT COMPLIANCE Subjects received 3 pill organizers, for each intervention period (+ one reserve), each containing a 2-weekly dose. Subjects therefore received more capsules than they were required to use. Remaining capsules were handed in and counted at the end of each study period. Treatment compliance was assessed by counting the returned capsules of the test products, calculating the non-used capsules and expressing test product actually used as a percentage of scheduled use. This assessment was leading for reporting of percentage compliance. A second way of assessing compliance was the registration of test product intake by the subject, in the daily online diary. Discrepancies between number of returned capsules and missing IP intake noted in the EPRO were checked with the subject during the visits. All discrepancies were solved. Missing IP intake was reported in the protocol deviation log.
Page 19 of 37
8.5 EFFICACY AND SAFETY VARIABLES
8.5.1 EFFICACY AND SAFETY MEASUREMENTS ASSESSED AND FLOW CHART At visits 2-5, the following measurements were performed (Table 1):
• Assessment of compliance • Measurement of body weight (height was measured at screening) • Handing in of spot fecal sample, collected by the subject at home, on the day (or 2 days) before
the visit and kept frozen during storage and transport • A fasting EDTA blood sample (10 mL) was taken
After centrifugation, plasma aliquots were stored at -80°C until further analysis. Plasma CRP concentrations were determined for all 4 visits. Approximately 2 times 1 gram of stool were collected, of which 200mg was needed for DNA isolation; the remainder was stored for potential exploratory analyses related to study product intake. Subjects reported information on compliance to intake of investigational/placebo product in an online subject diary. Stool consistency (Bristol Stool Score) of the collected stool samples was scored by the subjects in the online diary.
8.5.2 PRIMARY EFFICACY PARAMETERS The following primary outcome measures were defined in the protocol: o Change in fecal microbiota composition as assessed by 16S rDNA Illumina (MiSeq)
sequencing. o Change in fecal microbiota diversity
In order to obtain data on the abundance of microbial species, a bioinformatics pipeline is used that converts sequencing data on the abundance of microbial species. Standard output from the NIZO 16S bioinformatics pipeline, based on the Qiime 1.8 workflow, includes:
• Taxonomic classification of the sequencing reads and their relative abundance per (group of) sample(s) at the Phylum, Class, Order, Family, genus (and species) level.
• A variety of different microbial diversity metrics describing alpha-diversity (within a sample) and beta-diversity (between samples; by weighted and unweighted UNIFRAC analysis) to gain insight in differences between individuals and differences due to treatment.
• Identification and visualization of statistical significant differences in microbial composition at the different taxonomic levels (phylum, class, order, family, species) and at the OTU level to identify unique shifts in microbiota composition in different groups.
8.5.3 DRUG CONCENTRATION MEASUREMENTS Not applicable
Page 20 of 37
8.5.4 SECONDARY AND EXPLORATORY PARAMETERS The following outcome measure was defined in the protocol as exploratory outcome: o Change in plasma CRP concentration in fasting blood samples
High-sensitive CRP was determined by means of ELISA (kit K9710s from ImmunDiagnostik AG, Bensheim, Germany). The kit was validated at NIZO, with good reproducibility (R2>0.995), intra-assay variation: 6-7%; inter-assay variation: 7-10% (within range of kit specifications), and a limit of quantification of 185 ng/mL.
8.5.5 SAFETY VARIABLES Safety was monitored by registration of adverse events. Adverse events were defined as any undesirable experience occurring to a subject during the study, whether or not considered related to treatment. All adverse events reported spontaneously by the subject or observed by the investigator or her staff were recorded. All adverse events were recorded online from the first until the last study visit, using the following software: De research manager, http://deresearchmanager.nl, Deventer, The Netherlands; NEN 7510 and ISO 27001 compliant. Cloud9 is ISO 27001 certified. All AEs were followed until they had abated, or until a stable situation had been reached. AEs were graded as follows:
• Mild: sign or symptom, usually transient, requiring no special treatment and generally not interfering with usual activities.
• Moderate: sign or symptom, which may be ameliorated by simple therapeutic measures, may interfere with usual activity.
• Severe: sign or symptom intense or debilitating and interfering with usual activities without being immediately life-threatening. Recovery is usually aided by therapeutic measures.
In addition, the relation between the AE and the intake of IP was recorded, as well as the relation between the AE and study procedures. These relations were coded as follows:
1. None/not related 2. Unlikely 3. Possible 4. Probable 5. Definite
A serious adverse event was defined as any untoward medical occurrence or effect that: - Results in death; - Is life threatening (at the time of the event); - Requires hospitalisation or prolongation of existing inpatients’ hospitalisation; - Results in persistent or significant disability or incapacity; - Is a congenital anomaly or birth defect; or
Page 21 of 37
- Any other important medical event that did not result in any of the outcomes listed above due to medical or surgical intervention but could have been based upon appropriate judgement by the investigator. An elective hospital admission was not considered as a serious adverse event.
8.6 DATA QUALITY ASSURANCE Study staff was trained for study-specific activities. All samples were collected by trained personnel. Training was documented in training records. Completeness and quality of the data was monitored during 3 study visits (on 28 May, 27 June, 29 August 2019) and one close out visit (24 September 2019) by an external CRA, according to a monitor plan. Monitoring activities included
• Compliance with applicable regulations and Good Clinical Practices • Adherence to the study protocol • Adequacy and accuracy of record keeping and source documentation • Acceptability of storage conditions of clinical samples
The source data verification performed was based on the premise that this was a minimal risk study, and included:
o 100% of Informed Consent forms o Review screening and enrolment logs, for 100% of randomised subjects o 10% SDV for IP accountability o 33% of biological sample accountability o 33% SDV of eCRF o SAE review
Monitor reports were provided and followed up according to the monitor plan.
8.7 STATISTICAL METHODS
8.7.1 STATISTICAL AND ANALYTICAL PLANS Initial analysis during data pre-processing included unsupervised sample clustering (PCA / Unifrac clustering) in order to potentially identify samples that behave markedly different from the majority. In such an event metadata and potential errors during the sample processing were checked. Statistical differences in microbiota composition between groups and over time were assessed by univariate and multivariate statistical methods (see below). Input for the statistical analysis were the relative abundance data as produced in the bioinformatics pipeline, combined with a metadata file that describes sample properties (treatment, clinical parameters and confounders). The crossover design of the current study allows for both cross-sectional and longitudinal statistical analysis. Cross-sectionally, baseline and endpoint microbiota were compared between the treatment groups on both treatment phases by using unpaired statistics. In addition, all end point samples of
Page 22 of 37
both Lactobacillus LB product treatment phases were compared with all end point samples of both placebo treatment phases. This was also done for the baseline samples. Longitudinal analysis involved calculation of relative abundance (2log) ratios between baseline and endpoint samples of both treatment groups. Next, the ratios (change over time) were compared between the treatment groups using paired statistics. In addition, for each treatment phase and each treatment group, baseline samples were compared with the corresponding endpoint samples by applying paired statistics. Multivariate analysis involved both Principal Component Analysis (PCA) and Redundancy Analysis (RDA). PCA provides a first insight into the relative effect size of the different contrasts / confounders, as well as the identification of strong unexpected effects. In a RDA, metadata of choice (e.g. treatment) were explanatory variables and the method tries to separate samples in treatment groups based on common taxonomic patterns in samples of the treatment groups. The significance of the separation is determined by Monte Carlo permutation test. RDA can take into account the variation explained by covariates (confounders), e.g. age, gender etc. The effect of the covariate is then removed from the ordination before the effect of treatment is assessed. Univariate analysis to assess statistical differences between groups is done on the diversity metrics and on the relative abundance of every bacterial taxon. Non-parametric Mann-Whitney (unpaired) and Kruskal-Wallis (paired) tests were applied. P-values are two-tailed and in case of taxa comparisons, corrected for multiple testing using FDR. If samples are missing, cross-sectional analysis can still be performed. If one or more samples of an individual were missing, this subject was removed from longitudinal analysis. Nevertheless, if there is still both a baseline and endpoint sample available of the Lactobacillus LB product treatment phase, it is still possible to study the change over time (together with the other subjects), but which then cannot be directly compared to the placebo group. If metadata is missing, such as Bristol Stool Score, then it cannot be taken along as a confounder, unless the corresponding samples are removed from the analysis. Analysis can be done on all samples excluding the confounder, and/or on the samples only for which the confounding parameter is available. Change in CRP concentration (exploratory outcome) was compared for both treatments. The Kolmogorov–Smirnov test was used to determine whether the data are normally distributed. If they are not normally distributed, 10log-transformation was applied. An ANOVA model was used, including treatment, period, treatment × period interaction, gender (as fixed effects) and age as a continuous covariate. Subject is considered as a random effect. The carryover effect was assessed in the primary model by the treatment × period interaction effect. If the carryover effect was significant, then only the first period was to be considered for the analysis.
Page 23 of 37
P-values <0.05 were considered statistically significant. P-values were two-tailed and in case of taxa comparisons, corrected for multiple testing using FDR (Benjamini-Hochberg procedure).
8.7.2 DETERMINATION OF SAMPLE SIZE We are interested in the changes in fecal microbiota composition and diversity, after longer-term intake of Lactobacillus LB product, but there is no pre-formulated hypothesis to be tested. We therefore consider this an exploratory study. Inter- and intra-individual variation in the response to treatment, observed in this study, can be used in sample size calculation for future microbiota studies with this product. Intestinal microbiota is highly variable between and within individuals (Moore & Stanley, 2016) and responds to environmental changes and diet almost immediately, within 1–3 days. Although there is considerable variability in microbiota composition between people, the microbiota within individuals remained relatively stable, with ~ 70% of strains persisting in the individual over 1 year (Faith et al, 2013). This stable core of strains was present in individuals alongside elements of the microbiota that were more prone to a fast response to stimuli. Furthermore, there is evidence for substantial microbiota variation in each individual, both temporal and spatial. It may seem contradictory to refer to microbiota as stable over time but at the same time highly variable; however, the core microbiota represents a stable component in the community and is often defined as microbiota highly shared between individuals, as opposed to a highly responsive variable component of the community that is capable of mounting a very fast response to any external stimuli and thus allows the community to adapt to new conditions. To restrict the impact of interindividual variation, a crossover design was chosen. In addition, a number of determinants, that have been shown to be associated with microbiota compositional variation (see section 10.4) at the population level (Falony et al, 2016), were measured and taken into account in the statistical analysis. Most studies reporting effects of specific treatments on microbiota composition and diversity have used parallel designs, with relatively small sample sizes (varying from 12 – 25 subjects per treatment group). For the present pilot study, with crossover design, 30 subjects were enrolled.
8.7.3 INTERIM ANALYSIS BY DATA AND SAFETY MONITORING BOARD Not applicable.
8.8 CHANGES IN THE CONDUCT OF THE STUDY OR PLANNED ANALYSES Not applicable
Page 24 of 37
9. STUDY SUBJECTS
9.1 DISPOSITION OF SUBJECTS
Figure 1. Subject flow
interested334
passed prescreening
188
not invited 102
Invited for information
session86
Attended info session
65
Signed ICF65
Screen failures
eligible after
screening17 48
drop-outs, replaced
before V1
2
BMI too high
medication use
Health issues
Other: holiday,
no GP
Excluded by lot
Completed study
4 5 2 6 16 30
Page 25 of 37
9.2 PROTOCOL DEVIATIONS Table 2. Protocol deviations at subject level
Description of deviation Minor/major Subject-IDs IP-intake skipped, single occasions Minor 02, 03, 05, 06, 08, 12, 14, 20, 21, 28,
29, 31 IP-intake skipped repeatedly (7/140) Major 24 Antibiotic use Major 01, 30 Cooling agent thawed Minor 23, 28 Fecal sample thawed Major 24 Life style change Minor 02, 06, 31 IP returned at later visit Minor 02, 14, 26 Fecal sample delivered at later visit Minor 02 Urine in fecal sample Minor 05 Missing blood sample Major 05 (visit3, VTC 02, end of Period 1)
The only missing data point in the Data Master File is the plasma CRP concentration for subject 5, visit 3 (end of Period 1), because no blood sample could be collected from that subject. There were no further missing blood samples. All fecal samples, as well as all questionnaire data were complete. Sample quality may have been compromised for fecal samples from subjects 23 (second sample), 24 (fourth sample) and 28 (second sample), due to thawing during transportation. Other factors potentially affecting microbiota outcomes were antibiotic use and reported diarrhea. Sample quality was checked during preprocessing of the microbiota data. Microbiota QC showed clustering of samples per subject in the PCA, even when potential confounding factors were present. This means that there were no outlier samples over time within the individuals. Two subjects had used an antibiotic: Subject 01 (nitrofurantoin) and subject 30 (doxycycline), both in the wash-out period. No substantial influence of thawing of the samples (see above), or antibiotic treatment was observed. Subjects who suffered from diarrhea were not close to a fecal sampling time point. After discussion at the blind data review meeting, it was decided to include results of subjects 01 and 30 for period 1, and exclude their results for period 2 as antibiotic use was an exclusion criterion. For the main outcome parameters, additional analysis including subjects 01 and 30 for period 2 was agreed.
Page 26 of 37
10. EFFICACY EVALUATION
10.1 DATA SETS ANALYSED Full Analysis Population The Full Analysis Population includes all subjects who were randomised and received any study product. Based on the research questions it has been chosen to do a per-protocol analysis (see below). Intention-to-treat dataset In an intention-to-treat analysis all randomized samples are analysed, as this reflects best the net effect of a product prescribed to a certain population. However, for increased mechanistic insight of the product in relation to the microbiome, the non-compliant subjects should be excluded from the analysis (per-protocol dataset). In this study, up to the code break the analysis was performed on the ITT sample set (providing that samples were collected and pass sequencing QC), but after that only the Per-Protocol Set was analysed. Per Protocol Population Per protocol analysis was performed for all outcomes and for all randomized subjects having fulfilled the inclusion and exclusion criteria, who have completed the intervention and for whom results for the primary outcomes are available, excluding subjects with protocol violations or major protocol deviations (to be decided in the Blind Data Review meeting). During the Blind Data Review meeting on 08 October 2019 it was decided that the samples of the second period of subjects 01 and 30 were excluded from the PP-set due to antibiotic treatment. Since the effects of the antibiotics (nitrofurantoin, or a single dose of doxycycline) on the microbiota composition are probably limited, these samples were taken along in analysis on the ITT-set, to assess whether the result is similar to analysis on the PP-set or not. Analysis on the ITT-set will only involve assessment of treatment effect in multivariate analysis (cross-sectionally and longitudinally).
10.2 DEMOGRAPHIC AND OTHER BASELINE CHARACTERISTICS Table 3. Baseline characteristics of study population, by treatment order
Parameter Active-Placebo order n=14 Placebo-Active order n=15
Age (mean ± SD) 40.0 ± 12.0 41.9 ± 13.8
Gender (n/% male) 5/35.7% 5/33.3%
BMI (mean ± SD) 24.5 ± 3.2 24.6 ± 3.2
10.3 MEASUREMENT OF TREATMENT COMPLIANCE Overall, each subject had to use the test product on 140 moments: 2 periods x 5 weeks x 7 x 2 (twice daily). Self-reported compliance was as follows:
• 6 subjects missed 1/140 moments (0.7% non-compliance) • 3 subjects missed 2/140 moments (1.4%) • 2 subjects missed 3/140 moments (2.1%)
Page 27 of 37
• 1 subject missed 7/140 moments (5% non-compliance) It was concluded that there was no reason to exclude any subjects based on non-compliance.
10.4 EFFICACY RESULTS
10.4.1 STATISTICAL/ANALYTICAL ISSUES For 11 faecal samples with very low DNA yield, the DNA isolation and 16S amplicon PCR procedure was repeated on backup samples, before sending in all samples for sequencing. Sequencing results were received for all samples. Overall conclusion was that sequencing depth for each sample was sufficient and no samples had to be removed. For further analysis, replicate samples with the highest read count were selected. Due to a missing blood sample, one data point was missing for plasma CRP results, for the active treatment in period 1 (subject 05). Absolute CRP values were found not to be normally distributed. A 10log transformation was performed, resulting in values which were normally distributed. Four subjects (subjects 03, 09, 23, 29) had logCRP values larger than the mean +2 SD; however, there was no specific reason to exclude these values from statistical analysis. Five subjects presented with absolute CRP values >10 mg/L: 03, 09, 12, 29, 32. This cut-off value is often used to diagnose a state of infection. Underlying infection was a likely cause in subjects 12 (Pfeiffer) and 29 (common cold), whereas subject 09 reported occult blood loss after the end of study (later diagnosis showed no underlying pathology). The General Linear Model was re-run without these 5 subjects, to check whether this would have any influence on the effect of treatment.
10.4.2 PRIMARY OUTCOME: MICROBIOTA RESULTS Details of the microbiota analyses can be found in Appendix A. The ITT dataset included all samples, the PP dataset excluded 4 samples: samples of subjects 01 and 30 in period 2. The microbiota composition at the end of the Lactobacillus LB treatment was very similar to the composition after placebo treatment (Figure 2).
Page 28 of 37
Figure 2. Overview of microbiota compostion at the end of placebo (left panel) and Lactobacillus LB (right panel) treatment This visual evaluation was confirmed by RDA analysis at the genus level: there was no effect of treatment on microbiota composition at the end of the treatment periods (p=0.104). Looking at the change in microbiota composition within groups and within a treatment period, no change was observed in both placebo periods. The microbiota composition also did not change during Lactobacillus LB treatment in period 1 (p=0.71); however, during period 2 within the Lactobacillus LB treatment group, the variation in composition between start and end of this period explained was 4.8% (p=0.026). During the wash-out period of subjects that had started with placebo in period 1, and continued with Lactobacillus LB in period 2, a change in microbiota composition was observed that was borderline significant (p=0.05). RDA analysis of samples at the end of both treatment periods, but at the OTU level, resulted in a borderline significant difference between treatments (p=0.074, explained variation 0.66%). Lactobacillus LB treatment was associated with e.g. Veillonellaceae and Lactobacillaceae, while placebo treatment was associated with Lachnospiraceae (Fig. 3).
Page 29 of 37
Figure 3. RDA analysis of the effect of treatment on microbiota composition at the OTU level. End point samples of both intervention periods were included. Subject was included as covariate. Univariate analyses for specific taxons did not result in statistically significant differences between treatment, although Lactobacillus was significantly higher at the combined end points in the Lactobacillus LB product group compared to placebo (p=0.048), before correction for multiple testing was applied. When looking at the difference in change (ratios of relative abundances over time) of individual taxons, in the different study periods, a significant decrease in Enterobacteriaceae was observed during placebo treatment in period 2, while there was a slight increase in the Lactobacillus LB product group (p=0.0013; without correction for multiple testing). In the ITT group, for Lactococcus a non-significant trend (p=0.081; after FDR correction) was observed for an increase during placebo treatment and a decrease during Lactobacillus LB treatment, both only in period 2. Diversity of the microbiota appeared to increase during Lactobacillus LB treatment, in both periods, but differences compared to placebo were not statistically significant.
Page 30 of 37
10.4.3 EXPLORATORY OUTCOME: CRP RESULTS Plasma CRP values, both absolute and log-tranformed, at the start and end of both periods are presented by treatment order (Table 4). Table 4. CRP and logCRP per period and treatment order (mean ± SD)
Period 1 Period 2
week 2 week 6 week 9 week 13
Group 1 Active - Placebo
CRP 11.88 ± 34.71 5.95 ± 13.50 2.54 ± 2.20 2.18 ± 1.94
logCRP 0.28 ± 0.75 0.20 ± 0.19 0.22 ± 0.46 0.17 ± 0.41
Group 2 Placebo - Active
CRP 3.13 ± 4.01 2.59 ± 3.39 4.28 ± 8.49 3.53 ± 4.38
logCRP 0.17 ± 0.59 0.004 ± 0.67 0.15 ± 0.66 0.15 ± 0.70
Statistical testing of differences between treatments was performed on the log-transformed data. Data for period 1 and period 2 were combined, as no interaction with treatment order was observed.
Figure 4. Mean and SEM of logCRP (ng/mL) at start and end of active (A) and placebo (P) treatment periods Table 5. CRP and logCRP by treatment group (mean ± SD)
Baseline After 5 wk
consumption P-value*
CRP Active (n=29) 7.94 ± 24.71 4.70 ± 9.79
Placebo (n=30) 2.84 ± 3.19 2.38 ± 2.72
logCRP Active (n=29) 0.21 ± 0.70 0.17 ± 0.69 0.614
Placebo (n=30) 0.19 ± 0.52 0.09 ± 0.55 0.045
* Change within treatment period was tested with paired-sample t-test
Page 31 of 37
The change in plasma CRP within treatment period (Table 5, Figure 4) was statistically significant (p=0.045) for placebo treatment, but not for the active treatment. The change in the active treatment period compared to the placebo treatment period was tested in a General Linear Model including age and gender as potential confounders (Table 6). The overall effect of treatment on plasma CRP was not statistically significant (p=0.432). Table 6. General Linear Model including potential determinants of variation in logCRP
SS Degr. Of Freedom
MS F p
Intercept 0,11095 1 0,11095 1,0710 0,305506 Age 0,22933 1 0,22933 2,2138 0,142824 logCRP_start 11,58271 1 11,58271 111,8095 0,000000 Treatment order 0,00740 1 0,00740 0,0715 0,790251 Treatment 0,06505 1 0,06505 0,6280 0,431703 Gender 0,23362 1 0,23362 2,2552 0,139219 Treatment order * treatment 0,13959 1 0,13959 1,3475 0,251017 Error 5,38685 52 0,10359
No interaction with treatment order was observed (p=0.251 for treatment x treatment order), which allows the combining of results from period 1 and period 2. The only significant predictor of CRP value after 5 weeks consumption in this model was the baseline CRP value (p<0.0001). Five subjects had absolute CRP values >10 ng/mL, at one of the time points. Running the General Linear Model without these five subjects did not substantially change any of the findings above.
10.4.4 TABULATION OF INDIVIDUAL RESPONSE DATA See Data Master File.
10.4.5 EFFICACY CONCLUSIONS No consistent significant effects of Lactobacillus LB product on the gut microbiota composition of the healthy volunteers were detected. 5-week consumption of Lactobacillus LB product had no effect on plasma CRP concentrations, compared to placebo.
Page 32 of 37
11. SAFETY EVALUATION
11.1 ADVERSE EVENTS
11.1.1 BRIEF SUMMARY OF ADVERSE EVENTS AEs were registered on a daily basis during the total study period. A complete list of AEs is provided in Table 7. Table 7. Type and number of adverse events reported during active or placebo treatment, of in the wash-out period
Type of AE Placebo treatment
Active treatment
Wash-out Total Comment
Bladder infection/complaints
1 1 2
Cold sores 1 1 Bursitis 1 1 Pfeiffer’s disease 3 3 1 subject Sun allergy 1 1 Eczema 1 1 Athlete’s foot 1 1 Tick bite 1 1 Common cold 1 1 Sore throat 1 1 Muscle soreness 2 2 Back pain 2 2 Bruised ribs 1 1 Depression 1 1 Migraine 1 1 Headache 6 4 2 12 5 subjects Diarrhea 2 3 5 Stomach ache 2 1 3 Heartburn 1 1 Overall 16 20 5 41
The overall number of AEs is fairly evenly distributed over treatment periods. There is no apparent single or cluster of adverse events that occur more often during the active compared to the placebo period. Most (63%) AEs were considered mild. Severe AEs were observed only in the active treatment, not in the placebo treatment (Table 8), but none of the AEs in the active treatment were scored as possibly or definitely related to IP-intake (Table 9). Table 8. Severity of adverse events, by treatment period
Severity Placebo Active Wash-out All Comment Mild 11 11 4 26 Moderate 5 4 9 Severe 4 1 5 Missing 1 1 Bruised ribs
Page 33 of 37
Table 9. Reported relation of adverse events to IP-intake, by treatment period Relation with IP-intake
Placebo Active Wash-out All
No 15 18 5 38 Unlikely 2 2 Possible 1 1 Yes 0
11.1.2 ANALYSIS OF ADVERSE EVENTS Not applicable
11.1.3 LISTING OF ADVERSE EVENTS BY PATIENT See Data Master File
11.2 DEATHS, OTHER SERIOUS ADVERSE EVENTS AND OTHER SIGNIFICANT ADVERSE EVENTS
Not applicable
11.3 CLINICAL LABORATORY EVALUATION Not applicable
11.4 VITAL SIGNS, PHYSICAL FINDINGS, AND OTHER OBSERVATIONS RELATED TO SAFETY Not applicable
11.5 SAFETY SUMMARY AND CONCLUSIONS Overall, AEs were mild in majority and considered not to be related to the active treatment. The consumption of Lactobacillus LB product on a daily basis during 5 weeks in this healthy adult population was safe.
12. DISCUSSION AND OVERALL CONCLUSIONS In summary, no consistent significant effects were observed of a daily intake of 680 mg of Lactobacillus LB product during 5 weeks on the gut microbiota composition or blood CRP concentrations of healthy volunteers, compared to intake of a placebo product (sucrose). The study was designed as a crossover study, with a wash-out period of 2 weeks between the treatments. Because the variation in gut microbiota composition between individuals is large, a crossover design has the advantage of being able to compare changes within individuals. This reduces the overall variation, and consequently the number of subjects in the study population required to be able to detect relevant differences. A wash-out of 2 weeks was expected to be sufficient, and the statistical analysis confirmed there was no interaction with treatment order. Some minor but significant shifts in microbiota composition were found in one period only. Because of the lack of interaction with treatment order, both periods should be considered combined, and an effect in just one period cannot be explained biologically, except as normal biological variation within individuals. Effects that appeared to occur in the washout or placebo treatment period should also be interpreted as part of normal biological variation.
Page 34 of 37
A number of 30 subjects is very common for an exploratory study like this, aiming to look at shifts in microbiota composition and diversity. However, no actual power calculation could be performed in advance. In the multivariate analysis at the OTU level a trend (p = 0.074) was found for a treatment effect at the combined end points. Among taxa associated with the Lactobacillus LB treatment were e.g. Lactobacillaceae (generally considered beneficial), but also potentially pathogenic Enterobacteriaceae. In a larger study population a statistical significant difference may be found, but it should also be noted that further research would probably be required to study the biological relevance of the effect. In this study, the placebo product was sucrose, whereas lactose was used as carrier in the Lactobacillus LB product. The optimal choice for the placebo product would have been lactose, but technical restraints were leading in the choice for sucrose. Sucrose is completely digestible in the small intestine, and therefore no effects on the gut microbiome are expected to occur during placebo treatment. The study population consisted of healthy adult subjects. The gut microbiota composition of healthy subjects is relatively stable, although changes can be induced on the short term. The fact that no significant changes in microbiota composition and diversity were observed in this population may be explained by the fact that they were indeed healthy. This finding does not exclude the potential of Lactobacillus LB to be effective in restoring a healthy microbial community in subjects with gut dysbiosis, e.g. after a microbial infection.
Page 35 of 37
13. REFERENCES AIBMR Life Sciences, Inc. The Generally Recognized as Safe (GRAS) Status of Lactéol® LB strain. 2016 Berni Canani R, De Filippis F, Nocerino R, Laiola M, Paparo L, Calignano S, De Caro C, Coretti L, Chiariotti L, Gilbert JA, Ercolini D. 2017. Specific signatures of the gut microbiota and increased levels of butyrate in children treated with fermented cow's milk containing heat-killed Lactobacillus paracasei CBA L74. Appl Environ Microbiol 83:e01206-17. Ducrotte P. Lactéol® Technical and Scientific Synthesis. Nancy, France: ALN Editions; June 2009. Canducci F, Armuzzi A, Cremonini F, et al. A lyophilized and inactivated culture of Lactobacillus acidophilus increases Helicobacter pylori eradication rates. Aliment Pharmacol Ther. 2000;14(12):1625-1629. Chauvière G, Coconnier M-H, Kerneis S, Darfeuille-Michaud A, Joly B, Servin AL. Competitive exclusion of diarrheagenic Escherichia coli (ETEC) from human enterocyte-like caco-2 cells by heat-killed Lactobacillus. FEMS Microbiol Lett.1992;70(3):213-217. Coconnier M-H, Bernet MF, Chauvière G, Servin AL. Adhering heat-killed human Lactobacillus acidophilus, strain LB, inhibits the process of pathogenicity of diarrhoeagenic bacteria in cultured human intestinal cells. J Diarrhoeal Dis Res.1993;11(4):235-242. Coconnier M-H, Liévin V, Bernet-Camard MF, Hudault S, Servin AL. Antibacterial effect of the adhering human Lactobacillus acidophilus strain LB. Antimicrob Agents Chemother. 1997;41(5):1046-1052. Faith JJ, Guruge JL, Charbonneau M, Subramanian S, Seedorf H, Goodman AL et al. The long-term stability of the human gut microbiota. Science 2013; 341: 1237439. Falony G, Joossens M, Vieira-Silva S, et al. Population-level analysis of gut microbiome variation. Science. 2016 Apr 29;352(6285):560-4. doi: 10.1126/science.aad3503. Lahtinen SJ1, Forssten S, Aakko J, Granlund L, Rautonen N, Salminen S, Viitanen M, Ouwehand AC. Probiotic cheese containing Lactobacillus rhamnosus HN001 and Lactobacillus acidophilus NCFM® modifies subpopulations of fecal lactobacilli and Clostridium difficile in the elderly. Age (Dordr). 2012 Feb;34(1):133-43. doi: 10.1007/s11357-011-9208-6. Liévin-Le Moal, V. (2016) A gastrointestinal anti-infectious biotherapeutic agent: the heat treated Lactobacillus LB. Ther Adv Gastroenterol 2016, Vol. 9(1) 57–75 Liévin-Le Moal, V., Sarrazin-Davila, L. and Servin, A. (2007) An experimental study and a randomized, double-blind, placebo-controlled clinical trial to evaluate the antisecretory activity of Lactobacillus acidophilus strain LB against nonrotavirus diarrhea. Pediatrics 120: e795–e803. Moore RJ, Stanley D. Experimental design considerations in microbiota/inflammation studies. Clinical & Translational Immunology (2016) 5, e92; doi:10.1038/cti.2016.41 Salazar-Lindo, E., Figueroa-Quintanilla, D., Caciano, M., Reto-Valiente, V., Chauviere, G. and Colin, P. (2007) Effectiveness and safety of Lactobacillus LB in the treatment of mild acute diarrhea in children. J Pediatr Gastroenterol Nutr 44: 571–576. Simakachorn N, Pichaipat V, Rithipompaisam P, et al. Clinical evaluation of the addition of lyophilized, heat-killed Lactobacillus acidophilus LB to oral rehydration therapy in the treatment of acute diarrhea in children. J Pediatr Gastroenterol Nutr 2000;30: 68–72.
Page 36 of 37
14. APPENDICES
14.1 MICROBIOTA RESULTS To be added, PowerPoint slides as presented
14.2 DECLARATION AND SIGNATURE PAGE Authentication by the principal investigator I, the undersigned, hereby declare that to the best of my knowledge this report constitutes a true and complete representation of the procedures followed and of the results obtained in this study by NIZO, Ede, The Netherlands. The study was carried out under my overall supervision and conducted in accordance with the current assembly of the Declaration of Helsinki (Fortaleza, Brazil, October 2013; The requirements as described in the EU Clinical Trials Directive 2001/20/EC transposed in the Revision of the Dutch Medical Research involving Human Subjects Act (“Wet medisch-wetenschappelijk onderzoek met mensen”, WMO, effective as of March 1, 2006) and the current national regulations like the Civil Code, Act on safe guarding personal details and Act on archiving.
Alwine Kardinaal, PhD
Principal Scientist
Signature
Date
Page-37-of-37
