Team:TU-Eindhoven/Implementation

Implementation

“Knowledge is not power, the implementation of knowledge is power”

On this page you will find the proposed implementation of IBDetection. In chronological order, we want to firstly address the mechanism and used GMO, followed by the pill design that should deliver the bacteria. Then, we explore the implementation of the pill, going over all facets of application, including end-users such as patients and doctors, but also industry, safety, legislation, and more. Finally, we close our proposed implementation with a cost analysis.

As described in more detail on the Project Description page, IBDetection consists of a pill that can be ingested orally by the patient. It delivers live, genetically engineered bacteria to the patient’s intestine. There, the bacteria are released from the pill and migrate through the colon. At the site of inflammation, the bacteria will detect tetrathionate—a biomarker for inflammation—with high specificity, using the TtrS/R sensing system. This detection will result in the production of ARG gas vesicles in the cytoplasm of the bacteria. Results from the lab, as well as simulations from our model showed us that this proof-of-concept works, in particular the protein mechanism, and has the potential for application. Thus, we intend to build on this progress, and lay the groundworks for the future implementation of our mechanism.

The Expanded Mechanism

On our Proof of Concept and Model pages, the above described mechanism is implemented in vitro and in silico. As described on the Results page, the final experiments for the characterisation and reproducibility of our sensor still have to be performed. Nevertheless, based on the current results from the lab and the model, and the envisioned functioning of the system, a series of expansions are described below for future implementation of IBDetection.

The first expansion will consist of exchanging the current promoters in front of TtrS and TtrR for constitutive promoters, such as J23100 and J23109 respectively [1], for continuous expression of our these proteins. In particular, special attention must be paid to the constitutive promoter for TtrR expression, as overexpression of the TtrR protein results in nonspecific signal output as described on the Engineering Success segment Doxycycline. The exchange to constitutive promoters will mean our bacteria will be able to effectively operate in the intestine, where artificial induction is not feasible.

As described in Human Practices segment The Science, tetrathionate is not the only gut sulfate inflammation marker. The second expansion is, therefore, to incorporate an orthogonal thiosulfate sensor. Taking the considerable challenge of a three-plasmid co-transformation in mind, we instead envision a co-culture where one engineered species will contain the sensor for tetrathionate, and the other for thiosulfate. Specifically, the tetrathionate sensor will include our current system, i.e. TtrS, TtrR and ARG1, whereas the thiosulfate sensor will include ThsS, ThsR and ARG2. The two-component ThsS/R system functions in a similar fashion to the current sensing part as a transmembrane-regulator system, but for the detection of thiosulfate instead [1]. ARG2 functions similarly to ARG1, but gas vesicles produced by this gene can be collapsed with a different frequency pulse [2]. This means both types of gas vesicles, and thus both tetrathionate and thiosulfate concentrations, can be quantified during the same multiplexed ultrasound examination.

Lastly, to complete the system, and to ensure environmental safety, a kill switch will be incorporated in the plasmid design of both species. As explained in Safety, various types of kill switches are available, and the kill switch based on oxygen levels is the most suitable candidate for our user-case, as the human gut is sparse and the environment outside the body is rich in oxygen [3].

The GMO

We performed our lab work with E. coli BL21 (DE3), primarily because of its robust use in protein expression and availability in the lab. However, after consulting Bruno Pot, who works as microbiologist for the company Yakult Europe in the Netherlands, it became clear that this cell was not suitable for future application. Some E. coli bacteria have pro-inflammatory characteristics and are, therefore, associated with triggering IBD flare-ups [4]. Even though used for decenia in intestinal research, studies also revealed that the administration of E. coli Nissle 1917 bacteria are related to an increase in the risk for colon cancer [2,5]. Therefore, it was recommended to replace E. coli BL21 (DE3) with a more suitable bacterial strain, in particular the lactic acid bacteria L. plantarum WCFS1 strain [6]. These cells are already used for genetic research, have probiotic properties, and are present in the human gut microbiome [7]. When our system is incorporated into L. plantarum WCFS1, all safety aspects have to be considered thoroughly. The effect of the bacteria on the subject must be verified on multiple levels, including the microbiome and disease progression. The gut-on-a-chip—which is an in vitro platform for studying human-gut physiology, pathology, and pharmacology—can be used to study this effect first, whereafter preclinical and clinical trials must be conducted with an emphasis on quantification of disease severity and localization [8].

The Pill

After consulting gastroenterologists, the company Bioconnection, and our partner iGEM team BOKU Vienna, we reached a final pill design (Figure 1). This pill design contains a pH-sensitive and time-dependent coating that releases the bacteria into the terminal ileum and colon, where inflammation is most prevalent [9]. As releasing L. plantarum WCFS1 in the intestine comes with many safety and ethical concerns, an alternative pill design was developed. In this design, the bacteria are anchored inside the pill while ensuring sufficient residence time inside the intestine [10,11]. Detailed information on this alternative pill design can be found on our Human Practices segment Ethics. This design, however, comes with several challenges. Primarily, tetrathionate must diffuse into the pill while the gas vesicles produced by the bacteria inside the pill must still be detectable and quantifiable by ultrasound. Secondly, the minimal quantity of required cells per pill necessary for a sufficiency detectable signal when exposed to pathological concentrations of tetrathionate. Moreover, the maximally possible number of cells that can be included in the pill. Finally, the production process of the pill comes with additional challenges, e.g. how to keep the GMOs alive inside the pill. More on this can be found on our Human Practices segment The Pill.

Figure 1: Conceptional final pill design

The User

The primary end-users of IBDetection are IBD patients and gastroenterologists. From the patient interviews and general survey, made together with iGEM team BOKU Vienna, we learned that an ultrasound measurement was highly preferred over an endoscopy for monitoring IBD. Patients based this opinion mainly on the fact that an ultrasound procedure is non-invasive, and no fasting or taking laxatives prior to examination is required [12]. The ultrasound measurement must, however, be performed preferably in the morning to reduce the amount of intraluminal air and peristaltic movements. Moreover, gradually applied pressure using the ultrasound probe reduces intraluminal air [13]. This reduction of intraluminal air is preferred to minimize false-positive test results. For IBDetection, the ultrasound probe creates a specific frequency pulse to either image or collapse the produced gas vesicles. By image-analysis, the quantification of the gas vesicles can be performed by calculating the difference in contrast of the image before and after collapse. Consequently, laxation and fasting are not required for the effective application of IBDetection, as only the bacterially produced gas vesicles are affected by the specific frequency pulse, and all other signals are filtered.

Furthermore, there is a high GMO acceptance among questioned IBD patients under the proviso that it is safe and a clear explanation is given prior to usage. For gastroenterologists, however, some additional challenges were identified. Firstly, the imaging intervention in patients with obesity is challenging due to the limited penetration depth of the ultrasound equipment [14]. Therefore, additional attention must be paid during clinical trials for this specific patient base, especially considering the increasing prevalence of obesity in developed countries [15]. Second, some parts of the intestine, including the duodenum and proximal jejunum, cannot be visualized by the use of ultrasound [16,17]. However, this issue is minimal for IBDetection, as the target area is set on the terminal ileum and colon, the region where inflammation is predominantly present [9]. Lastly, experienced sonographers are required for proper ultrasound imaging. Unfortunately, trained operators are currently in high demand, which means the introduction of IBDetection would further contribute to an increase in sonographer demand [18].

The Real World

For the implementation of IBDetection into the real world; entrepreneurial, ethical, safety, and legislative aspects were considered.

As for entrepreneurship, we received clear advice from Paul Vernooij, Investment Analyst at BOM, to make IBDetection a financial success. We were advised to reduce risks as much as possible, particularly in the preclinical and clinical phases. As mentioned before, we aim to first thoroughly validate and characterize the pharmacokinetics, efficacy, and safety of our expanded system, GMO, and pill in vitro. This way, risk of failure is minimized when the preclinical phase happens, and it showcases the potential of IBDetection to future investors. Then, we plan to apply for a patent of our expanded mechanism, giving us the ability to commercialize IBDetection. These points would make IBDetection attractive to potential investors by reducing capital risks.

With Johannes Rath, ethics, biosafety, and biosecurity advisor, we went over many facets of ethics. Some we already integrated into our patient interviews. Johannes pointed out safety and ethics are closely related. Additionally, he mentioned the challenge to assess various parameters, such as patient comfort. Patient comfort cannot be accessed through laboratory results, and therefore, other methods must be used. We took this very seriously when pursuing our goals. As per Johannes’ advice, we put further emphasis on the safety of IBDetection, and in particular also considered the safety of implementation, such as production and distribution. To address these questions, we approached the company Bioconnection. It was brought up that for the production process of IBDetection, it is of high importance to take the industry-accepted Good Manufacturing Practices into account, which has to be met if we want our product to enter the market. For distribution, we plan to follow their advice and freeze-dry the product into a dormant state for easy, off-the-shelf distribution of IBDetection, instead of either a short-lived product that has to be distributed by order or same-day production and delivery for patients in need. This means production and distribution of IBDetection can be done cheaper, and on a larger scale.

In addition, we also had to consider GMO legislation, which could be the greatest challenge of all. As discussed with EuropaBio Industrial Biotech Officer Lucie McMurtry, the Eurobar survey showed that there is a high GMO acceptance among Europeans while most concerns were about antibiotics, hormones and steroids residues. Thus, we aim to filter out antibiotics after large scale culturing, and avoid the use of hormones and steroids throughout the process. Furthermore, we intend to inform the end-users in detail about the exact contents of IBDetection. Lucie also pointed out that the current European GMO legislation does not allow the application of live GMOs outside of specially regulated labs, limiting the implementation of IBDetection in the current European climate. Acknowledging this fact, but setting it aside, a cost analysis was performed to determine the cost-effectiveness and financial perspective of the implementation of IBDetection.

The Cost Analysis

A cost analysis was performed to estimate and inspect the cost-effectiveness of IBDetection. This analysis covers factors such as estimated material costs, wages, time consumption, and specificity of both the current golden standard methods and that of IBDetection. In concreto, the average costs of the current golden standard monitoring tract, i.e. from the fecal calprotectin test to the endoscopy, was compared to the monitoring tract including IBDetection as an intermediate procedure after a positive calprotectin test result. For the current monitoring methodology, the IBDoc Calprotectin Home Test was chosen as the golden standard for the fecal calprotectin test, with a true specificity of 74% [19–21]. Average endoscopy and ultrasound procedure costs and durations were based on statistics from literature, which were 800 dollars & 30 minutes, and 80 dollars & 20 minutes respectively [22–25]. Wages were determined similarly, and based on wages of gastroenterologists, medical sonographers, and nurses in the United States in 2019 [26–28]. A cost estimation of our pill was done in consultation with experts from the industry, resulting in a price of 2 dollars per pill for industry-scale production.

Two major assumptions were made for the cost analysis, the first being that all performed tests, i.e. the calprotectin test, ultrasound procedure, and endoscopy, all have equal and maximal sensitivity and are not factored into the calculation. Literature states a 97-100% sensitivity for the calprotectin test, pointing to this being a realistic assumption [29]. The second assumption concerns the specificity of IBDetection. Literature states a high specificity of the two-component system, however, no exact percentage was stated [1]. Thus, we assume the definition of high specificity to be between 80 and 99%, and do the analysis for this entire range. Because the true prevalence of IBD-positive (IBD+) patients during flare-ups that go through the entire tract were not documented well in literature, we decided to do the analysis for the entire relevant range of IBD+ patients, i.e. from 50 to 100% IBD+. As can be inferred from the results (Figure 2), everything under 50% IBD+, i.e. where the majority of patients going through the tract, in fact, do not have IBD, results in IBDetection being the significantly cheaper method. However, considering the likelihood of a patient being IBD+ when suffering from symptoms, we found it relevant to only assess the cost analysis for cases where the majority of patients going through the tract do in fact suffer from IBD. Below, an example calculation can be seen, with the specificity of IBDetection set to 99%, and the IBD+ prevalence set to 80%.

Example Calculation

Table 1: Example calculation of the cost analysis with IBDetection specificity set to 99%, and IBD+ to 80%.

As can be seen in Table 1, a calculation is done for different values of IBDetection specificity and IBD+ prevalence, here an example with 99% and 80% respectively, resulting in a total cost analysis that is two-dimensional (Figure 2). The analysis goes as follows, two calculations are done for the golden standard tract and two for the tract including IBDetection. Per tract one calculation is done for an IBD+ patient, and one for an IBD- patient.

In the case of IBD+, the cost calculation amounts to the total sum of all procedures for the respective tracts. For the golden standard, this includes the calprotectin test and the endoscopy, including wages. For the IBDetection tract, this includes the calprotectin test, the ultrasound examination and the endoscopy, including wages and pill cost. In this example, this results in the total costs of 939 and 1032 dollars respectively.

In the case of IBD-, the patient goes through the tracts sequentially, starting with a guaranteed cost for the calprotectin tests in both tracts, and a subsequent chance of a follow-up examination that is proportional to the specificity of the calprotectin test in the golden standard tract. In the IBDetection tract, the first follow-up examination is replaced with IBDetection, and patients have a subsequent chance of a follow-up examination that is proportional to the specificity of IBDetection. In this example, 266 and 57 dollars respectively.

The average cost of the golden standard is then calculated by summing the product of the respective costs for IBD+ and IBD- with their respective prevalence, in this example the sum of 80% chance of 939 dollars and 20% chance of 266 dollars. The same is done for the IBDetection tract, in this example the sum of 80% chance of 1032 dollars and 20% chance of 57 dollars. This amounts to an average cost per tract per patient of 804 dollars for the golden standard tract, and 837 dollars for the tract that includes IBDetection. These average costs are then subtracted, resulting in a final cost difference in dollars. This difference is then converted to a percentage increase over the current golden standard for the final statistic. Hence, a positive percentage reflects an increase in costs, and a negative percentage a decrease (in comparison to the costs of the golden standard tract).

Thus, the cost analysis primarily depends on the specificities of the calprotectin test and IBDetection, and the prevalence of IBD- patients, as this is where IBDetection intervention happens and yields not only financial benefit, but also a gain in Quality of Life (QoL), a reduction in procedure time and a reduction in required staff. For each of these cases, two nurses are no longer required, the patient does not have to use strong laxatives or fasten prior to the examination, and the vast majority avoids an unnecessary invasive follow-up endoscopy.

In Figure 2, the total cost analysis is plotted as a two-dimensional heatmap, where the IBD+ prevalence is varied from 50 to 100%, and the IBDetection specificity from 80 to 99%. Each box represents the percentage difference in costs when comparing the golden standard tract to that including IBDetection. Thus, negative percentages mean the IBDetection intervention results in cost savings, and positive percentages extra costs.

Figure 2: cost analysis of IBDetection, expressed in average extra costs in percentages, ranging IBD positive (IBD+) patient prevalence from 50 to 100%, and IBDetection specificity from 80 to 99%.

From Figure 2 we can observe the effect of IBD+ prevalence and IBDetection specificity on the final change in costs, and it appears IBD+ prevalence is the major factor in our cost analysis. IBDetection will be most expensive when all patients going through the tract are IBD+, resulting in an average cost increase of 9.1% compared to the golden standard. Here, the intervention of IBDetection would not be beneficial as all patients test positive and have to undergo both the ultrasound and the endoscopy, which is a highly unlikely scenario. IBDetection implementation is estimated to yield an average decrease in costs when IBD+ prevalence falls below 65%. This means that on average, for each case below this threshold our method will be less expensive than the current golden standard, on top of the gain in QoL and other benefits as mentioned before.

Taking a more specific and realistic estimation, with IBD+ prevalence between 70 and 90%, and an IBDetection specificity between 90 and 95%, we would have a worst-case scenario where the implementation of IBDetection results in an average increase in costs of 7%, and a best-case scenario of an increase in costs of less than 1%. Thus, we estimate that it is very likely IBDetection will be marginally more expensive than the current golden standard. This, however, does not factor in the gain in QoL, the freeing of nurses that are in high demand, and the reduction in procedure time (both the examination and the preparation prior to examination).

We’d like to present the analogy of a croissant (a suitable analogy given the planned location of Paris for this year’s Giant Jamboree). You either pay a dollar for a plain croissant, or you pay a dollar 7 cents and receive a chocolate croissant that is delivered faster to you, without the hassle of having to order it, and demands less labor to prepare. With this analogy in mind to support our train of thought, based the cost analysis above, we conclude IBDetection is a cost-effective method for ruling out these false-positive test results, yielding a significant gain in QoL and time-efficiency, while reducing demand on staff.

References

1. Daeffler KN-M, Galley JD, Sheth RU, Ortiz-Velez LC, Bibb CO, Shroyer NF, et al. Engineering bacterial thiosulfate and tetrathionate sensors for detecting gut inflammation. Mol Syst Biol. 2017 Apr 3;13(4):923.

2. Bourdeau RW, Lee-Gosselin A, Lakshmanan A, Farhadi A, Kumar SR, Nety SP, et al. Acoustic reporter genes for noninvasive imaging of microorganisms in mammalian hosts. Nature. 2018 Jan 3;553(7686):86–90.

3. Albrecht JA. Oxygen Control of Respiratory Gene Expression in Escherichia Coli. 1996. 430 p.

4. Palmela C, Chevarin C, Xu Z, Torres J, Sevrin G, Hirten R, et al. Adherent-invasive Escherichia coli in inflammatory bowel disease. Gut [Internet]. 2018 Mar [cited 2021 Oct 17];67(3). Available from: https://pubmed.ncbi.nlm.nih.gov/29141957/

5. Wassenaar TM. E. coli and colorectal cancer: a complex relationship that deserves a critical mindset. Crit Rev Microbiol. 2018 Sep;44(5):619–32.

6. Bermudez-Brito M, Borghuis T, Daniel C, Pot B, de Haan BJ, Faas MM, et al. L. plantarum WCFS1 enhances Treg frequencies by activating DCs even in absence of sampling of bacteria in the Peyer Patches. Sci Rep. 2018 Jan 29;8(1):1785.

7. Mercenier A, Pavan S, Pot B. Probiotics as Biotherapeutic Agents: Present Knowledge and Future Prospects. Curr Pharm Des. 2003 Feb 1;9(2):175–91.

8. Office of the Commissioner. The drug development process [Internet]. 2020 [cited 2021 Oct 18]. Available from: https://www.fda.gov/patients/learn-about-drug-and-device-approvals/drug-development-process

9. Advances in oral nano-delivery systems for colon targeted drug delivery in inflammatory bowel disease: Selective targeting to diseased versus healthy tissue. Nanomedicine. 2015 Jul 1;11(5):1117–32.

10. Thiolated pectins: In vitro and ex vivo evaluation of three generations of thiomers. Acta Biomater [Internet]. 2021 Aug 18 [cited 2021 Oct 17]; Available from: http://dx.doi.org/10.1016/j.actbio.2021.08.016

11. Gericke M, Liebert T, Heinze T. Interaction of ionic liquids with polysaccharides, 8 - synthesis of cellulose sulfates suitable for polyelectrolyte complex formation. Macromol Biosci [Internet]. 2009 Apr 8 [cited 2021 Oct 17];9(4). Available from: https://pubmed.ncbi.nlm.nih.gov/19116894/

12. Pinto PN, Chojniak R, Cohen MP, Yu LS, Queiroz-Andrade M, Bitencourt AG. Comparison of three types of preparations for abdominal sonography. J Clin Ultrasound [Internet]. 2011 May [cited 2021 Oct 17];39(4). Available from: https://pubmed.ncbi.nlm.nih.gov/21337584/

13. Magdalena Andrzejewska MG. The role of intestinal ultrasound in diagnostics of bowel diseases. Przegla̜d Gastroenterologiczny. 2018;13(1):1.

14. Uppot RN. Technical challenges of imaging & image-guided interventions in obese patients. Br J Radiol. 2018 Sep;91(1089):20170931.

15. Trishnee Bhurosy RJ. Overweight and Obesity Epidemic in Developing Countries: A Problem with Diet, Physical Activity, or Socioeconomic Status? The Scientific World Journal [Internet]. 2014 [cited 2021 Oct 17];2014. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4212551/

16. Kucharzik T, Maaser C. Intestinal ultrasound and management of small bowel Crohn’s disease. Therap Adv Gastroenterol. 2018 May 1;11:1756284818771367.

17. Smereczyński A, Ultrasound Self-Study Club, Department of Genetics and Pathology, Pomeranian Medical University, Szczecin, Poland, et al. Pitfalls in ultrasound imaging of the stomach and the intestines [Internet]. Vol. 18, Journal of Ultrasonography. 2018. p. 207–11. Available from: http://dx.doi.org/10.15557/jou.2018.0031

18. Ultrasound brings many advantages, but trained sonographers are too few [Internet]. 2008 [cited 2021 Oct 17]. Available from: https://healthcare-in-europe.com/en/news/ultrasound-brings-many-advantages-but-trained-sonographers-are-too-few.html

19. About IBDoc® [Internet]. [cited 2021 Oct 17]. Available from: https://www.ibdoc.net/about-ibdoc/

20. Berinstein JA, Steiner CA, Bousvaros A, Tiongco FP, Greenberg E, Hardi R, et al. The Clinical Accuracy of the BÜHLMANN fCAL ELISA in the Differentiation of Inflammatory Bowel Disease From Irritable Bowel Syndrome: A Multicenter Prospective Case–Control Study. Crohn’s Colitis 360 [Internet]. 2019 Oct 5 [cited 2021 Oct 17];1(3). Available from: https://academic.oup.com/crohnscolitis360/article/1/3/otz037/5581795

21. Kennedy NA, Clark A, Walkden A, Chang JCW, Fascí-Spurio F, Muscat M, et al. Clinical utility and diagnostic accuracy of faecal calprotectin for IBD at first presentation to gastroenterology services in adults aged 16-50 years. J Crohns Colitis. 2015 Jan;9(1):41–9.

22. Moore LE. The advantages and disadvantages of endoscopy. Clin Tech Small Anim Pract. 2003 Nov;18(4):250–3.

23. Abdominal ultrasound [Internet]. [cited 2021 Oct 17]. Available from: https://www.mayoclinic.org/tests-procedures/abdominal-ultrasound/about/pac-20392738

24. Radiological Society of North America (RSNA), American College of Radiology (ACR). Ultrasound - Abdomen [Internet]. [cited 2021 Oct 17]. Available from: https://www.radiologyinfo.org/en/info/abdominus

25. Wat kost een behandeling in het ziekenhuis [Internet]. [cited 2021 Oct 17]. Available from: https://www.zilverenkruis.nl/consumenten/zorg-regelen/wat-kost-een-behandeling-in-het-ziekenhuis

26. Diagnostic Medical Sonographer Salary [Internet]. [cited 2021 Oct 17]. Available from: https://money.usnews.com/careers/best-jobs/diagnostic-medical-sonographer/salary

27. Nurse salary [Internet]. 2020 [cited 2021 Oct 17]. Available from: https://nursinglicensemap.com/resources/nurse-salary/

28. Site built by: Salary.com. Physician - Gastroenterology Salary [Internet]. [cited 2021 Oct 17]. Available from: https://www.salary.com/research/salary/benchmark/physician-gastroenterology-salary

29. Ricciuto A, Griffiths AM. Clinical value of fecal calprotectin. Crit Rev Clin Lab Sci. 2019 Aug;56(5):307–20.

IBDetection

Non-Invasive Monitoring of IBD using Genetically Engineered Bacteria as Living Sensors.

Project

Description

Lab

Model

Human Practices

About Us

Team

Partnership

Attributions

Sponsors

Contact

PO-box 513, 5600MB Eindhoven, Netherlands

igem@tue.nl

© 2021 Copyright: iGEM TU Eindhoven