Patient Advocacy Leads to Synthetic Antibodies for Mycobacteria

A small biotech company is applying a novel computational approach to design antibodies to tackle antibiotic-resistant bacterial infections

By Monica Berrondo and Varun Thvar

Rarely do small biotech companies pursue projects instigated by patients themselves. But patient advocacy groups in the rare disease space aim to change that. These groups are led by patients who are anxious to cure their own disease or their friends and family who are eager to help others handle the struggles of a new or rare disease. In late 2018, I was connected serendipitously to one such family.

I met junior high school student, Varun Thvar, in the Fall of 2018 while raising money in the San Francisco Bay Area for my small company, Macromoltek. I started the company several years earlier with the dream of personalized medicine and a lofty goal to cure any disease at the push of a button, relying on the power of computational biology.

Varun’s mother had developed a very rare infection caused by the bacteria, Mycobacterium abscessus (MABC). His mother’s infection had sparked Varun to learn more about the drug discovery process in general, and drug discovery for the MABC infection in particular. The family had set up a small foundation to find a cure for the disease and learned about our accelerated process of drug discovery. Varun had been inspired by the bacille Calmette-Guérin (BCG) vaccine for tuberculosis and the use of antibodies instead of antibiotics in the treatment of infectious disease, but learned that MABC did not yet have any antibody-based treatments available. When Varun contacted me, I was intrigued by the problem as well as the use of the project as a case study for Macromoltek’s technology. As a result, it became clear that designing an antibody for MABC could validate our capabilities while also tackling other important pathogens.

Varun’s mother had to endure an exhausting and excruciating process for both diagnosis and treatment. The initial presentation was a sudden onset of a 104-degree Fahrenheit fever. As the family learned, diagnosis of the MABC infection was a circuitous process that required ruling out various other causes before a bronchoscopy eventually led to a MABC diagnosis. Her treatment consisted of a harsh combination of antibiotics.

After an initial positive response to the treatment, the MABC infection developed immunity to the antibiotic combination within just 4 weeks. A second bronchoscopy was performed to ascertain which remaining antibiotics could be effective against the MABC strain. Unfortunately, the results showed that very few treatment options remained — the MABC infection had become highly drug-resistant. Consequently, the treatment cocktail method of delivery was updated and all antibiotics were delivered intravenously through a peripherally inserted central catheter (PICC) line in a twice-daily regimen of 2-3 hours each time. This course of treatment continued for the next 9 months, after which a lung resection was used to remove the affected area. Fortunately, after a 12-month ordeal, Varun’s mother recovered and became the inspiration for Varun to reduce the suffering of others through drug discovery.

MABC Defined

MABC is a rapidly-growing bacteria that is responsible for a number of severe pulmonary infections, especially in patients with chronic lung disease. MABC infections are characterized by a resistance to antibiotic treatments and comorbidities with other respiratory illnesses. As a result, these infections are a serious concern, similar to what we have seen with other respiratory infections, such as COVID-19.

An analysis of 107 patients between 2001-2008  identified that patients with MABC infections were predominantly female, non-smokers, and with an average age of 60.1 MABC infections are found in both patients with and without existing respiratory conditions, such as lung airway abnormalities. Although environmental reservoirs of MABC have not been confirmed, the bacteria have been found in biofilms in many sources, including shower heads. Insufficient evidence has been collected to suggest changes in patient behavior could aid in preventing MABC infections.

The increasing profile of MABC infections is due to the infection’s uniquely strong antibiotic resistance. While some patients receive surgical treatment, the MABC infection is too extensive for effective surgical resection for most patients. For other patients, multi-drug cocktails, including clarithromycin, are a standard treatment, but the clinical efficacy of such approaches varies on a case-by-case basis.

As in this example, treating MABC infections requires huge resources and effort for each patient. As the treatment is not scalable, it is fortunate that MABC infections are not contagious or frequent. However, as we have seen with COVID, a single mutation or variant could increase transmissibility, potentially creating an epidemic of dire proportions.

Further, MABC’s increased antibiotic resistance and ability to acquire immunity to antibiotics, sometimes over the course of weeks, make it part of the challenging trend of “superbugs” that can defy traditional treatments. Additional treatment avenues, such as phage therapy, have been explored to address MABC. However, obstacles remain in scalability as the strategy is not a ‘one-size-fits-all’.2

Despite the challenges, the pursuit of alternative treatments for MABC might also lead to new insights into cures for other more prevalent diseases like tuberculosis, which is increasingly becoming antibiotic resistant and is dormant in about a third of humanity.3

Process and Discovery

After learning about this disease and meeting Varun, we worked together to develop an antibody against MABC. We invited Varun to visit our company and spend the summer driving the project (Figure 1). Varun joined our team as an intern, not only to provide insight into the specific needs of those affected by MABC, but also to actively contribute to the project through his own research and analysis. This experience not only allowed us to gain a deeper understanding of the disease and potential treatment options, but also greatly impacted Varun, as he was able to work firsthand on the entire process of designing an antibody from scratch and improving the efficacy in the quest for a cure.

Scientist in Macromoltek laboratory
Figure 1. Varun testing antibodies in the Enzyme-Linked Immunosorbent Assay test in the Macromoltek laboratory.

At Macromoltek, we use computational biology to design antibodies. Antibody drugs play a critical role in infectious diseases, cancer, autoimmune diseases, and inflammation. However, experimental methods for the generation of therapeutic antibodies, such as using immunized mice or phage display are time-consuming and cannot always be targeted to a specific antigen epitope. In both cases, the process is akin to searching for a needle in a haystack versus tailoring the antibody to the problem of interest. The design of an antibody against a unique epitope on the surface of a protein is an important application of computational methods for antibody design.4-6 Towards this end, we have devised algorithms for computational library creation and selection. As we computationally search the space around a particular epitope, we can design antibodies to bind only to that epitope without the need to produce millions of antibodies in the lab. This search leads to a small number of antibodies for further experimental testing. Our algorithms combine structural biology and machine learning techniques to reduce the search space to something tractable and epitope-directed.

As a small biotech company, we rely on collaborations with academic institutions or large pharmaceutical companies to give us targets to design around or against. These targets are often of high value and priority to pharmaceutical companies because they affect a large number of people. Without dedicated personnel for different diseases, we are caught with the dilemma of either focusing our discovery platform on a single disease or working on projects that outside parties bring us.

There are multiple benefits to antibody treatments compared to antibiotic treatments.7,8 First, antibodies are more effective than antibiotics due to the larger surface area and stronger binding to the pathogen. Second, antibody treatments also deliver immunity from future infections, as they can stay in the bloodstream without harming other cells. Third, antibody treatments have fewer side effects than antibiotics and do not negatively impact other organs, such as the liver or kidneys, due to their ability to be tailored to specific pathogens and bacteria, an ability that antibiotics lack.

To create functional antibodies targeting MABC, we designed antibodies using computational methods to find potential candidates. After identifying a target protein and creating a computational model, we used deep learning algorithms to generate a map of possible epitopes on the protein surface. For each potential epitope, we used structure-based methods to search the sequence and conformational space on several antibody scaffolds to generate paratopes that matched the selected epitope. This process generated tens of thousands of sequence/structure pairs that were refined and optimized to ensure the final antibody sequences can easily be made in the lab and have biophysical properties to make them into a drug. The library of antibodies and their structures were then subjected to a second deep learning model to select the antibodies with the best epitope/paratope match for experimental validation in efficacy and specificity.

In comparison to the traditional trial and error method of antibody discovery that can take up to two years to yield a hit, this entire process was done over the course of a summer and resulted in two strong candidates. With traditional methods, tens of thousands to millions of antibodies would need to be filtered in the lab. This same amount of antibodies were “tested” computationally in a more efficient manner. This method of designing custom antibodies provides a blueprint of possible treatment vectors for various diseases that could help cure more prevalent diseases, such as tuberculosis, COVID-19, and potentially cancer.

Looking back at the beginnings of this project, we started with the goal of preventing individuals diagnosed with the MABC infection from experiencing the tortuous treatment process that Varun’s mother endured. To this end, we have made initial process developing antibody hits against a protein that is crucial to the life-cycle of the disease. The next step is for these resulting antibodies to be tested in vivo to show efficacy and move into clinical trials. Only then can such an approach be implemented into the existing treatment structure. Varun’s journey with our company has been a powerful reminder of the importance of patient advocacy and the potential for biotech start-ups to make a meaningful impact in the fight against rare diseases. The dedication and perseverance of individuals like Varun can inspire others and drive innovation in the field of medicine.

 

References

  1. Jarand J, Levin A, Zhang L, et al. Clinical and Microbiologic Outcomes in Patients Receiving Treatment for Mycobacterium abscessus Pulmonary Disease. Clinical Infectious Diseases 2011;52(565-571, doi:10.1093/CID/CIQ237
  2. Shahraki AH, Mirsaiedi M. Phage Therapy for Mycobacterium Abscessus and Strategies to Improve Outcomes. Microorganisms 2021;9(3):596; doi: 10.3390/microorganisms9030596
  3. Cohen A, Mathiasen VD, Schon T, et al. The global prevalence of latent tuberculosis: a systematic review and meta-analysis. Eur Respir J 2019;54(3), doi:10.1183/13993003.00655-2019
  4. Bradbury A, Pluckthun A. Reproducibility: Standardize antibodies used in research. Nature 2015;518(7537):27-9, doi:10.1038/518027a
  5. Fleishman SJ, Whitehead TA, Ekiert DC, et al. Computational design of proteins targeting the conserved stem region of influenza hemagglutinin. Science 2011;332(6031):816-21, doi:10.1126/science.1202617
  6. Strauch EM, Fleishman SJ, Baker D. Computational design of a pH-sensitive IgG binding protein. Proc Natl Acad Sci U S A 2014;111(2):675-80, doi:10.1073/pnas.1313605111
  7. Safir MC, Bhavnani SM, Slover CM, et al. Antibacterial Drug Development: A New Approach Is Needed for the Field to Survive and Thrive. Antibiotics (Basel) 2020;9(7), doi:10.3390/antibiotics9070412
  8. Zurawski DV, McLendon MK. Monoclonal Antibodies as an Antibacterial Approach Against Bacterial Pathogens. Antibiotics (Basel) 2020;9(4), doi:10.3390/antibiotics9040155

 

Monica Berrondo ([email protected]) is CEO and Varun Thvar is an intern at Macromoltek, Austin, Texas, USA.

GEN Biotechnology, published by Mary Ann Liebert, Inc., is the new, marquee peer-reviewed journal publishing outstanding original research and perspectives across all facets of the biotech industry. This article was originally published in the February 2023 issue of GEN Biotechnology, Volume 2, Issue 1.