Startups Meet Pharma Cohort 2020 – The Amgen MM challenge
Choosing the right treatment for Multiple Myeloma patients is challenge #5 set by Amgen and four startups are shortlisted to work with them.
Sixteen startups from Germany, Austria, UK, France, Belgium, Netherlands, Ireland, Poland, Italy, Finland, Greece, Spain, Estonia and Israel have been selected for the Startups Meet Pharma 2020 cohort. This year, five challenges have been released in collaboration with four EIT Health Partners: Amgen, Bayer, Beiersdorf and Boehringer Ingelheim.
Let’s take a closer look at Challenge#5: Choosing the right treatment for Multiple Myeloma patients.
Multiple Myeloma (MM) patients often experience relapses and then, it is very difficult to choose the right second line treatment, as guidelines are very open for all available possibilities and no direct head-to head comparison of the newest medical options is available. There are 3 novel options available for relapsed MM (for second and further lines): a) Daratumumab – a monoclonal antibody targeting CD38, b) Elotuzumab – a monoclonal antibody, not available in Europe, c) Carfilzomib – a next generation PI inhibitor, very effective, more effective as second line rather than latter lines, targets proteosoma, and d) Ixazomib – an oral PI inhibitor, not as effective, but offering convenience of once daily dosing at home, targets proteosoma. Amgen is now looking for novel technologies to assess the available data or tap into other data that might be available through registries, in order to make these recommendations (a decision-making tool).
A total of 7 applications were received for this challenge and four of them were shortlisted to participate in the programme: Proteona Gmbh (Germany), BioXplor LTD (Ireland), Real Research Sp. Z oo (Poland) and KAZAAM Lab SRL (Italy).
Here is a sneak peak into their solutions:
Your solution to the challenge: Proteona is using single cell proteogenomics to build a database of multiple myeloma tumour responses to various therapies and then mining that database with proprietary AI tools to identify the best treatment option for new cancer patients. A key challenge in multiple myeloma management is patient and tumour heterogeneity. Each patient has a unique set of tumour cells, requiring a unique treatment plan. Proteona developed a single cell multi-omics analysis platform tool for assisting personalized multiple myeloma treatment decision making. It measures protein, overall gene and mutation expression in multiple myeloma patient samples on the single cell level, therefore fully capturing the heterogeneity of each cancer. For each sample, Proteona is able to extract thousands of data points from each cell and these data are compared to a proprietary database of tumour cells with known treatment outcomes. The information is used to automatically identify the cancer cell subtypes present and recommend which therapy option is best to kill those specific tumour cells. By running this analysis at the single cell level, we are able to identify how unique clones may behave, thus delay or entirely avoid an additional relapse.
What makes it a good fit? Proteona addresses the key challenges in multiple myeloma treatment selection.
- Understand heterogeneity. It is difficult to make a treatment decision for multiple myeloma because of the large biological variations within each patient as well as between different individuals. We capture intra-patient heterogeneity by analysing up to ten thousand cell tumor and normal cells from each sample. In addition, we access inter-patient heterogeneity by building a multiple myeloma database that comprises patient sample analysis data as well as known treatment responses. Therefore, we provide a solution to truly understand and address the variability in multiple myeloma.
- Manage relapse. Multiple myeloma is still considered an incurable disease due to constant relapse. Proteona helps to manage multiple myeloma relapse by monitoring the evolution of tumor clones during the course of treatment on single cell level. In this way, we are able to detect a minor population of tumor cells that escapes the current treatment regime, which may be responsible for a potential relapse.
- Evidence based guidance to treatment selection. Currently, selecting the right treatment relies heavily on clinician experience. The tumor heterogeneity as well as the wide selection of treatment regimes make a comprehensivehead-to-head comparison in a clinical trial setting unfeasible. Single cell proteogenomic analysis provides millions of data points per sample, which gives rich information for advanced data mining and analysis. By mapping data from patient samples to our proprietary database, we are able to flag out individual cancer clones, and identify how similar tumor cells from existing dataset responded to therapy, hence suggesting potential treatment strategy based on past evidence.
Major expectations from the programme: 1) Obtaining feedback on our scientific and business approach, 2) Sharing of insight and experience with other startup companies in the programme, 3) Boosting the visibility and recognition of Proteona and 4) Access to the EIT network of experienced biotech/biomedical professionals.
Long-term vision for your startup: We believe that single cell multi-omics will be an important driver in precision medicine. Proteona aims to be the leader in using single cell multi-omics to address clinical questions.
Your solution to the challenge: “Choosing the right treatment for Multiple Myeloma (MM) patients – in collaboration with Amgen” is PHOENIX, a decision support system able to integrate and analyze huge amounts of data, coming both from public databases on molecular interactions, genotype-phenotype associations, biological ontologies, and other databases own of health providers (e.g., on Multiple Myeloma – MM).
What makes it a good fit? The key feature of PHOENIX is its ability of bringing into light information which is hidden in the integrated data, then providing suggestions for the choice of the most appropriate therapy to the so obtained patient categories. We propose to integrate and organize data on MM patients, in particular on their behavioral, clinical and genetic profiles, together with other data coming from external databases, and to categorize patients according to specific features, extracted by PHOENIX, able to cluster individuals with similar responses to the considered therapies. Specific services could be added to PHOENIX in order to suggest the most appropriate therapy for new MM patients, according to their category.
Major expectations from the programme: This program allows us to have the chance of applying our services on very important use-cases. The challenge of providing a successful solution for a specific clinical problem is highly motivating for our company. Moreover, the opportunity of working side to side with important pharma companies would have a great impact for KAZAAM Lab, and hopefully would provide new ideas to our research team.
Long-term vision for your startup: Our vision is that KAZAAM Lab becomes a reference in the global market of Precision Medicine for the developing of decision support software services based on the integration and analysis of big omics data. We expect to collaborate in partnership with pharmaceutical companies, clinical institutes and research laboratories.
Your solution to the challenge: We want to give multiple myeloma patients a bigger chance to recover by testing samples from the individual cancer-patients to predict tumor response for a different therapies. It will make a treatment more personalised. Testing different therapies on samples from real patients before giving actual therapy will allow to choose treatment which gives the best result for every patient.
What makes it a good fit? Choosing a good treatment for a cancer and especially multiple myeloma is very difficult. There are many treatment options but mostly we are not sure that we will choose good therapy. Our solution to this problem is to test efficiency of different therapies before giving them to patients. It will allow to choose the therapy with the highest effectiveness and give it to the patient. Potentially our test will be quick and available for everyone. We can test the efficiency of any available drug with accuracy of results and repeatability of results. Also using samples from real patients for a new drug development allow us to get more effective drug testing.
Major expectations from the programme: Our main expectations are to learn how to cooperate with global-size company with international reach-out and help to improve in new drug development and cancer treatment. Our dream would be to test closed drug programs that have been rejected but there is a possibility that drug may can work and help but have not been sufficiently tested and rejected too fast. We would like to test them on our 3D cell-culture system with patient derived cells. Our system is fully repeatable and much better mimic natural environment as well as better captures the morphology of the tumor compared to standard 2D test using cancer cell lines. It is possible that our system will give different results than previously observed with standard in vitro testing.
Long-term vision for your startup: In long term vision we want to supply 3D cell culture solution for pre-clinical drug testing as well as for testing different chemo therapies for cancer-patients. From the patients perspective there are available tests for bacterial sensitivity to antibiotics and we want want to introduce similar solutions for chemotherapy testing and give them possibility of getting the best treatment.
Your solution to the challenge: BioXplor combines public and private data to prioritize candidates for drug repositioning and combinations based on safety and efficacy data using natural language processing and machine learning.
What makes it a good fit? As with Multiple Myeloma, drug combinations are emerging as the future therapeutic strategy for oncology and many other complex diseases, yet today there is no system, database or platform to combine different sources of efficacy and safety data to prioritize multi-therapeutics or multi-target therapeutics. BioXplor is building a computational platform to solve this problem.
Major expectations from the programme: To enable Amgen to identify drug combinations with it’s approved, late-stage, and pre-clinical candidates to prioritize novel multi-therapeutic candidates for multiple myeloma, and other diseases.
Long-term vision for your startup: BioXplor’s long-term vision is to build a data-driven platform to de-risk and prioritize drug combinations for complex diseases via partnering with Pharma and Biotech assets.