By combining truly personalized science with artificial intelligence, we can predict how effective different treatment options would be in battling your dog's cancer.
Every patient is unique, so two patients with the same medical condition may respond differently to the available treatment options. Precision medicine is an emerging approach for improving the success rate of medical interventions by identifying the treatment options most likely to be effective for individual patients.
What is functional precision medicine?
In addition to conventional precision medicine approaches based on biomarkers and patient information, ImpriMed uses an innovative and powerful approach called “functional precision medicine.” Our approach is “functional” because we directly test how effectively anticancer drugs disrupt the normal functioning of a patient's live cancer cells.
ImpriMed’s approach to precision oncology is unique in several ways.
The result is a unique precision medicine product that enables you and your veterinarian to quickly find the best drugs for treating your pet's cancer.
How can we get live cancer cells to the lab for testing? This is a real challenge for most companies. The logistics associated with keeping cells alive long enough to be tested are difficult to overcome.
ImpriMed innovated past these challenges by developing a proprietary transport medium that keeps cells healthy during shipping and a web portal to make ordering, processing, and receiving samples easy and fast.
Drug sensitivity shows us how effective a drug can be for each patient. Our drug panel currently includes 13 anticancer drugs that are commonly used for treating canine lymphoma and leukemia.
Flow cytometry quantifies the levels of 10 different proteins found on and within tumor cells. This test provides prognostic and diagnostic information for leukemia and lymphoma.
PARR (PCR for Antigen Receptor Rearrangement) tells us the genetic lineage of the malignancy. This test lets us give veterinary oncologists detailed information about the expansion of B-cell and T-cell clones.
To determine which drugs elicit a positive clinical response, we conducted an initial study to collect real-world data. We used the data from this clinical study to train AI models so they can predict the ways in which individual patients will respond to a panel of commonly used anticancer drugs. After the initial clinical study, we partner with our amazing network of customers to continuously grow our database of clinical outcomes. These data are added to our existing database and used to further refine the AI models.
Our outcomes database now includes more than 2,500 canine patients and is growing by the month. Due to ImpriMed’s continuous learning process, our AI models are constantly increasing in accuracy with more time and more customers.
Based on our established technology and workflow, we performed clinical studies to evaluate correlation between our predictive value and the actual clinical response to various anticancer drugs.
Published in Haematologica, Sept 2023
Prognostic value of European Leukemia Net 2022 criteria and genomic clusters using machine learning in older adults with acute myeloid leukemia
Published in npj Precision Oncology, May 2023
ML-based sequential analysis to assist selection between VMP and RD for newly diagnosed multiple myeloma
Published in Veterinary Sciences, July 2024
Prognostic Utility of the Flow Cytometry and Clonality Analysis Results for Feline Lymphomas
Read the Full ArticlePublished in Frontiers in Oncology, Feb 2024
Multimodal machine learning models identify chemotherapy drugs with prospective clinical efficacy in dogs with relapsed B-cell lymphoma
Read the Full ArticlePublished in Veterinary Sciences, Dec 2021
Predicting Dynamic Clinical Outcomes of the Chemotherapy for Canine Lymphoma Patients Using a Machine Learning Model
Read the Full ArticlePublished in Veterinary and Comparative Oncology, Oct 2020
Predicting likelihood of in vivo chemotherapy response in canine lymphoma using ex vivo drug sensitivity and immunophenotyping data in a machine learning model
Read the Full Article