A startup in the tech industry aims to utilize artificial intelligence (AI) for assisting veterinarians in administering targeted drugs for treating cancer in dogs. However, the effectiveness of this tool is up for debate. When Cory Padilla noticed that his dog, Copper, had swollen lymph nodes in the neck, he knew it was time to take his beloved pet to the vet. Padilla shared with The Daily Beast that along with the swollen lymph nodes, Copper, a mix of beagle and German shepherd, was experiencing bouts of diarrhea and had droopy eyes. Nonetheless, cancer was not on Padilla's radar as a possible cause. "I thought it could be another disease commonly found in hounds, " Padilla said. However, his worries were confirmed by the veterinarian, who diagnosed Copper with lymphoma, one of the most prevalent dog cancers, after aspirating the swollen lymph nodes. "My heart sank, " Padilla remarked. He had every reason to be concerned since untreated canine lymphoma leads to the demise of most dogs within three to four weeks. Padilla immediately turned to the internet to educate himself about the disease and potential treatment options. It was during this research that he came across ImpriMed, a company established in 2017. ImpriMed employs AI and detailed analysis of live cancer cells to generate personalized predictions regarding drug efficacy for dogs with lymphoma. After the live cancer cells are received at ImpriMed's laboratory in California, they undergo flow cytometry testing, a technique that aids veterinarians in distinguishing between different lymphoma subtypes. The testing also includes cell analysis and drug sensitivity testing, which ranks 13 common anti-cancer drugs used for treating lymphoma and leukemia. "Our approach goes beyond selecting the appropriate treatment, " explained Sungwon Lim, CEO and co-founder of ImpriMed, to The Daily Beast. "It also helps avoid unnecessary expenses on treatments that might not effectively combat your pet's specific cancer. In the long run, this can save time and money, reducing the trial-and-error process often associated with cancer treatment. " Lim clarified that ImpriMed's predictions are made using proprietary AI algorithms that rely on treatment outcome data collected nationwide, as well as information obtained from live cancer cells. This allows ImpriMed to provide personalized predictions of drug efficacy for their clients and veterinarians to help determine the most effective course of treatment. Padilla was amazed by how AI analyzed the cancer at a cellular level and provided a panel of recommended chemotherapy treatments that would be most effective for his dog. Canine lymphoma is commonly treated with the CHOP protocol, a chemotherapy regimen consisting of a combination of four drugs: doxorubicin, cyclophosphamide, vincristine, and prednisone. "We know that 90 to 95 percent of dogs respond to CHOP, and this has been unequivocally proven in the literature to be the gold standard of treatment, " stated Jack O'Day, a veterinary oncologist in Fairfax, Virginia, who incorporates ImpriMed into his precision treatment practices. However, CHOP has its limitations. It does not account for the individuality of each patient or the financial constraints of owners. CHOP treatment can cost between $10, 000 and $25, 000, and not all dogs respond to it in the same way. "I wanted to go with CHOP, but it would cost an average of $500 to $600 per week, " Padilla explained. "I don't have that kind of money. " This is what made ImpriMed's service so appealing. It provided Padilla with valuable insights into Copper's lymphoma type and ranked the effectiveness of various drug options. "Knowing the most effective treatment allowed me to make decisions based on that information instead of blindly attempting different treatments, " Padilla said, adding that ImpriMed helped save both time and money during Copper's treatment process. While ImpriMed shows promise, it is not without its own costs.

The average ImpriMed Prediction Profile, which ranks anti-cancer drugs, costs pet owners between $1, 000 and $1, 500. This can be prohibitively expensive for many owners. Lim stated that ImpriMed offers its service to veterinary clinics at a fixed price, but the final cost varies depending on each clinic's pricing model. Another pet owner, Susie Araiza from Texas, shared with The Daily Beast that she did not find the ImpriMed Prediction Profile particularly helpful when it came to navigating treatment options for her dog, Precious, who experienced a lymphoma relapse. For instance, the first drug recommended by ImpriMed caused Precious to become anemic, necessitating an emergency iron injection from her veterinarian. "We stopped giving her that drug, and she returned to normal, " Araiza recalled. "I found it strange that the very first drug. . . she couldn't tolerate. " Araiza mentioned that giving Precious the second drug recommended on ImpriMed's panel, doxorubicin, resulted in her dog being hospitalized for over a week. Ultimately, Araiza felt it would have been better to spend the money she used for the panel directly on treatment rather than relying solely on its recommendations. "I thought it was going to kill her, " Araiza remembered. "The oncologists told us they would probably never give her that drug again. " Araiza's experience, along with experiences shared by other pet owners in social media groups dedicated to canine lymphoma, raises concerns about the effectiveness of the panel without considering the complete clinical picture of an animal. Veterinary oncologist Daniela Korec informed The Daily Beast that she primarily uses ImpriMed for flow cytometry, an advanced type of cell analysis, rather than predictive drug panels. "I review the Prediction Profiles, and I believe it's important to note that our knowledge of drug testing with lymphoma still provides a lot of predictive abilities, " said Korec. Although none of her patients have undergone Prediction Profiles with ImpriMed, Korec supports the technology with the caveat that owners also consider her expertise as a clinician treating the animal, rather than solely relying on the information provided in the report. Parminder Basran, an associate research professor and medical physicist at Cornell University's College of Veterinary Medicine, warns against depending solely on AI for patient care trajectory, especially in broad areas like cancer treatment. "These are tough questions that veterinary medical experts are better equipped to handle, rather than an AI system, " Basran, whose research lab explores the use of AI in veterinary healthcare, told The Daily Beast. Basran finds the use of AI models in veterinary medicine "exciting and certainly very interesting, " but acknowledges that there is still "a lot of work to do" before fully trusting AI systems in both veterinary and human medicine. While pet owners, veterinary oncologists, and medical AI experts hold varying degrees of confidence in ImpriMed's predictions, Lim stands by his product, despite acknowledging the challenges faced in treating animals. "Unfortunately, some dogs do not respond to any available drugs, so even if we predict a high probability of success for a particular drug, it may still yield no positive response, " he explained. "Furthermore, like all medical tests and AI predictions, our drug response predictions can sometimes be incorrect. " Nevertheless, Lim believes that continuously pushing the boundaries is the key to advancing modern medicine. This is why ImpriMed's team incorporates real-world statistics to assess whether their product offers clinical benefits in everyday use and conducts post-market clinical research studies. In December 2023, the company released a study based on combined data from 60 independently treated dogs across the country. The study reported three times longer patient survival and four times higher complete remission rates in patients treated in high accordance with ImpriMed's AI predictions. This is encouraging news for Lim. For the first time, pet owners and veterinarians have a tool that has the potential to specifically target an animal's cancer using an AI-predicted panel of effective drugs. However, there is still a need to address issues of accessibility and accuracy. While the prospect is enticing, treating a dog's disease, not to mention investing in predictive AI technology, may never be a reality for many pet owners due to the cost of cancer treatment and the research that advances curative therapies. Nevertheless, when it comes to our pets, many of us would spare no effort to help them. Lim concluded, "Our objective is to provide veterinary oncologists and their clients with the best tools available to fight cancer in their beloved pets. "