An artificial intelligence (A.I.) platform called CURATE.AI may be used to determine the optimal doses of investigational therapy ZEN-3694 and Xtandi (enzalutamide) to be given to patients with metastatic castration-resistant prostate cancer (mCRPC).
The approach led to a complete and durable halt of disease progression in a patient included in a Phase 1/2 trial.
The research, “Modulating BET Bromodomain Inhibitor ZEN-3694 and Enzalutamide Combination Dosing in a Metastatic Prostate Cancer Patient Using CURATE.AI, an Artificial Intelligence Platform,” appeared in the journal Advanced Therapeutics.
Cancer chemotherapy is typically administered at fixed doses. Despite being a standard for clinical care, this approach does not take into account the changing response to medications during the course of treatment, which affect its effectiveness and tolerability.
Also, each patient responds to combination treatment differently, as drug synergy — medications working together to boost efficacy — is patient-specific.
To address this limitation, the team, led by the National University of Singapore (NUS) conducted a one-year study at the UCLA Institute of Urologic Oncology.
They developed CURATE.AI, which has been used to optimize liver transplant immunosuppression and tuberculosis therapy, among other indications. This A.I. platform intends to increase treatment effectiveness by rapidly identifying the doses of optimal medications.
CURATE.AI uses the patient’s own clinical data — including treatment doses and changes in tumor size, as well as blood levels of cancer biomarkers — to develop an individualized profile for the best possible treatment outcome.
“As a patient’s body and the cancer itself evolve during treatment, the CURATE.AI profile evolves as well, enabling the clinical and engineering teams to optimize care for the entire duration of treatment, an unprecedented advance for combination therapy,” Dean Ho, the study’s senior author, said in a press release.
The patient described in this study was participating in a Phase 1b/2a clinical trial (NCT02711956) assessing the combination of ZEN-3694 and Xtandi (by Pfizer and Astellas Pharma) in mCRPC patients. ZEN-3694 is a BET inhibitor developed by Zenith Epigenetics that suppresses levels of key cancer genes.
The patient had failed prior Zytiga (abiraterone acetate) and started the combination six weeks after discontinuing his prior treatment.
Doses of both ZEN-3694 and Xtandi were initially adjusted to better manage the patient’s reported quality of life. These initial doses and the patient’s level of prostate specific antigen (PSA) — a common test to screen for prostate cancer — were then used to create his personalized CURATE.AI profile.
This led to identifying a ZEN-3694 dose 50% lower than the dose used prior to artificial intelligence. Importantly, dosage guidance with the A.I. platform led to the lowest PSA levels over the course of the study, an indicator of treatment efficacy.
Slight increases in ZEN-3694 along treatment enabled marked reductions in PSA levels, also showing the treatment candidate’s role in suppressing cancer.
Then, computed tomography (CT) imaging of the patient’s lesions revealed that disease worsening had stopped. Patient care is still ongoing.
“This study represents an important gateway toward the use of AI to sustain and optimize treatment efficacy,” the investigators wrote.
“Using CURATE.AI to dynamically modify drug doses and successfully treat a metastatic cancer patient represents a landmark breakthrough for the use of AI to truly personalize patient care,” Ho said. The scientist also said that this is expected to “dramatically improve response rates for all combination therapies” in cancer and all other diseases.
“We can also expect CURATE.AI to markedly reduce the costs of drug development,” Ho added. “We are excited that CURATE.AI could ultimately enhance patient accessibility to important new combination therapies, saving lives in the process.”
Allan Pantuck, the study’s lead author, said that unlike the frequent switch of medications when patients treated with fixed-dose approaches no longer respond to treatment, “CURATE.AI has shown that patients can still respond to the therapies that have seemingly stopped working by continuously identifying the patient’s optimal dosing parameters.”
Several clinical trials of CURATE.AI-guided combination therapy for oncology and other applications are underway. Overall, the CURATE.AI scientists expects to use the platform in the prevention of transplant rejection, as well as in adult and pediatric cancers, cardiovascular medicine, diabetes, infectious diseases, among other indications.
Of note, two of the study’s authors are inventors on pending and issued patents covering artificial intelligence-guided combination therapy. The research was conducted in collaboration with Zenith.