Artificial Intelligence–Supported Precision Medicine Platform Improves Treatment Outcomes in Hematologic Cancers

The single-cell functional precision medicine treatment, an artificial intelligence strategy helps to create therapies for patients with hematologic cancers.

Results from the EXALT-1 study highlighted the potential real-world impact of utilizing an artificial intelligence (AI)–supported precision medicine platform to determine the most effective therapies for patients with late-stage hematological cancers, according to a press release from Exscientia.1

By using the company’s single-cell functional precision medicine (scFPM) platform at a median follow-up of 23.9 months, 54% of patients experienced a 1.3-fold enhanced clinical progression-free survival (PFS) benefit. Of these 30 patients, 40% had responses that lasted 3 times longer than expected. The PFS for patients utilizing scFPM was found to have significantly increased (P =.0093).

"The EXALT-1 study was the first-ever prospective study of its kind, demonstrating the potential of a functional precision medicine platform to identify the optimal therapy for an individual patient. The results of the trial are clear: patients who were treated using recommendations guided by single-cell drug screenings from real-time biopsies remained progression-free longer as compared [with] their previous line of therapy,” principal investigator Philipp P. Staber, MD, PhD, a hematologist and oncologist at Medical University Vienna, said in a press release.

The platform utilizes AI technology to perform a single cell, high content analysis of individual patient tissue samples in order to garner clinically important insights upon which treatments can be based.

The study evaluated 76 patients, 56 of whom received treatment according to scFPM and 20 were treated according to physician’s choice. Patients had a median age of 64 years, and the median number of treatment lines prior to enrollment was 3. From sampling to the scFPM report, the median time was 5 days, and median time to treatment was 21 days.2

The study was comprised of patients who had common and rare hematologic malignancies such as acute myeloid leukemia (25%), B-cell non-Hodgkin lymphoma (46%), and T-cell non-Hodgkin lymphoma (28%).

After 12 months on treatment, 23% of patients were progression free compared with 5% on the previous treatment. Patients had an objective response rate of 55% among those whose treatment was guided with scFPM as well as 60% for the lymphoid subgroup and 41% for myeloid neoplasms.

In the primary analysis cohort, investigators reported that the PFS ratio was 1.47. Additionally, 14% of patients received an allogenous hematopoietic stem cell transplantation or donor lymphocyte infusion after reaching a complete response with scFPM-guided therapy; this did not translate to a PFS benefit vs complete responders who did not receive the consolidation treatment.

Pretreatment performance status that was altered via benefit garnered from guidance with scFPM with a PFS ratio of 1.3 or greater was reached in 62% of patients who had an ECOG score of 1 or less and in 35% of patients who had an ECOG score of more than 1. For those with a score of 1 or less, the median PFS was 207 days compared with 29 days for those whose score was greater than 1.

Patients with T-cell non-Hodgkin lymphoma had an increased median PFS of 235 days compared with 60 days for those with B-cell non-Hodgkin lymphoma. Those with TP53 variants had a significantly shorter PFS than those without.

Characteristics that did not have an impact on PFS included age, sex, lineage, number of previous treatment lines, disease subgroup, and time from sampling to treatment.


  1. Publication of EXALT-1 trial in Cancer Discovery demonstrated first AI-supported functional precision medicine platform to improve cancer treatment outcomes. News Release. Exscientia. October 11, 2021. Accessed October 14, 2021.
  2. Kornauth C, Pemovska T, Vladimer GI, et al. Functional precision medicine provides clinical benefit in advanced aggressive hematological cancers and identifies exceptional responders. Cancer Discov. Published online October 11, 2021. doi:10.1158/2159-8290.CD-21-0538