Real-World Outcomes of an Automated Physician Support System for Genome-Driven Oncology

Abstract

Purpose: Matching patients to investigational therapies requires new tools to support physician decision making. We designed and implemented Precision Insight Support Engine (PRECISE), an automated, just-in-time, clinical-grade informatics platform to identify and dynamically track patients on the basis of molecular and clinical criteria. Real-world use of this tool was analyzed to determine whether PRECISE facilitated enrollment to early-phase, genome-driven trials.

Materials and methods: We analyzed patients who were enrolled in genome-driven, early-phase trials using PRECISE at Memorial Sloan Kettering Cancer Center between April 2014 and January 2018. Primary end point was the proportion of enrolled patients who were successfully identified using PRECISE before enrollment. Secondary end points included time from sequencing and PRECISE identification to enrollment. Reasons for a failure to identify genomically matched patients were also explored.

Results: Data were analyzed from 41 therapeutic trials led by 19 principal investigators. In total, 755 patients were accrued to these studies during the period that PRECISE was used. PRECISE successfully identified 327 patients (43%) before enrollment. Patients were diagnosed with 29 tumor types and harbored alterations in 43 oncogenes, most commonly ERBB2 (21.3%), PIK3CA (14.1%), and BRAF (8.7%). Median time from sequencing to enrollment was 163 days (interquartile range, 66 to 357 days), and from PRECISE identification to enrollment 87 days (interquartile range, 37 to 180 days). Common reasons for failing to identify patients before enrollment included accrual on the basis of molecular alterations that did not match pre-established PRECISE genomic eligibility (140 [33%] of 428) and external sequencing not available for parsing (127 [30%] of 428).

Conclusion: PRECISE identified 43% of all patients accrued to a diverse cohort of early-phase, genome-matched studies. Purpose-built informatics platforms represent a novel and potentially effective method for matching patients to molecularly selected studies.

FIG 1.

Evolution of PRECISE (Precision Insight Support Engine) functionality. Throughout the development of PRECISE, multiple functionalities were gradually enhanced. The initial iteration (version 1) of PRECISE involved generating cohorts on the basis of complex genetic and clinical criteria defined by the study’s principal investigator (PI), which could then be sent to the PI at defined intervals. The capability of PRECISE was later enhanced (version 2) to enable PI notifications triggered by certain events of interest, such as an upcoming patient appointment or computed tomography scan. Present day (version 3) PRECISE can also incorporate a patient’s prior treatment history and allows for direct notification of the patient’s treating oncologist that a patient may be eligible for a study, often prompting an exchange between the treating oncologist and PI that initiates the patient’s future enrollment. Future development of PRECISE includes harnessing machine learning algorithms and continuous feedback loop analytics to enhance efficiency and accuracy of trial–patient matches. MD, medical doctor; Peds, pediatrics.

FIG 2.

Source of patient enrollment by genome-driven study. (A) In aggregate, 43% (327 of 755) of all patient enrollments were facilitated by Precision Insight Support Engine (PRECISE). (B) Each column depicts patient enrollments by study principal investigator, with patient enrollment facilitated by PRECISE shaded in blue and patient enrollment not facilitated by PRECISE shaded in red. The absolute number of patients in each category is labeled above (non-PRECISE) and below (PRECISE enrollment) each column. (C) Each column represents a unique study, with the absolute number of patients in each category labeled above (non-PRECISE) and below (PRECISE enrollment) each column.

FIG 3.

Time courses to patient enrollment. (A) Scatterplot depicting the timing of enrollment by individual patient relative to study activation (opening for accrual). Each series of three dots on a single vertical axis represents a single patient’s course, from initial sequencing (blue dot) to identification by PRECISE (Precision Insight Support Engine; red dot) to enrollment in the study (gray dot). (B) Box and whisker plot showing the interval of time from sequencing to enrollment in the study (left; median, 163 days) and from identification by PRECISE to enrollment in the study (right; median, 87 days).

FIG 4.

CONSORT diagram. Understanding factors that contribute to cohort patient attrition. Diagram depicts factors that led to attrition from a representative Precision Insight Support Engine (PRECISE) cohort for a phase I and II study. Reasons for permanent or temporary exclusion from study qualification are depicted in blue (criteria that is available for use by PRECISE), red (criteria that is sometimes available and could potentially be captured as an extractible structured data element), or teal (criteria that is not yet available for use by PRECISE). (*) Listed as alive in electronic medical record but actually deceased.

FIG A1.

Source of patient enrollment by principal investigator (PI), study characteristic, and PRECISE (Precision Insight Support Engine) cohort creation date. (A) Each column depicts patient enrollments by study PI, with PIs with the most years in practice on the left and the fewest years in practice on the right. Patient enrollment facilitated by PRECISE is shaded in blue and patient enrollment not facilitated by PRECISE is shaded in red. The absolute number of patients in each category is labeled above (non-PRECISE) and below (PRECISE) enrollment each column. (B-D) Each column represents a unique study, with the type of study (pilot, phase I, phase I and II, or phase II; [B]), tumor type eligibility (multiple, breast, lung, or other; [C]), and date of cohort creation (earliest on the left; [D]). The earliest cohort was created on April 16, 2014, and the latest cohort was created on October 11, 2017. The absolute number of patients in each category is labeled above (non-PRECISE) and below (PRECISE enrollment) each column.