Adult

Example 2

Twenty-eight (28) healthy, adult male and female (non-childbearing potential) subjects were enrolled in the study in total; 14 subjects in each study part (Parts 1 and 2). A minimum of 8 female subjects were enrolled in the study (i.e., a minimum of 4 female subjects per study part). Each subject participated in either Part 1 or Part 2, but not both.

Part 1

On Day 1 of Treatment Period 1, a single oral dose of 20 mg mitapivat sulfate was administered. Serial blood samples for plasma assay of mitapivat concentrations and its CYP3A4 metabolite, referred to herein as the “Metabolite” (structure below),

[Figure (not displayed)]
were collected from predose to 120 hours following administration of mitapivat sulfate. In Treatment Period 2, an oral dose of 200 mg itraconazole was administered once daily (QD) for 9 consecutive days (Day 1 through Day 9 of Treatment Period 2) with a single oral dose of 20 mg mitapivat sulfate coadministered on Day 5. Serial blood samples for plasma assay of mitapivat and the Metabolite concentrations were collected from predose to 120 hours following coadministration of mitapivat sulfate and itraconazole on Day 5.

In Treatment Period 1, mitapivat sulfate was administered orally with approximately 240 mL of water. In Treatment Period 2, on Days 1 to 4, itraconazole was administered alone immediately followed by approximately 220 mL of water, and on Day 5, itraconazole was administered first (no water) and was immediately followed by mitapivat sulfate administration with approximately 220 mL of water. Study drugs (mitapivat sulfate and itraconazole) were administered following an overnight fast of at least 10 hours on Day 1 of Treatment Period 1 (mitapivat sulfate only) and Day 5 of Treatment Period 2 (mitapivat sulfate and itraconazole), and subjects remained fasted for 4 hours after dosing. On all other dosing days, itraconazole was administered following a predose fast of at least 4 hours and subjects remained fasted for at least 2 hours after dosing.

Part 2

On Day 1 of Treatment Period 1, a single oral dose of 50 mg mitapivat sulfate was administered. Serial blood samples for plasma assay of mitapivat and the Metabolite concentrations were collected from predose to 120 hours following administration of mitapivat sulfate. In Treatment Period 2, an oral dose of 600 mg rifampin was administered QD for 12 consecutive days (Day 1 through Day 12 of Treatment Period 2) with a single oral dose of 50 mg mitapivat sulfate coadministered on Day 8. Serial blood samples for plasma assay of mitapivat sulfate and the Metabolite concentrations were collected from predose to 120 hours following coadministration of mitapivat and rifampin on Day 8.

In Part 2, study drugs were administered with approximately 240 mL of water on all dosing days including the coadministration of mitapivat sulfate and rifampin on Day 8 of Treatment Period 2. Mitapivat sulfate and rifampin was administered following an overnight fast of at least 10 hours on Day 1 of Treatment Period 1 (mitapivat sulfate only) and Day 8 of Treatment Period 2 (both mitapivat sulfate and rifampin) and subjects remained fasted for 4 hours after dosing. On all other dosing days, rifampin was administered following a predose fast of at least 4 hours and subjects remained fasted for at least 2 hours after dosing. There was a washout period of 7 days between the mitapivat sulfate dose in Treatment Period 1 and the first itraconazole (Part 1) or rifampin (Part 2) dose in Treatment Period 2. All study drugs were consumed within 5 minutes for both Part 1 and Part 2.

Example 1

The MCA-miner method disclosed herein in FIGS. 2A-2C, when used together with BRL, offers the power of rule list interpretability while maintaining the predictive capabilities of already established machine learning methods.

The performance and computational efficiency of the new MCA-miner is benchmarked against the “Titanic” dataset, as well as the following five (5) datasets available in the UCI Machine Learning Repository: “Adult,” “Autism Screening Adult,” “Breast Cancer Wisconsin (Diagnostic),” “Heart Disease,” and “HIV-1 protease cleavage,” which are designated as Adult, ASD, Cancer, Heart, and HIV, respectively. These datasets represent a wide variety of real-world experiments and observations, thus enabling the improvements described herein to be compared against the original BRL implementation using the FP-Growth miner.

All six benchmark datasets correspond to binary classification tasks. The experiments were conducted using the same set up in each of the benchmarks. First, the dataset is transformed into a format that is compatible with the disclosed BRL implementation. Second, all continuous attributes are quantized into either two (2) or three (3) categories, while keeping the original categories of all other variables. It is worth noting that depending on the dataset and how its data was originally collected, the existing taxonomy and expert domain knowledge are prioritized in some instances to generate the continuous variable quantization. A balanced quantization is generated when no other information was available. Third, a model is trained and tested using 5-fold cross-validations, reporting the average accuracy and Area Under the ROC Curve (AUC) as model performance measurements.

Table 1 presents the empirical result of comparing both implementations. The notation in the table follows the definitions above. To strive for a fair comparison between both implementations, the parameters rmax=2 and smin=0:3 are fixed for both methods, and in particular for MCA-miner μmin=0:5 and M=70 are also set. The multi-core implementations for both the new MCA-miner and BRL were executed on six parallel processes, and stopped when the Gelman & Rubin parameter satisfied {circumflex over (R)}≤1.05. All the experiments were run using a single AWS EC2 c5.18×large instance with 72 cores.

It is clear from the experiments in Table 1 that the new MCA-miner matches the performance of FP-Growth in each case, while significantly reducing the computation time required to mine rules and train a BRL model.

Example 2

In the following experiments, a mouse model of RVO, which induces reproducible retinal edema was used. RVO is the model that was used for testing anti-VEGF therapies for DME. Brown et al., Ophthalmology 117, 1124-1133 el 121 (2010); and Campochiaro et al., Ophthalmology 117, 1102-1112 e1101 (2010). I n this model, Rose Bengal, a photoactivatable dye, is injected into the tail veins of adult C57B16 mice and photoactivated by laser of retinal veins around the optic nerve head. A clot is formed and edema or increased retinal thickness develops rapidly. Inflammation, also seen in diabetes, also develops.

Fluorescein leakage and maximal retinal edema, measured by fluorescein angiography and optical coherence tomography (OCT), respectively, using the Phoenix Micron IV, is observed 24 h after RVO. Retinal edema is maintained over the first 3 days RVO. By day 4 the edema decreases and the retina subsequently thins out. In addition to edema formation there is evidence of cell death in the photoreceptor cell layer by day 2 after RVO.

In this example, mice were anesthetized with intra-peritoneal (IP) injection of ketamine and xylazine. One drop of 0.5% alcaine was added to the eye as topical anesthetic. The retina was imaged with the Phoenix Micron IV to choose veins for laser ablation using the Phoenix Micron IV image guided laser. One to four veins around the optic nerve head were ablated by delivering a laser pulse (power 50 mW, spot size 50 μm, duration 3 seconds) to each vein.