Exhibit 99.2

SELECTED EFFICACY AND BLEEDING/SPOTTING OUTCOMES FROM THE SECURE TRIAL: A PHASE 3 STUDY OF AG200-15, AN INVESTIGATIONAL WEEKLY TRANSDERMAL CONTRACEPTIVE PATCH

Anita L. Nelson,(1) Andrew M. Kaunitz,(2) Robin Kroll,(3) James A. Simon,(4) Alfred N. Poindexter,(5) Joseph A. Chiodo,(6) Lisa Flood,(6) Elizabeth I.O. Garner(6)

(1)Los Angeles Biomedical Research Institute, Los Angeles, CA, USA; (2)University of Florida College of Medicine-Jacksonville, Jacksonville, FL, USA; (3)Seattle Women’s, Seattle, WA, USA; (4)George Washington University School of Medicine, Washington, DC, USA; (5)Baylor College of Medicine, Houston, TX, USA; (6)Agile Therapeutics, Princeton, NJ, USA

Objective: To assess selected efficacy and bleeding/spotting outcomes with the use of AG200-15 (levonorgestrel and ethinyl estradiol patch).

Design: The SECURE trial was a single-arm, open-label, 1-year (13-cycle), healthcare-company funded, Phase 3, IRB-approved study conducted at 102 US sites (Clinical Trials Identifier: NCT02158572). SECURE had no eligibility restrictions for weight or body mass index (BMI).

Materials and Methods: Pre-specified efficacy analyses in various subgroups were assessed using the Pearl Index (PI). Study subjects receiving AG200-15 recorded vaginal bleeding (requiring use of at least one tampon or sanitary pad) and spotting (requiring use of pantyliners only or no sanitary protection) daily in electronic diaries.

Results: A total of 2,031 subjects wore at least one patch and were included in the safety population (median age 26.0 years [range, 18, 60]; median BMI 26.8 kg/m2; median weight 71.9 kg); 1,736 subjects age <35 years were included in the efficacy analysis. A total of 2,017 subjects provided information on scheduled and unscheduled bleeding/spotting episodes by cycle (Table 1). The PI for women <35 years of age was 4.80 (95% CI, 3.55, 6.06). When analyzed by BMI, the PI was 4.22 and 5.41 in women with a BMI <26.8 kg/m2 and >26.8 kg/m2, respectively. Similarly, when analyzed by weight, the PI was 3.83 and 5.84 in women with a weight <71.9 kg and >71.9 kg, respectively. When analyzed by race, the PIs for subjects identifying as white, black/African American, and “other” were 5.06, 4.52 and 3.64, respectively; whereas when analyzed by ethnicity, the PIs for subjects identifying as Hispanic/Latino and non-Hispanic/Latino were 3.65 and 5.07, respectively. The most common hormone-related treatment-emergent adverse events were nausea (4.1%), headache (3.6%) and mood swings/changes/depression (2.8%).

Table 1. Mean Number of Bleeding/Spotting Episodes and Days Per Episode (±SD) By Cycle

An episode of bleeding and/or spotting was defined as one or more consecutive days of bleeding/spotting bounded on either end by >2 days of no bleeding or spotting. Unscheduled bleeding/spotting was defined as bleeding/spotting on days when not wearing a patch.

Conclusions: The efficacy of AG200-15, as assessed by the PI, differed by subgroups based on BMI, weight, and race/ethnicity. The mean number of bleeding/spotting episodes were generally similar throughout the study, while the length of episodes generally decreased.

Annotations

Results: A total of 2,031 subjects wore at least one patch and were included in the safety population (median age 26.0 years [18, 60]; median BMI 26.8 kg/m2; median weight 71.9 kg) and 1,736 subjects age <35 years were included in the efficacy analysis. [Table 14.1.1.1 run date 28Feb17; Table 14.1.2 run date 16Feb17] A total of 2,017 subjects provided information on scheduled and unscheduled bleeding/spotting episodes per cycle (Table 1). [Table 14.2.7.1.1 run date 23Mar17] The PI for women <35 years of age was 4.80 (95% confidence interval: 3.55, 6.06). [Table 14.2.1.2.1 run date 16Feb17; Figure 1 run date 31Mar17] When analyzed by BMI, the PI was 4.22 and 5.41 in women with a BMI <26.8 kg/m2 and >26.8 kg/m2, respectively. [Figure 2, run date 31Mar17] Similarly, when analyzed by weight, the PI was 3.83 and 5.84 in women with a weight <71.9 kg and >71.9 kg, respectively. [Figure 3, run date 31Mar17] When analyzed by race, the PIs for subjects identifying as white, black/African American, and “other” were 5.06, 4.52 and 3.64, respectively; whereas when analyzed by ethnicity, the PIs for subjects identifying as Hispanic/Latino and non-Hispanic/Latino were 3.65 and 5.07, respectively. [Figure 4, run date 31Mar17] The most common hormone-related treatment-emergent adverse events were nausea (4.1%), headache (3.6%) and mood swings/changes/depression (2.8%). [51882 ACOG poster ]

Table 1. Mean Number of Bleeding/Spotting Episodes (±SD) and Mean Number of Days Per Episode By Cycle [Table 14.2.7.1.1 run date 23Mar17][Table 14.2.7.2.1 run date 23Mar17][Table 14.2.7.3.1 run date 23Mar17][Table 14.2.7.4.1 run date 23Mar17][Table 14.2.7.5.1 run date 23Mar17][Table 14.2.7.6.1 run date 23Mar17]

An episode of bleeding and/or spotting was defined as one or more consecutive days of bleeding/spotting bounded on either end by > 2 days of no bleeding or spotting. [CSR draft1 24Mar17/p44]