Baseline and end of treatment values are the calculated means of the daily 24-hour API scores for each participant during the last 7 days prior to randomization (baseline) and the last 7 days on treatment within each period (end of treatment). Participants rated their API over the preceding 24 hours, using an 11-point PI-Numerical Rating Scale (0=no pain, 10=pain as bad as you can imagine). LOCF was used if less than 4 days of diary data were provided. Change from baseline was calculated as end of treatment minus baseline. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for 24-hour API assessments during the GEn 3600mg treatment period, and was therefore not included in this analysis.

Baseline and end of treatment values are the calculated means of the daily 24-hour API scores for each participant during the last 7 days prior to randomization (baseline) and the last 7 days on treatment within each period (end of treatment). Participants used a hand-held diary to rate their average pain intensity over the preceding 24 hours, using an 11-point PI-NRS (0=no pain, 10=pain as bad as you can imagine). LOCF was used if less than 4 days of diary data were provided. The by period summary is provided as a sensitivity analysis for the primary analysis.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for 24 hour API while taking GEn 3600 mg in the first period, and was therefore not included in this analysis.

Day-time is defined as the time between rising in the morning and going to bed at night. Participants recorded day-time API on a daily basis in the evening before bedtime using an 11-point PI-NRS (0=no pain, 10=pain as bad as you can imagine). Baseline and end of treatment scores are as defined for the primary endpoint. Change from baseline is calculated as the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for day-time pain assessments during the GEn 3600 mg treatment period, and was therefore not included in this analysis.

Day-time worst pain is defined as the participant's assessment of their worst pain intensity between rising in the morning and going to bed at night. Day-time worst pain was recorded in the evening before bedtime using an 11-point PI-NRS (0=no pain, 10=pain as bad as you can imagine). Baseline and end of treatment scores are as defined for the primary endpoint. Change from baseline is calculated as the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for day-time pain assessments during the GEn 3600 mg treatment period, and was therefore not included in this analysis.

Current pain is defined as the participant's assessment of pain intensity "right now." Participants recorded their current evening pain intensity in the evening before bedtime using an 11-point PI-NRS (0=no pain, 10=pain as bad as you can imagine). Baseline and end of treatment scores are as defined for the primary endpoint. Change from baseline is calculated as the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for current evening pain during the GEn 3600 mg treatment period, and was therefore not included in this analysis.

Night-time is defined as the time between going to bed in the evening and rising in the morning. Participants recorded night-time API on a daily basis in the morning upon wakening using an 11-point PI-NRS (0=no pain, 10=pain as bad as you can imagine). Baseline and end of treatment scores are as defined for the primary endpoint. Change from baseline is calculated as the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for night-time pain assessments during the GEn 1200 mg treatment period, and was therefore not included in this analysis.

Night-time worst pain is defined as the participant's assessment of their worst pain intensity between going to bed and rising in the morning. Participants recorded night-time worst pain in the morning upon wakening using an 11-point PI-NRS (0=no pain, 10=pain as bad as you can imagine). Baseline and end of treatment scores are as defined for primary endpoint. Change from baseline = the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for night-time pain assessments during the GEn 1200 mg treatment period, and was therefore not included in this analysis.

Current pain is defined as the participant's assessment of pain intensity "right now." Participants recorded their current morning pain intensity in the morning upon wakening using an 11-point PI-NRS (0=no pain, 10=pain as bad as you can imagine). Baseline and end of treatment scores are as defined for the primary endpoint. Change from baseline is calculated as the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for current morning pain during the GEn 1200 mg treatment period, and was therefore not included in this analysis.

Baseline and end of treatment (EOT) scores are the calculated means of the 24-hour average pain scores for each participant during the last 7 days prior to randomization (Baseline) and the 7 days prior to the last on-treatment completed diary (EOT). Percent reduction from baseline was calculated as the [(EOT score minus baseline score) divided by the baseline score], multiplied by 100. The PI-NRS is an 11-point scale (0=no pain, 10=pain as bad as you can imagine) by which a participant assesses their 24-hour average pain intensity. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, there was one participant who did not provide post-baseline data for 24-hour API assessments during the GEn 3600 mg treatment period, and was therefore not included in this analysis.

Baseline and end of treatment scores are the calculated means of the 24-hour average pain scores for each participant during the last 7 days prior to randomization (Baseline) and the 7 days prior to the last on-treatment completed diary (end of treatment). Percent reduction from baseline was calculated as the [(end of treatment score minus the baseline score) divided by the baseline score], multiplied by 100. The PI-NRS is an 11-point scale (0=no pain, 10=pain as bad as you can imagine) by which a participant assesses their 24-hour average pain intensity. Data are summarized by period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, there was one participant who did not provide post-baseline data for 24-hour API while taking GEn 3600 mg in the first period, and was therefore not included in this analysis.

Mean daily use of rescue medication (milligrams of acetaminophen) was calculated by determining the average number of tablets taken per day of rescue medication (Commercial Tylenol) during treatment and multiplying that by 500 mg. Baseline and end of treatment scores are as defined for the primary endpoint. Change from baseline is calculated as the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, there was one participant who did not provide post-baseline data for rescue medication usage during the GEn 3600 mg treatment period, and was therefore not included in this analysis.

The PGIC is a single-item questionnaire designed to provide an overall assessment of treatment from the participant's perspective since the start of the study. It is measured on a 7-point scale, where 1=very much improved and 7=very much worse. A participant is considered a responder if they have a response of "very much improved" or "much improved" Data are summarized by dose, independent of treatment period.

ITT Population. There were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. There were many participants who did not respond to the questionnaire and thus could not be included in the analysis.

The PGIC is a single-item questionnaire designed to provide an overall assessment of treatment from the participant's perspective since the start of the study. It is measured on a 7-point scale, where 1=very much improved and 7=very much worse. A participant is considered a responder if they have a response of "very much improved" or "much improved". Data are summarized by dose within each treatment period.

ITT Population. There were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. There were many participants who did not respond to the questionnaire and thus could not be included in the analysis.

The CGIC is a single-item questionnaire designed to provide an overall assessment of treatment from the clinician's perspective since the start of the study. It is measured on a 7-point scale, where 1=very much improved and 7=very much worse. A participant is considered a responder if they have a response of "very much improved" or "much improved." Data are summarized by dose, independent of treatment period.

ITT Population. There were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. There were many participants without a response to this questionnaire and thus could not be included in the analysis.

The CGIC is a single-item questionnaire designed to provide an overall assessment of treatment from the clinician's perspective since the start of the study. It is measured on a 7-point scale, where 1=very much improved and 7=very much worse. A participant is considered a responder if they have a response of "very much improved" or "much improved." Data are summarized by dose within each treatment period.

ITT Population. There were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. There were many participants without a response data to this questionnaire and could thus not be included in the analysis.

Participants assessed sleep interference due to pain on a daily basis using the 11-point NRS (0=pain does not interfere with sleep, 10=pain completely interferes with sleep). Baseline and end of treatment scores are as defined for the primary endpoint. Change from baseline is calculated as the end of treatment score minus the baseline score. An ANCOVA with baseline value, BMI, grouped center as covariates was used. Data are summarized by dose, independent of treatment period.

Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for sleep interference during the GEn 1200 mg treatment period, and was therefore not included in this analysis.

The BPI assesses the severity and interference of pain; and consists of 6 items assessed on an 11-point NRS (0=no impact to 10=greatest impact). 2 summary scores are calculated: BPI Severity Score (average of first 4 items) and BPI Interference Score (average of 7 responses to item 6); where scores range from 0 to 10 (0=no impact to 10=greatest impact). Analysis of this endpoint is based on the change from baseline (BL) (EOMT score minus the BL score) using an ANCOVA model with BL value, BMI, grouped center as covariates. Data are summarized by dose, independent of treatment period.

ITT Population. There were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. There were many participants who did not respond to the questionnaire and could thus not be included in the analysis.

Steady-state average (Cave, ss), maximum (Cmax, ss), and minimum (Cmin,ss) plasma concentration of gabapentin in each participant were estimated using the gabapentin plasma concentration data and with the aid of a population pharmacokinetic model. Dispersion is represented by the fifth to ninety-fifth percentile, though labeled as "Full Range." A total of 10 blood samples were collected per participant over the Baseline, Period 1, and Period 2 at various timepoints during the dosing interval. Plasma concentration of gabapentin in these samples was measured.

Drug concentration data were available from 89 ITT Population participants. Data from 7 of these participants had one concentration with less than half of the first percentile of the concentrations observed at ss and were defined as non-compliant and were excluded from the pharmacokinetic (PK) analysis.