Percent change in BMD (expressed as areal density in grams per centimeter square [g/cm^2]) as specified by dual energy X-ray absorptiometry (DEXA) scans of the left 'total hip' which included the femoral neck, trochanter and inter-trochanter areas was assessed by areal density at Baseline and Week 48. The estimated value in the statistical analysis is this difference and the upper and lower limit values shown are the 95% confidence intervals. Baseline was considered as Day 1 and percent change from Baseline was calculated as Value at Week 48 minus Baseline value divided by Baseline value multiplied by 100. An analysis of covariance (ANCOVA) model was used to compare the difference. The analysis was performed on Intent-to-Treat exposed DEXA (ITT-ED) Population which comprised of all participants in the ITT-E Population who received at least one dose of study treatment, and who were registered for the 202094 study.

ITT-ED Population. Only those participants with data available at specified time point were analyzed.

Percent change in BMD (expressed as areal density in g/cm^2) as specified by DEXA scans of the 'lumbar spine' which included the first lumbar vertebra (L1) to the fourth lumbar vertebra (L4) was assessed by areal density at Baseline and Week 48. The difference is adjusted percent change from Baseline to Week 48 between treatment groups. The estimated value in the statistical analysis is this difference and the upper and lower limit values shown are the 95% confidence intervals. Baseline was considered as Day 1 value and percent change from Baseline was calculated as Value at Week 48 minus Baseline value divided by Baseline value multiplied by 100. An ANCOVA model was used to compare the difference in percentage change from Baseline at week 48 in lumbar spine BMD between the DTG+RPV and CAR arms.

ITT-ED Population. Only those participants with data available at specified time point were analyzed.

Percent change in BMD (expressed as areal density in g/cm^2) as specified by DEXA scans of left 'total hip' which included femoral neck, trochanter and inter-trochanter areas and 'lumbar spine' which included L1 to L4 was assessed by areal density. Percent change from Baseline is post-dose value minus Baseline value divided by Baseline value multiplied by 100. BMD parameters at Weeks 48, 100 and 148 reflect data adjusted following the ongoing longitudinal and cross-calibration of multiple DEXA scanner instruments in this study. Data presented through Week 48 only represent results of Week 48 Primary Endpoint analysis which applied DEXA scanner calibrations though Week 48, with no subsequent calibration applied. In the final analysis conducted at Week 148, DEXA scanner calibration data acquired from Day 1 to Week 148 was applied to all raw DEXA BMD data at Weeks 48, 100 and 148. Hence, actual values of Week 48 DEXA data may vary slightly between Weeks 48 and 148 analyses.

ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X) in category titles.

Percent change in BMD (expressed as areal density in g/cm^2) as specified by DEXA scans of the left 'total hip' which included the femoral neck, trochanter and inter-trochanter areas was assessed by areal density at indicated time points. Percent change in BMD as specified by DEXA scans of the 'lumbar spine' which included the first lumbar vertebra (L1) to the fourth lumbar vertebra (L4) was assessed by areal density at indicated time points. The last pre-switch value (Week 48) was considered as LS Baseline and percent change from LS Baseline was calculated as post-dose visit value minus LS Baseline value divided by LS Baseline value multiplied by 100. The analysis was based on Late-Switch Intent-to-Treat Exposed DEXA (LS-ITT-ED) Population which comprised of all participants in the LS-ITT-E Population, and who were registered for the DEXA study.

LS ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X) in category titles.

Total hip and lumbar spine BMD was assessed by T-scores and Z-scores. Day 1 was considered as Baseline. Change from Baseline was calculated as the value at Week 48 minus Baseline. DEXA scans of the left 'total hip' (femoral neck, hip, inter-trochanter areas, trochanter) and 'lumbar spine' (lumbar vertebral column) were performed. T-score is the number of standard deviations above or below the mean BMD of a 30-year-old participant of the same sex. Caucasian reference values were used for all participants to calculate T-scores. T-score values > -1.0 are considered normal, T-score values <= -1.0 to > -2.5 indicate osteopenia, T-score values <= -2.5 to <-3.5 indicate osteoporosis and T-score values <= -3.5 indicate severe osteoporosis. The Z-score is the number of standard deviations above or below the mean BMD for a reference population of same age and sex and in this study. Caucasian reference values were used in calculation of Z-scores.

ITT-ED Population. Only those participants with data available at specified time point were analyzed.

T-score is the number of standard deviations above or below the mean BMD of a 30-year-old participant of same sex. Caucasian reference values were used to calculate T- and Z- scores. T-score values: > -1.0 is normal; <= -1.0 to > -2.5 indicate osteopenia; <= -2.5 to <-3.5 indicate osteoporosis; <= -3.5 indicate severe osteoporosis. Z-score is the number of standard deviations above or below the mean BMD for a reference population of same age and sex in this study. Change from Baseline is post-dose visit value minus Baseline value. Data for Week 48 only represent final results of Week 48 Primary Endpoint analysis which applied DEXA scanner calibrations through Week 48, with no subsequent calibration applied. In the final analysis conducted at Week 148, DEXA scanner calibration data acquired from Day 1 to Week 148 was applied to all raw DEXA BMD data at Weeks 48, 100 and 148. Hence, actual values of Week 48 DEXA data may vary slightly between Weeks 48 and 148 analyses.

ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X) in category titles.

The last pre-switch value (Week 48) was considered as LS Baseline and change from LS Baseline was calculated as the post-dose visit value minus LS Baseline value. DEXA scans of the left 'total hip' (femoral neck, hip, inter-trochanter areas, trochanter) and 'lumbar spine' (lumbar vertebral column) were performed. T-score is the number of standard deviations above or below the mean BMD of a 30-year-old participant of the same sex. Caucasian reference values were used for all participants to calculate T-scores. T-score values > -1.0 are considered normal, T-score values <= -1.0 to > -2.5 indicate osteopenia, T-score values <= -2.5 to <-3.5 indicate osteoporosis and T-score values <= -3.5 indicate severe osteoporosis. The Z-score is the number of standard deviations above or below the mean BMD for a reference population of same age and sex and in this study. Caucasian reference values were used in calculation of Z-scores.

LS ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X in category titles).

Total hip and lumbar spine BMD (expressed as areal density in g/cm^2) assessed by third agent class (INSTI, NNRTI, PI) at indicated time points. Percent change from Baseline was calculated as value at Week 48 minus Baseline value divided by Baseline value multiplied by 100. Value at Day 1 was considered as Baseline. An ANCOVA model adjusted for Baseline BMD values was used to compare the difference in percent change from Baseline to Week 48 in total hip BMD or in lumbar spine BMD between the DTG+RPV and CAR arms by third agent class: INSTI, NNRTI or PI.

ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X in category titles).

Total hip and lumbar spine BMD was assessed by Baseline third agent class (INSTI, NNRTI, PI) using T-scores and Z-scores at Baseline and Week 48. DEXA scans of hip and spine were performed. Value at Day 1 was considered as Baseline. Change from Baseline was calculated as the value at Week 48 minus Baseline value. T-score is the number of standard deviations above or below the mean BMD of a 30-year-old participant of the same sex. Caucasian reference values were used for all participants to calculate T-scores. T-score values > -1.0 are considered normal, T-score values <= -1.0 to > -2.5 indicate osteopenia, T-score values <= -2.5 to <-3.5 indicate osteoporosis and T-score values <= -3.5 indicate severe osteoporosis. The Z-score is the number of standard deviations above or below the mean BMD for a reference population of same age and sex and in this study. Caucasian reference values were used in calculation of Z-scores.

ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X in category titles).

Total hip and lumbar spine BMD (expressed as areal density in g/cm^2) assessed by third agent class (INSTI, NNRTI, PI) at indicated time points. Percent change from Baseline was calculated as post-dose value minus Baseline value divided by Baseline value multiplied by 100. BMD parameters expressed as areal density (g/cm^2) at Weeks 48, 100 and 148 reflect data adjusted following the ongoing longitudinal and cross-calibration of the multiple DEXA scanner instruments in this study. Data and analyses presented through Week 48 only represent the final results of Week 48 Primary Endpoint analysis which applied DEXA scanner calibrations though Week 48, with no subsequent calibration applied. In the final analysis conducted at Week 148, DEXA scanner calibration data acquired from Day 1 to Week 148 was applied to all raw DEXA BMD data at Weeks 48, 100 and 148. Hence, the actual values of Week 48 DEXA data may vary slightly between the Week 48 and Week 148 analyses.

ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X in category titles).

T-score is the number of standard deviations above or below the mean BMD of a 30-year-old participant of same sex. Caucasian reference values were used to calculate T- and Z-scores. T-score values > -1.0 is normal; <= -1.0 to > -2.5 indicate osteopenia; <= -2.5 to <-3.5 indicate osteoporosis; <= -3.5 indicate severe osteoporosis. Z-score is the number of standard deviations above or below the mean BMD for a reference population of same age and sex in this study. Change from Baseline is the post-dose value minus Baseline value. Data for Week 48 only represents final results of Week 48 Primary Endpoint analysis which applied DEXA scanner calibrations through 48, with no subsequent calibration applied. In the final analysis conducted at Week 148, DEXA scanner calibration data acquired from Day 1 to Week 148 was applied to all raw DEXA BMD data at Weeks 48, 100 and 148. Hence, actual values of Week 48 DEXA data may vary slightly between Weeks 48 and 148 analyses.

ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X in category titles).

Total hip and lumbar spine BMD (expressed as areal density in g/cm^2) assessed by third agent class (INSTI, NNRTI, PI) at indicated time points. The last pre-switch value (Week 48) was considered as LS Baseline and percent change from LS Baseline was calculated as post-dose value minus LS Baseline value divided by LS Baseline value multiplied by 100.

LS ITT-ED Population. Only those participants with data available at specified time point were analyzed (represented by n=X in category titles).

Total hip and lumbar spine BMD was assessed by Baseline third agent (INSTI, NNRTI, PI) using T-scores and Z-scores at indicated time points. DEXA scans of hip and spine were performed. The last pre-switch value (Week 48) was considered as LS Baseline and change from LS Baseline was calculated as the post-dose value minus LS Baseline value. T-score is the number of standard deviations above or below the mean BMD of a 30-year-old participant of the same sex. Caucasian reference values were used for all participants to calculate T-scores. T-score values > -1.0 are considered normal, T-score values <= -1.0 to > -2.5 indicate osteopenia, T-score values <= -2.5 to <-3.5 indicate osteoporosis and T-score values <= -3.5 indicate severe osteoporosis. The Z-score is the number of standard deviations above or below the mean BMD for a reference population of same age and sex and in this study. Caucasian reference values were used in calculation of Z-scores.

LS ITT-ED Population.Only those participants with data available at specified time point were analyzed (represented by n=X in category titles).