Time to First-Line Antiretroviral Therapy Failure and Its Predictors among People Living with HIV in Tanzania

This retrospective cohort study of over 36,000 Tanzanian patients reveals that while dolutegravir-based first-line antiretroviral therapy initially offers strong protection against treatment failure compared to non-dolutegravir regimens, this protective effect diminishes over time, underscoring the critical need for sustained virological monitoring and highlighting age and gender as significant predictors of failure.

The Gist

Imagine the fight against HIV as a long-distance marathon. For people living with HIV, taking daily medication (Antiretroviral Therapy, or ART) is like running that race. The goal is to keep the virus "suppressed" (asleep) so the runner can stay healthy for decades. But sometimes, the virus wakes up, and the medication stops working. This is called treatment failure.

This paper is like a massive race report from Tanzania, looking at thousands of runners to see how long they stayed on track and what caused some to stumble.

Here is the breakdown of the study in simple terms:

1. The Big Picture: A National Race

The researchers looked at data from 36,764 people in Tanzania who started HIV treatment between 2017 and 2021. They wanted to answer two questions:

• How long does the medicine last before it stops working?
• What factors make it more likely for the medicine to fail?

They used a special statistical tool (like a high-tech stopwatch) to track exactly when the failure happened, rather than just taking a snapshot of people at one specific time.

2. The Different "Shoes" (Medication Regimens)

In a marathon, runners wear different shoes. In this study, the "shoes" were different types of HIV medication. The main contenders were:

• The Old Guard (NVP): Nevirapine-based drugs.
• The Standard (EFV): Efavirenz-based drugs.
• The New Champion (DTG): Dolutegravir-based drugs.
• The Heavy Duty (PI): Protease Inhibitors.

3. The Key Discovery: The "Fresh Start" Effect

The most exciting finding was about the New Champion (DTG).

Think of DTG like a brand-new, high-tech running shoe.

• The First 6 Months: When people first put on these new shoes, they were incredibly fast and stable. The risk of the virus waking up was 63% lower compared to the older drugs. It was like having a superpower at the start of the race.
• The Middle (12 Months): The shoes were still great, but the superpower started to fade a little. The protection was still there, but not as strong as day one.
• The Long Haul (24 Months): Here is the twist. After two years, the "superpower" of the new shoes seemed to disappear. The protection wasn't any better than the older drugs anymore.

The Lesson: The new drug is amazing at getting you started, but it doesn't guarantee you'll win the whole race on its own. You still have to keep running (taking your meds) perfectly.

4. Who is Most Likely to Trip?

The study also looked at the runners themselves to see who was at higher risk of stumbling:

• Age: Younger runners were slightly more likely to fail than older ones. (Perhaps younger people find it harder to stick to a strict daily routine).
• Gender: Women were slightly more likely to experience failure than men.
• The "Old Shoes" (NVP): People still wearing the old Nevirapine shoes had a much higher chance of stumbling compared to those on the standard Efavirenz drugs.

5. Why Did the "Superpower" Fade?

The researchers suggest that while the new drug (DTG) is very strong, it isn't magic. Over time, if a person misses doses or if the virus starts to mutate (change its shape to hide from the drug), the protection wears off.

It's like having a very strong lock on your front door. It keeps burglars out for the first year. But if you leave the door open sometimes, or if the burglar learns how to pick the lock, eventually, the lock might not be enough.

6. The Takeaway for the Real World

This study tells us three important things for Tanzania and similar places:

1. Keep using the new drugs (DTG): They are the best choice to start with because they work so well initially.
2. Don't get complacent: Just because the new drug works great at first doesn't mean you can stop paying attention. You have to keep taking your meds every single day.
3. Keep checking the pulse: Doctors need to keep testing patients' blood (viral load) regularly, especially after the first year, to catch any "stumbles" early before the virus takes over completely.

In short: The new medicine is a fantastic head start, but winning the marathon requires consistent effort and regular check-ins for the entire journey, not just the beginning.

Technical Summary

1. Problem Statement

Despite widespread access to antiretroviral therapy (ART) in sub-Saharan Africa, sustaining long-term viral suppression remains a critical public health challenge. While previous studies have identified predictors of treatment failure, there is a significant gap in nationally representative, longitudinal data regarding the time-to-event (durability) of first-line ART, particularly for newer dolutegravir (DTG)-based regimens. Most existing evidence relies on cross-sectional data, pediatric cohorts, or second-line therapy, failing to capture the evolving risk of virological failure over time in adult populations within Tanzania.

2. Methodology

• Study Design: Retrospective cohort study utilizing a time-to-event analytical framework.
• Data Source: Routinely collected national programmatic data from the National Care and Treatment Clinic (CTC-2) database in Tanzania, managed by the National AIDS, Sexually Transmitted Infections and Hepatitis Control Programme (NASHCoP).
• Study Population:
  • Inclusion: People Living with HIV (PLHIV) aged ≥11 years who initiated first-line ART between January 2017 and December 2021.
  • Exclusion: Patients with missing baseline data, those <11 years, transfers without records, or those lost to follow-up/died before completing 6 months of therapy.
  • Sample Size: A simple random sample of ~50,000 was drawn; the final analytic cohort comprised 36,764 individuals.
• Outcome Definition:
  • Primary Outcome: Time to first-line ART failure.
  • Failure Definition: First documented virological failure defined as a viral load ≥1,000 copies/mL occurring at least 6 months after ART initiation.
• Statistical Analysis:
  • Descriptive: Kaplan-Meier methods to estimate failure-free survival; log-rank tests for group comparisons.
  • Multivariable Modeling: Cox proportional hazards models to identify predictors (age, gender, regimen).
  • Advanced Modeling: An extended Cox model with a time-varying coefficient was employed to address non-proportional hazards specifically observed in DTG-based regimens. This allowed for the estimation of hazard ratios (HR) at specific time points (6, 12, and 24 months).
  • Stratification: Models were stratified by geographic region to account for heterogeneity.

3. Key Contributions

• National Scale Analysis: This is one of the first studies to utilize the entire national CTC-2 database to evaluate first-line ART durability across Tanzania, providing a robust, population-level perspective.
• Time-Varying Effect of DTG: The study moves beyond static comparisons by modeling the temporal dynamics of DTG effectiveness. It demonstrates that the protective effect of DTG is not constant but changes significantly over the duration of treatment.
• Methodological Rigor: By addressing non-proportional hazards through time-varying coefficients, the study offers a more accurate representation of regimen performance than standard Cox models, which assume constant hazard ratios over time.
• Real-World Evidence: The use of routine programmatic data provides "real-world" evidence of ART outcomes, complementing clinical trial data.

4. Key Results

• Cohort Characteristics: The cohort included 36,764 individuals with 789 failure events. The majority were on Efavirenz (EFV) regimens (26,061), followed by DTG (7,180), Nevirapine (NVP) (3,041), and Protease Inhibitors (PI) (482).
• Baseline Predictors:
  • Gender: Male gender was associated with a significantly lower hazard of failure compared to females (AHR 0.46).
  • Age: Increasing age at ART initiation was associated with a modest reduction in failure hazard.
  • Regimen (Static Model): In baseline models, NVP-based regimens showed a significantly higher risk of failure compared to EFV (AHR 1.66). DTG showed a lower risk compared to non-DTG regimens (AHR 0.75), though this was not statistically significant when compared strictly to EFV in the baseline model.
• Time-Varying DTG Effect (Critical Finding):
  The study revealed a distinct temporal pattern for DTG-based regimens compared to non-DTG regimens:
  • 6 Months: Strong protective effect (HR 0.37; 95% CI 0.22–0.60).
  • 12 Months: Protective effect attenuated but remained significant (HR 0.67; 95% CI 0.50–0.90).
  • 24 Months: The protective effect disappeared, with the hazard ratio exceeding unity (HR 1.22; 95% CI 0.88–1.70), indicating no significant difference in failure risk compared to non-DTG regimens at this later stage.
• Follow-up Disparities: Median follow-up time varied significantly by regimen, with DTG patients having the shortest follow-up (442 days) compared to NVP/PI patients (~1,700 days), reflecting the timeline of national guideline transitions.

5. Significance and Implications

• Clinical & Programmatic Strategy: The findings validate the initial switch to DTG-based regimens due to their superior early protection against virological failure. However, the attenuation of this benefit over time highlights that early success does not guarantee long-term durability.
• Monitoring Recommendations: The study underscores the critical need for sustained, long-term virological monitoring, particularly for patients on DTG beyond the first year. Relying solely on early success metrics may mask emerging resistance or adherence issues.
• Targeted Interventions: The higher failure hazard observed in females and younger patients suggests a need for differentiated care strategies and enhanced adherence support for these specific demographics.
• Policy Impact: The results support the continued prioritization of DTG but call for a shift from cross-sectional performance metrics to longitudinal survival analysis in national HIV program monitoring to better detect delayed treatment failures.

Limitations Noted: The study relied on routine data with missing clinical covariates (e.g., baseline CD4, WHO stage), potential misclassification of transient viremia as failure, and the exclusion of patients lost to follow-up before 6 months, which may slightly underestimate early failure rates.