Latent tuberculosis treatment

Abbreviations

Abbreviation	Definition
TB	Tuberculosis
LTBI	Latent tuberculosis infection
TPT	Tuberculosis preventive therapy
PLHIV	People living with HIV
U5 HHC	Children under 5 who are exposed to active TB in household
NSP	National Strategic Plan
WHO	World Health Organization

TB is the leading infectious cause of deaths globally, with particularly high disease burden in low to middle-income countries. Treatment for LTBI patients can reduce their risk of progression to (active/symptomatic) TB. WHO guidelines recommend that high-risk populations such as PLHIV and U5 HHC, should receive TPT in order to achieve TB elimination per the WHO End TB strategy. [WHO_High_TB_Burden_Country_Lists] [Stop_TB_Partnership] [Global_Tuberculosis_Programme]

Intervention Overview

CSU client has a long history in TB control and offers programmatic support with partners from the global Stop TB initiative. As part of its commitment to reduce the burden of TB, they developed rifapentine as part of a short-course combination therapy, which may increase adherence, and treatment completion rates compared to alternative long-course LTBI treatment isoniazid. Modeling the scale-up of two treatment options listed below can allow us to compare the impact of these two treatments for LTBI on reducing TB burden.

• Treatment option 1 (6H): daily isoniazid monotherapy administered daily for six months

• Treatment option 2 (3HP): A short-course combination treatment, rifapentine and isoniazid, administered for three months once a week

The table below provides parameters and related data sources for LTBI intervention model.

Treatment parameters

Parameter	Data source	Note
Treatment coverage	NSP and WHO reports
Treatment efficacy	Systematic literature review	Assume same efficacy for 6H and 3HP
Treatment adherence	Systematic literature review

Treatment Coverage Data

We aim to project the scale-up of TPT reaching 95% coverage of two high-risk populations in five high-burden countries in order to access the TB-related health outcomes over a five-year timeframe. We used LTBI treatment coverage target from country-specific NSP to inform the linear projection over 5 years, assuming all countries presented below can reach 95% coverage in 2025.

Baseline coverage data

Location	Baseline treatment	Alternative treatment	U5 HHC coverage 2019 (%)	U5 HHC coverage 2025 (%)	PLHIV coverage 2019 (%)	PLHIV coverage 2025 (%)
South Africa	6H	3HP	69	95	73	95
India	6H	3HP	50	95	50	95
Philippines	6H	3HP	57	95	70	95
Ethiopia	6H	3HP	41	95	67	95
Peru	6H	3HP	54	95	21	95

Vivarium Modeling Strategy

The treatment model links interventions (6H and 3HP) to the likelihood of advancing from LTBI to active TB, thus affecting TB incidence, deaths, and DALYs.

Todo

Add an overview of the Vivarium modeling section.

Intervention Affected Outcomes

Outcome	Outcome type	Outcome ID	Affected measure	Effect size measure	Effect size	Note
TB (all-form)	GBD cause	cause_id 934, 946, 947, 948, 949, 950	incidence rate of developing active TB from LTBI	Relative risk	Detailed in table Effect Size

Restrictions

Restriction	Value	Note
Male only	False
Female only	False
Age group start	0 or 2	0 for 6H treatment population; 2+ for 3HP treatment population.
Age group end	125
Treatment eligibility	Population with LTBI that are considered high-risk for TPT	High-risk population: PLHIV or U5 HHC

Treatment adherence and efficacy

An adherent patient in this simulation is defined as:

• Simulated patient treated with 6H: participants who take drug ≥ 70%-85% of days prescribed.

• Simulated patient treated with 3HP: participants who completed 11 out 12 courses.

\(RR_{no{\_}treatment} = \frac{\text{active TB incidence for no treatment population}}{\text{active TB incidence for adherent population}}\)

\(RR_{non{\_}adherent} = \frac{\text{active TB incidence for non-adherent population}}{\text{active TB incidence for adherent population}}\)

Where,

• \(RR_{no{\_}treatment}\) denotes the risk ratio of developing active TB between a population without treatment and a demographically identical population adherent to per-protocol treatment;

• \(RR_{non{\_}adherent}\) denotes the risk ratio of developing active TB between a population with “non-adherent treatment” (e.g., in a randomized control trial, the intention-to-treat group minus per-protocol group) and a population adherent to per-protocol treatment.

Adherence

Population	Measure	Value	Distribution	Note
Adherent to 6H	6H Adherence rate in percentage points	62 (95%UI 32 - 86)		Derived from observational studies
Adherent to 3HP	3HP Adherence rate in percentage points	77 (95%UI 50 - 94)		Derived from observational studies

Effect Size

Population	Measure	Value	Distribution	Note
No treatment	Relative risk of incidence rate ratio between no treatment population and adherent population	3.3 (95%UI 1.6 - 8.7)		RR is applicable for both 6H and 3HP
Non-adherent to treatment	Relative risk of incidence rate ratio between non-adherent population and adherent population	1.5 (95%UI 1.2 - 1.8)		RR is applicable for both 6H and 3HP

As defined in table above, compared to someone who is adherent to the treatment, someone who receives treatment and is not adherent is 1.5 times more likely to transition from LTBI to active TB. Compared to someone who is adherent to treatment, someone who does not receive treatment is 3.3 times more likely to advance to active TB.

Treatment Algorithm

To estimate the effect of different treatment scenarios, each individual patient in the simulation is assigned a treatment state indicating their current treatment (e.g., treated with 6H, treated with 3HP, not treated). Each patient’s treatment state changes over time, depending on adherence and duration of treatment regimen. The TPT coverage parameter defines how likely eligible individuals (LTBI patients with HIV or under 5 exposed to active TB in household) are to be treated. Once a patient is initiated, the treatment pathway is defined based on adherence and efficacy. Simulants that receive treatment are assigned a reduced risk of progressing to active TB for the duration of the simulation depending on the treatment (6H or 3HP) and whether or not the simulated individual is adherent. The simulation assumes no one has had TPT prior to the day 1 of the simulation, and no retreatment for those that have already received treatment, regardless of adherence. The adherence and efficacy parameter values are detailed in Treatment adherence and efficacy section. Based on the available evidence and guidelines, which do not recommend rifapentine nor 3HP for children under two, [Global_Tuberculosis_Programme] [LTBI_Treatment_Guidelines] [Recommendations-for-Use-of-3HP-Regimen] [LTBI-Treatment-FAQs] eligible children under 2 are treated with 6H instead.

Variables in Treatment Algorithm

Variable	Choice	Measure	Data source	Note
with_ltbi	True, False	Prevalence and incidence of LTBI	GBD
treatment_status	Treated with 6H, Treated with 3HP, Not treated	Treatment coverage	NSP and WHO reports	1) No one has received treatment on day 1 of the simulation; 2) no retreatment for those who have received treatment during the simulation period.
u5_hhc_status	Exposed, Unexposed	Demographic age structure, Prevalence of active TB in household	GBD, Household survey data (e.g., DHS, IPUMS)
hiv_status	Positive, Negative	Prevalence of HIV	GBD
adherence_status	Adherent, Non-adherent	Treatment completion rate	Systematic literature review	See values in Table Adherence
treatment_outcome (no treatment population)	Advance to active TB, Stay in LTBI	Incidence of active TB, Treatment efficacy derived from active TB incidence ratio between no treatment population and adherent population	GBD, Systematic literature review	See \(RR_{no{\_}treatment}\) in Table Effect Size
treatment_outcome (non-adherent population)	Advance to active TB, Stay in LTBI	Incidence of active TB, Treatment efficacy derived from active TB incidence ratio between non-adherent population and adherent population	GBD, Systematic literature review	See \(RR_{non{\_}adherent}\) in Table Effect Size
treatment_outcome (adherent population)	Advance to active TB, Stay in LTBI	Incidence of active TB	GBD	Reference treatment group, \(RR_{adherent} = 1\)

Assumptions and Limitations

1. In this simulation, patients are not eligible for retreatment after receiving TPT, regardless of whether the simulated patient completes the full course of treatment or not.

2. We assume same efficacy for 6H and 3HP because there are literature evidence demonstrated that 6HP is non-inferior to 3HP on protecting patients progress from LTBI to active TB. [citation]

3. 3HP is not suitable for children under two. Instead, we treat them with 6H.

4. We assume on one has had TPT prior to the beginning of the simulation.

5. For LTBI patients, only high-risk individuals (PLHIV and U5 HHC) are considered eligible for TPT.

Validation and Verification Criteria

References

[WHO_High_TB_Burden_Country_Lists]

WHO High TB Burden Country Lists 2016-2020. https://www.who.int/tb/publications/global_report/high_tb_burdencountrylists2016-2020.pdf (accessed Jan 15, 2020)

[Stop_TB_Partnership]

Stop TB Partnership | High Burden Countries. http://www.stoptb.org/countries/tbdata.asp (accessed Jan 15, 2020)

[Global_Tuberculosis_Programme] (1,2)

Latent tuberculosis infection: updated and consolidated guidelines for programmatic management. 2018

[LTBI_Treatment_Guidelines]

Sterling TR. Guidelines for the Treatment of Latent Tuberculosis Infection: Recommendations from the National Tuberculosis Controllers Association and CDC, 2020. MMWR Recomm Rep 2020; 69. DOI:10.15585/mmwr.rr6901a1.

[Recommendations-for-Use-of-3HP-Regimen]

Borisov AS. Update of Recommendations for Use of Once-Weekly Isoniazid-Rifapentine Regimen to Treat Latent Mycobacterium tuberculosis Infection. MMWR Morb Mortal Wkly Rep 2018; 67. DOI:10.15585/mmwr.mm6725a5.

[LTBI-Treatment-FAQs]

Latent TB Infection Treatment FAQs for Clinicians | Tools for Health Care Providers | Professional Resources & Tools | TB | CDC. 2020; published online Feb 19. https://www.cdc.gov/tb/education/FAQforProviders.htm (accessed Feb 25, 2020)