Abstract
Cryptococcal antigen (CrAg) screening is recommended for patients with advanced HIV to reduce AIDS-related mortality. For asymptomatic CrAg-positive persons, fluconazole pre-emptive therapy is standard, despite a ∼25% failure rate. Single-dose liposomal amphotericin B (AmBisome) is non-inferior to standard treatment for cryptococcal meningitis. We evaluate the threshold of efficacy necessary for AmBisome + fluconazole to be cost-effective as pre-emptive therapy for CrAg-positive persons.
We created a decision analytic model to evaluate CrAg screening and treatment in HIV-infected persons with CD4 < 100 cells/μL. Costs were estimated for screening, pre-emptive therapy, and hospitalization for an example low-income country (Uganda) and middle-income country (South Africa). We used a discounted price range of AmBisome® at 
16.25 to 
40 per 50 mg vial for both Uganda and South Africa. We estimated AmBisome efficacy from 75 to 95%. Parameter assumptions were based on prospective CrAg screening studies and clinical trials in Africa. Disability adjusted life years (DALYs) were calculated using the age-specific life expectancy in Uganda, per WHO Global Health Observatory data. We modeled the theoretical efficacy of adjunctive AmBisome to determine cost per DALY averted.
In South Africa, at 
16.25 per vial cost and a minimum efficacy of 85%, adjunctive AmBisome is cost-saving compared to fluconazole monotherapy. Compared to fluconazole pre-emptive therapy in Uganda, AmBisome + fluconazole would cost 
475, 
220, or 
136 per DALY averted if meningitis-free survival efficacy was 80, 85, or 90% at 
24 per vial cost.
Investing in AmBisome may be cost-effective in low-income settings compared to using fluconazole pre-emptive therapy alone, if efficacy is 85% or greater. AmBisome pre-emptive therapy appears more cost-efficient in middle-income settings where hospitalization costs for meningitis, and GDP per capita are higher.
Lay Summary
We evaluate the efficacy necessary for AmBisome + fluconazole to be cost-effective to prevent cryptococcal meningitis. We found that if AmBisome pre-emptive therapy has an efficacy of 85% or greater, it is likely to be cost-effective in low-income settings.
Introduction
The Joint United Nations Programme on HIV/AIDS (UNAIDS) has a vision of zero AIDS-related deaths by 2030.1 In 2020, an estimated 680 000 people died from AIDS-related illnesses.2 Cryptococcal meningitis remains a leading cause of death in persons with advanced HIV infection, accounting for 15% of AIDS-related mortality.3 In sub Saharan Africa, cryptococcal meningitis mortality is high at 50 to 70% due to delays in presentation to care, the need for complex management requiring serial lumbar punctures, and limited access to optimal antifungal medications.4–6 However, cryptococcal antigen (CrAg) can be detected in the blood weeks before onset of symptoms of meningitis, and CrAg positivity in the blood is an independent predictor of future meningitis or death.7–9 Screening for CrAg in persons with advanced HIV, and pre-emptively treating those CrAg-positive with fluconazole has been evaluated in a randomized controlled trial in Zambia and Tanzania, and, alongside adherence counseling, demonstrated a 28% relative mortality reduction.10 CrAg screening and pre-emptive treatment for patients with advanced HIV is now recommended by the World Health Organization (WHO) and numerous national HIV programs.11,12
However, despite the survival benefit with fluconazole pre-emptive therapy, 25% of asymptomatic CrAg-positive persons still develop meningitis and/or die.13,14 The majority of deaths of CrAg-positive persons are from cryptococcal infection.14 As fluconazole monotherapy is inadequate, enhanced antifungal therapy has the potential to improve survival by aggressively treating subclinical, disseminated infection. Single dose liposomal amphotericin B (AmBisome®) at 10 mg/kg dose has been demonstrated to be safe and effective for treatment of visceral leishmaniasis.15 A phase II study found a single-dose AmBisome was non-inferior to a longer 14-day regimen of amphotericin B deoxycholate in the treatment of cryptococcal meningitis, and recent results presented at the International AIDS Society conference in July 2021 demonstrate that one high-dose of AmBisome is non-inferior to standard induction treatment for HIV-associated cryptococcal meningitis.16 High dose AmBisome (in conjunction with fluconazole) is also currently being evaluated for pre-emptive treatment in asymptomatic cryptococcal infection (Clinicaltrials.gov: NCT03945448). Thus, there now exists a medicine with known safety, potentially efficacious via a single dose, and is newly available, where the burden of cryptococcosis is the highest.
There is an urgent need for more potent pre-emptive antifungal therapy for CrAg-positive persons, as 25% of CrAg + develop breakthrough cryptococcal meningitis and/or die with the current standard of care fluconazole.10,13 Amphotericin B is the recommended therapy against Cryptococcus neoformans.17 However the traditional form of amphotericin B has toxic side effects including anemia and kidney dysfunction. Thus IV fluids are required, along with close monitoring of electrolytes.17 Liposomal amphotericin (AmBisome) has similar efficacy but with better tolerability and less severe side effects. The biggest barrier to AmBisome access in low and middle income countries (where the burden of cryptococcal infection is the highest) is its high cost.18,19
Cost-effectiveness studies are a useful tool in the assessment of health care interventions and recommendations as they offer a logical way of informing decisions relating to uptake of new health care strategies compared to available standards. Based on a detailed analysis of Ugandan and South African costs for CrAg screening and meningitis treatment from the perspective of their respective Ministries of Health, the objective of this study is to evaluate the costs and mortality reductions (lives saved) with varying efficacy of AmBisome + fluconazole pre-emptive therapy for cryptococcal antigenemia in low- and middle- income countries. We also explored how the basic model results could change with alternative meningitis treatment regimens.
Methods
A decision analytic model was developed to evaluate CrAg screening and treatment as two separate stages: 1) screening for CrAg; and 2) treatment for CrAg-positive persons, which includes pre-emptive treatment of those asymptomatic CrAg-positive with AmBisome + fluconazole, as well as hospitalization and treatment for meningitis for those symptomatic CrAg-positive, and those missed by the screening program (Supplementary Figure). The full details of this model had been published previously,20 but is briefly summarized as below. The main change from the previous model is the inclusion of costs of adjunctive AmBisome for pre-emptive therapy. Uganda and South Africa were chosen to represent low-income countries and middle-income countries in the region, respectively.
CrAg screening for those with CD4 ≤100 cells/μl
CrAg screening identified CrAg-positive individuals in the HIV-infected population with a CD4 cell count ≤100 cells/μl.20 Table 1 describes all input parameters and sources of data for CrAg screening population. The model results presented as the ‘base case’ use the parameter assumptions in Table 1.
Table 1.
Transition probabilities used in the CrAg screening and treatment models in Uganda and South Africa.
| Population screened | Probability in Ugandan model | Probability in South African model | Source |
|---|---|---|---|
| CD4 count ≤100 cells/μl | 0.16 | 0.093 | 34,35 |
| CD4 ≤100 cells/μl CrAg screened | 1 | 1 | Assumption |
| Return to clinic for CrAg results | 1 | 1 | Assumption |
| Asymptomatic CrAg + receives pre-emptive treatment | 1 | 1 | Assumption |
| CrAg prevalence | |||
| CrAg negative | 0.9075 | 0.9481 | 13 |
| Prior CCM | 0.0019 | 0.0019 | 13 |
| Incident CrAg + | 0.0906 | 0.05 | 13 |
| Symptomatic CrAg+ | 0.0102 | 0.01 | 13 |
| Asymptomatic CrAg+ | 0.0804 | 0.04 | 13 |
| Of incident asymptomatic CrAg+ | |||
| High titer ≥ 1:160 | 0.39 | 0.39 | 13 |
| Low titer < 1:160 | 0.61 | 0.61 | 13 |
| CrAg + outcomes | |||
| CrAg + high titer, no pre-emptive treatment or ART, develops CM | 1.0 | 1.0 | 19 |
| CrAg + low titer, no pre-emptive treatment or ART, subsequently develops CM | 0.50 | 0.50 | 8,9 |
| CrAg + symptomatic presents to hospital | 0.73 | 0.73 | 36 |
| CrAg + fails fluconazole, presents to hospital | 0.80 | 0.80 | Assumption |
| Symptomatic Meningitis outcomes | |||
| CM who present to hospital | 0.80 | 0.80 | Assumption |
| 2-week survival with Amphotericin + fluconazole (14 days) | 0.82 | 0.82 | ACTA30 |
| 10-week survival with Amphotericin + fluconazole (14 days) | 0.57 | 0.57 | 30 |
| CM = cryptococcal meningitis | |||
For those with a CD4 count ≤100 cells/μl, we assumed 100% were CrAg screened and 100% of those tested returned for their CD4 and CrAg results (CrAg negative or CrAg positive) (Supplementary Figure). CrAg positive persons were disaggregated into three categories: (1) asymptomatic and thus eligible for pre-emptive treatment with AmBisome + fluconazole; (2) symptomatic with meningitis who should be hospitalized; and (3) CrAg-positive due to a history of cryptococcal infection. Those with a history of cryptococcal meningitis remain CrAg positive for years, and this lab result was considered an artifact. CrAg testing was presumed to occur using the lateral flow assay (LFA) (Immy, Norman OK), which has > 99% sensitivity and specificity.21
Pre-emptive treatment for asymptomatic CrAg positive persons
For asymptomatic CrAg positive persons, we assumed that all who returned for results were eligible for pre-emptive treatment (Supplemental Figure). Pre-emptive treatment was one dose of AmBisome at 10 mg/kg + fluconazole. We assumed an average weight of 50 kg per patient. Thus, 500 mg of AmBisome was given, along with fluconazole 800 mg daily for 2 weeks, followed by 400 mg daily for 8 weeks, followed by 200 mg daily for 6 months, per WHO guidelines.11 The efficacy of this AmBisome regimen is unknown, thus we modeled efficacy (defined as survival without development of meningitis) from a range of 75 to 95% to evaluate costs and benefit. The lower range of 75% was chosen because fluconazole has ∼75% meningitis-free survival,10 thus anything lower than 75% efficacy would be irrelevant compared with current standard of care.
Treatment for cryptococcal meningitis
In the base case model, those hospitalized for cryptococcal meningitis were treated with amphotericin + fluconazole for 14 days, per the standard of care in Uganda and South Africa.
Health outcomes
Patients were assumed to have three primary outcomes: survived, died, or unknown. The unknowns were a small number of persons who developed meningitis, survived hospitalization but then developed recurrent meningitis during their post-hospital maintenance treatment phase. These patients either died without returning to a hospital or presented back to a hospital for further care.
Deaths avoided (lives saved) was used as the main health outcome, along with disability adjusted life years (DALYs). Disability was assumed to be insignificant in persons with asymptomatic cryptococcal antigenemia; by definition they are asymptomatic. Among those with symptomatic cryptococcal meningitis, most survivors have near baseline neurologic function after 12 months.22 Thus, there is not significant disability associated with this infection, rather high mortality. As a result, our determination of DALYS was predominantly focused on deaths avoided. We assumed the average age of those presenting to care was 32 based on the average age screened for CrAg in a recent Ugandan trial,13 and the average life expectancy for this age group was an additional 31 years.23 With a 3% discount rate, and 31 years of life lost from a death, 20.6 DALYs were lost per death (all from years of life lost), so 20.6 DALYs were avoided per death prevented.
Ugandan screening and treatment costs
For the base case results, the screening and treatment costs were based on the information in Table 2. All costs were reported in 2017 US dollars and assumed to have been borne fully by the Ministry of Health. The 2017 annual average exchange rate was 3 611 Ugandan shillings (UGX) for 1 US dollar.24 CrAg testing was estimated at 
3.41 per test, based on the cost of the lateral flow assay, import costs per the manufacturer, shipping, and labor.
Table 2.
Input costs for CrAg screening and treatment.
| Cost in Uganda (USD) | Cost in South Africa (USD) | Notes | |
|---|---|---|---|
| CrAg test cost |
3.41 |
4.28 |
Test, plus import, shipping, and labor20,37 |
| Fluconazole 200 mg tablet |
0.14 |
0.07 |
25 |
| Total fluconazole course |
39.06 |
19.36 |
Including 6 months maintenance on 200 mg daily |
| AmBisome 500 mg |
162.50 to  400 |
162.50 to  400 |
10 vials for a 50 kg patient |
| Total pre-emptive treatment course: AmBisome + Fluconazole |
444.04 to  844.04 |
469.36 to 869.36 |
Includes IV fluids, nursing, and administrative fee for infusion |
| Hospitalization | |||
| Amphotericin 50 mg per day × 14 days |
152.46 |
152.46 |
25 |
| Hospitalization total |
535.80 |
3144.00 |
20,26 |
| Post hospitalization consolidation and maintenance with fluconazole |
66.78 |
33.39 |
1 year |
Fluconazole costs of 
0.14 per 200 mg tablet were based on information from the Joint Medical Store (JMS), a major Ugandan national supplier.25 The cost of a 50 mg vial of AmBisome was ranged from 
16.25 to 
40. This range reflects the price to approved logistic suppliers for the treatment of visceral leishmaniasis and cryptococcal meningitis. Thus assuming a single dose of 10 mg/kg of AmBisome for a 50 kg patient in Uganda, the cost would range from 
162.50 to 
400. Additional fluconazole pre-emptive therapy costed 
39.90. We also included a 
5 administrative fee which included 1 liter of intravenous fluids given prior to AmBisome, and nursing care for 2 h to administer the AmBisome.
Hospitalization costs for cryptococcal meningitis have been described in detail previously.20 Including post-hospitalization consolidation and maintenance fluconazole therapy for 1 year, the total cost of hospitalization and therapy for a person with cryptococcal meningitis in Uganda was estimated at 
600 US dollars.
South African screening and treatment costs
Screening and treatment costs for South Africa were based on the parameters highlighted in Table 2. Costs were reported in 2017 US dollars, and were assumed to have been borne by the Ministry of Health. The 2017 annual average exchange rate was 13.33 Rand to 1 US dollars. The cost of a 50 mg vial of AmBisome was the same as in Uganda (ranged from 
16.25 to 
40 per vial), and the cost of fluconazole was 
0.07 per 200 mg tablet.25 Ultimately the total cost of pre-emptive AmBisome + fluconazole ranged from 
231 to 
469 per vial including a 
50 administrative fee including nursing care during the infusion, and IV fluids.
Hospitalization costs for meningitis in South Africa was taken from a publication, that evaluated both urban and rural hospitalization costs in South Africa for HIV-infected persons, and stratified by CD4 cell count.26 Converting this to 2017 USD, the cost of hospitalization for a person with CD4 < 100 cells/μl was 
2 940 (range 
2 016 to 
3 696). We added the cost of 14 days of amphotericin + fluconazole for the treatment of cryptococcal meningitis, for a total hospital cost of 
3144.
ICER calculation & cost-effective threshold
The incremental cost-effectiveness ratio (ICER) is calculated by dividing the change in cost by the change in effectiveness of one strategy, compared to next best strategy. The ICER units are provided in US dollars per DALY.
Cost-effectiveness threshold is typically determined by local governments depending on their willingness to pay for specific interventions. While the WHO CHOICE guidelines used to recommend GDP per capita as a threshold for ‘highly cost-effective’, recent literature has discussed several reasons why such a blanket threshold is not appropriate.27,28 For the purpose of our analysis, we used a publication by Woods et al. that provided country-specific estimates for cost-effectiveness thresholds.29 We used the upper limit of the range provided. Thus, the cost-effectiveness threshold was 
293/DALY in Uganda, and 
1 474/DALY in South Africa.
Sensitivity analyses
We performed a set of sensitivity analyses to explore alternative meningitis treatment regimens. In these two-way sensitivity analyses, all parameters were kept the same as in Tables 1 and 2, but hospitalization costs and survival varied in accordance with previous published outcomes.30
Two alternative strategies for meningitis treated were a) Fluconazole 1200 mg daily for 14 days, and b) Amphotericin 1 mg/kg + flucytosine 50 mg/kg for 7 days.11 While flucytosine is not yet available in sub Saharan Africa, it may become available in the future.
Finally, we performed an analysis removing hospitalization costs entirely. While the perspective of this analysis is from that of the ministry of health of each country, it is not clear that these ministries actually pay hospital costs. Depending on the type of facility, hospitalization costs may be paid by the government, by the patient, or by insurance. Thus, in the scenario that hospital costs are not being paid by a public payer, it would be valuable to evaluate the potential costs and benefit of AmBisome + fluconazole pre-emptive therapy in this context. Here we included costs of screening, pre-emptive medications, and costs of medications in the hospital for meningitis.
Results
CrAg screening and pre-emptive treatment in Uganda
In the base case model analysis using the assumptions in Tables 1 and 2, of 1 million persons receiving CD4 testing, 160 000 patients with CD4 < 100 cells/μl are CrAg screened for a cost of 
540 495. Without any CrAg screening or pre-emptive treatment, 3 466 would die at a cost of 
3.8 million, due to the cost of hospitalization of persons identified as CrAg + but not treated, who subsequently develop meningitis. While this is not the recommended standard of care, many locations in low and middle income countries have not implemented CrAg screening.
In a scenario of CrAg screening with fluconazole pre-emptive treatment (with presumed efficacy of 75%) which is current standard of care, the cost is 
3 million (due to cost of fluconazole, and those hospitalized with meningitis) and one could expect 1 761 deaths. Single-dose AmBisome + fluconazole at 75% efficacy is thus not cost-effective compared to fluconazole, given that it would be more expensive for the same efficacy. This scenario was dominated and thus removed from further analysis. As the efficacy of AmBisome increases, there are fewer deaths, and costs are reduced, given fewer people break through pre-emptive therapy with meningitis and require hospitalization (Table 3). At a price of 
24 per vial, and an efficacy of 85% or greater, AmBisome + fluconazole would cost 
220. (or less) per DALY averted in Uganda (Fig. 1). At a cost-effectiveness threshold of 
269 per DALY averted, at a cost of 
16.25 to 
24 per vial, and a minimal efficacy of 85%, AmBisome + fluconazole would be considered a cost-effective strategy.
Table 3.
Varying efficacy of AmBisome + fluconazole pre-emptive treatment in Uganda and South Africa.
| Uganda | South Africa | |||||||
|---|---|---|---|---|---|---|---|---|
| Regimen | Cost of 50 mg AmBisome vial (USD) | Efficacy | Cost of screening, treatment, & meningitis for 1 million PLWH (USD) | Deaths | Incremental cost-effectiveness ratio ( /DALY)* |
Cost of screening, treatment, & meningitis for 1 million PLWH (USD) | Deaths | Incremental cost-effectiveness ratio ( /DALY*) |
| Fluconazole | 75% |
2 994 544 |
1761 | Reference |
4 445 597 |
693 | Reference | |
| AmBisome + fluconazole |
16.25 |
75% |
5 149 052 |
1761 | ∞ |
5 236 097 |
693 | ∞ |
| AmBisome + fluconazole |
16.25 |
80% |
4 880 159 |
1540 |
414 |
4 819 889 |
629 |
284 |
| AmBisome + fluconazole |
16.25 |
85% |
4 611 266 |
1319 |
178 |
4 403 681 |
565 | − 16 |
| AmBisome + fluconazole |
16.25 |
90% |
4 342 373 |
1098 |
99 |
3 987 474 |
501 | − 116 |
| AmBisome + fluconazole |
16.25 |
95% |
4 073 481 |
876 |
59 |
3 571 266 |
437 | − 166 |
| AmBisome + fluconazole |
24 |
75% |
5 312 159 |
1761 | ∞ |
5 524 397 |
693 | ∞ |
| AmBisome + fluconazole |
24 |
80% |
5 157 253 |
1540 |
475 |
5 108 189 |
629 |
503 |
| AmBisome + fluconazole |
24 |
85% |
5 002 348 |
1319 |
220 |
4 691 981 |
565 |
93 |
| AmBisome + fluconazole |
24 |
90% |
4 847 443 |
1098 |
136 |
4 275 774 |
501 | − 43 |
| AmBisome + fluconazole |
24 |
95% |
4 692 538 |
876 |
93 |
3 859 566 |
437 | − 111 |
| AmBisome + fluconazole |
32 |
75% | 6 341 279 | 1761 | ∞ |
5 821 997 |
693 | ∞ |
| AmBisome + fluconazole |
32 |
80% | 6 186 373 | 1540 |
701 |
5 405 789 |
629 |
728 |
| AmBisome + fluconazole |
32 |
85% | 6 031 468 | 1319 |
334 |
4 989 581 |
565 |
206 |
| AmBisome + fluconazole |
32 |
90% | 5 876 563 | 1 098 |
211 |
4 573 374 |
501 |
32 |
| AmBisome + fluconazole |
32 |
95% | 5 721 658 | 876 |
149 |
4 157 166 |
437 | − 55 |
| AmBisome + fluconazole |
40 |
75% | 8 204 252 | 1761 | ∞ |
6 119 597 |
693 | ∞ |
| AmBisome + fluconazole |
40 |
80% | 7 935 359 | 1540 |
1085 |
5 703 389 |
629 |
954 |
| AmBisome + fluconazole |
40 |
85% | 7 666 466 | 1319 |
513 |
5 287 181 |
565 |
319 |
| AmBisome + fluconazole |
40 |
90% | 7 397 573 | 1098 |
322 |
4 870 974 |
501 |
108 |
| AmBisome + fluconazole |
40 |
95% | 7 128 681 | 876 |
227 |
4 454 766 |
437 |
2 |
*Compared to fluconazole pre-emptive treatment alone. Abbreviations: DALY = disability adjusted life year saved, PLWH = people living with HIV. For reference Ugandan cost-effectiveness threshold is 
293/DALY. South African cost-effectiveness threshold is 
1 474/DALY.
Figure 1.
Cost per DALY averted by efficacy of AmBisome + fluconazole pre-emptive treatment regimen in Uganda.
CrAg screening and pre-emptive treatment in South Africa
In a scenario of CrAg screening with fluconazole pre-emptive treatment (with presumed efficacy of 75%), the cost is 
4.4 million (due to cost of fluconazole, and those hospitalized with meningitis) and one could expect 693 deaths (Fig. 2). As described above, AmBisome + fluconazole at 75% efficacy is dominated, and was removed from the analysis. As the efficacy of AmBisome + fluconazole increases, there are fewer deaths, and costs are reduced, given fewer people break through pre-emptive therapy with meningitis and require hospitalization. The results of the various scenarios are reported in Table 3. At 
16.25 per vial, AmBisome is cost-saving at an efficacy of 85% or greater compared to fluconazole monotherapy. At a cost of 
24 per vial, AmBisome is cost-saving at an efficacy of 90% or greater. And at 
32 per vial, AmBisome is cost-saving at an efficacy of 95% or greater. In summary, investing in AmBisome could save lives and save money compared to using fluconazole pre-emptive therapy alone, given the high efficacy, and reduced hospitalization costs.
Figure 2.
Cost per DALY averted by efficacy of AmBisome + fluconazole pre-emptive treatment regimen in South Africa. At 90% efficacy this regimen is cost saving at 
24 per vial in South Africa. Cost savings is depicted by bars extending below the X-axis.
Sensitivity analyses
Alternative treatment regimens for cryptococcal meningitis
If those hospitalized with meningitis were treated with fluconazole 1200 mg daily alone, the cost of hospitalization would cost 
300 (after eliminating the costs of amphotericin, the necessary supplies and laboratory monitoring) in Uganda, and 
2945 in South Africa. We assumed 65% 2-week survival with fluconazole monotherapy for cryptococcal meningitis, and 45% 10-week survival.31 In settings where fluconazole monotherapy is used for meningitis induction therapy, a costly but effective regimen of AmBisome + fluconazole pre-emptive therapy is generally favorable both in low and middle income settings, given poor efficacy of the meningitis treatment regimen (Supplementary Table 1).
If treatment for meningitis was 1 week of amphotericin + flucytosine,11 and hospitalization was reduced to 7 days instead of 14 days, the total cost of hospitalization in Uganda was 
418, and 
3144 in South Africa. Survival with amphotericin + flucytosine for meningitis treatment was presumed to be 88% at 2 weeks, and 72% at 10 weeks.30 At 
16.25 per vial, and 85% efficacy or greater this regimen was cost-saving in South Africa (Supplementary Table 2). Using a regimen of Amphotericin + flucytosine for meningitis resulted in the fewest deaths overall, given the high efficacy of this regimen.
Elimination of hospital costs
In both countries, after removing hospital costs (but including medication costs for meningitis and for pre-emptive treatment), the cost per death averted with AmBisome + fluconazole pre-emptive therapy is highlighted in Supplemental Table. Compared to the base case analysis it costs more to avert one death with AmBisome.
Discussion
In this study, we found that use of a one time, high dose of AmBisome + fluconazole to prevent cryptococcal meningitis is cost-saving at an efficacy of 85% or greater in South Africa, if priced at 
16.25 per vial. If priced at 
24 per vial, and at a minimum efficacy of 90%, high dose AmBisome + fluconazole has the potential to be cost-saving in South Africa and presumably other middle-income countries. When using the cost-effectiveness threshold of 
269/DALY in Uganda, high dose AmBisome + fluconazole would be cost-effective in Uganda (and other low-income settings) at a cost of 
24 per vial or less, and an efficacy of 85% or greater.
Cost-effectiveness conclusions did not dramatically change when incorporating alternative therapies to treat meningitis or with eliminating hospital costs from our analysis. At a minimum of 80% efficacy of AmBisome + fluconazole and a cost of 
40 per vial, it cost as much as 
1085 per DALY averted in Uganda. These results provide a framework for ministries of health and international organizations to budget and consider use of liposomal amphotericin in low and middle income settings in sub Saharan Africa. If high dose AmBisome proves efficacious in the treatment of asymptomatic cryptococcal antigenemia, the effectiveness in outpatient settings outside of clinical trials will be needed. Implementation and operational research on storage, administration, training of healthcare workers, and effectiveness in rural clinics will be valuable. Areas of future research include whether co-administration of IV fluids is needed, whether laboratory monitoring is necessary. If neither IV fluids of laboratory monitoring is needed, then administration could potentially occur in more limited settings.
AmBisome for the prevention of cryptococcal meningitis may additionally have a role in populations where fluconazole is contraindicated, for example, in pregnant and breastfeeding women. Strategies to prevent cryptococcal meningitis have largely excluded pregnant women for this reason.32 AmBisome as monotherapy should be investigated in such populations.
A major limitation of our analysis is that CrAg screening was dependent on CD4 testing. In the move towards HIV test-and-treat, CD4 testing is becoming obsolete, as it is no longer required to decipher who should start ART. Thus, many persons entering HIV care are no longer getting CD4 testing.33 In the absence of CD4 testing, CrAg screening for everyone entering care is cost-effective in Uganda,20 but we did not specifically evaluate this in the context of the new pre-emptive treatment regimen.
Additional limitations of our analysis are related to input parameters of our model. We used data from trials in sub Saharan Africa, but some parameters are unknown. For example, the proportion with meningitis who present to hospital is highly variable. In our model we assumed that 80% of people with symptomatic meningitis would present to the hospital. Though, we expect that estimates will vary from country to country, and even within a country, depending on public or private setting. Thus, there is likely variability in our estimates when applied to other settings. However, in all of our sensitivity analyses, regardless of the posited proportion hospitalized, cryptococcal treatment regimen, public or private hospitalization location, CrAg screening is always favorable.
In summary, more effective antifungal prophylactic therapies are needed. Single dose AmBisome is promising as pre-emptive therapy, and clinical trials are underway. Here we present the theoretical efficacy of AmBisome that would be needed to demonstrate a cost-effectiveness, thereby implementation on a national scale can be considered. Overall, single-dose AmBisome pre-emptive therapy appears more cost-efficient in a middle income setting where hospitalization costs for meningitis, and GDP per capita are higher.
Supplementary Material
Acknowledgments
RR is supported by the National Institute of Allergy and Infectious diseases (K23AI138851). BL supported by grant from CDC foundation. Gilead Sciences Inc. has provided funding to the Meningitis foundation, with which RR, DM, and DB are affiliated. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the funding agencies. The source of this information is the Global Cryptococcal Antigen Screening Initiative, a project of the CDC Foundation funded by a grant from Pfizer Inc.
Contributor Information
Radha Rajasingham, Division of Infectious Diseases & International Medicine, University of Minnesota, MN 55455, USA.
Elizabeth Nalintya, Infectious Diseases Institute, Makerere University, Kampala, Uganda.
Dennis M Israelski, Medical Affairs, Global Patient Solutions, Gilead Sciences, Inc., CA 94404, USA.
David B Meya, Infectious Diseases Institute, Makerere University, Kampala, Uganda.
Bruce A Larson, Department of Global Health, Boston University School of Public Health, Boston, MA 02118, USA.
David R Boulware, Division of Infectious Diseases & International Medicine, University of Minnesota, MN 55455, USA.
Declaration of interest
The authors report no conflicts of interest. The authors alone are responsible for the content and the writing of the paper.
References
- 1. UNAIDS . Forty years on and new UNAIDS report gives evidence that we can end AIDS. 2021. https://www.unaids.org/en/resources/presscentre/pressreleaseandstatementarchive/2021/june/20210603_global-commitments-local-action (accessed September 23, 2021). [Google Scholar]
- 2. 2020 U. Global AIDS Monitoring 2021. 2020. https://www.unaids.org/sites/default/files/media_asset/global-aids-monitoring_en.pdf (accessed March 22, 2021). [Google Scholar]
- 3. Rajasingham R, Smith RM, Park BJet al. Global burden of disease of HIV-associated cryptococcal meningitis: an updated analysis. Lancet Infect Dis 2017; 17: 873–881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Beyene T, Zewde AG, Balcha Aet al. Inadequacy of high-dose fluconazole monotherapy among cerebrospinal fluid cryptococcal antigen (CrAg)-Positive human immunodeficiency virus-infected persons in an Ethiopian Crag Screening Program. Clin Infect Dis 2017; 65: 2126–2129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Butler EK, Boulware DR, Bohjanen PR, Meya DB.. Long term 5-year survival of persons with cryptococcal meningitis or asymptomatic subclinical antigenemia in Uganda. PLoS One 2012; 7: e51291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Jarvis JN, Bicanic T, Loyse Aet al. Determinants of mortality in a combined cohort of 501 patients with HIV-associated Cryptococcal meningitis: implications for improving outcomes. Clin Infect Dis 2014; 58: 736–745. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Meya DB, Manabe YC, Castelnuovo Bet al. Cost-effectiveness of serum cryptococcal antigen screening to prevent deaths among HIV-infected persons with a CD4+ cell count < or = 100 cells/microL who start HIV therapy in resource-limited settings. Clin Infect Dis 2010; 51: 448–455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jarvis JN, Lawn SD, Vogt M, Bangani N, Wood R, Harrison TS.. Screening for cryptococcal antigenemia in patients accessing an antiretroviral treatment program in South Africa. Clin Infect Dis 2009; 48: 856–862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Letang E, Muller MC, Ntamatungiro AJet al. Cryptococcal antigenemia in immunocompromised human immunodeficiency virus patients in rural tanzania: a preventable cause of early mortality. Open Forum Infect Dis 2015; 2: ofv046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Mfinanga S, Chanda D, Kivuyo SLet al. Cryptococcal meningitis screening and community-based early adherence support in people with advanced HIV infection starting antiretroviral therapy in Tanzania and Zambia: an open-label, randomised controlled trial. Lancet 2015; 385: 2173–2182. [DOI] [PubMed] [Google Scholar]
- 11. World Health Organization . Guidelines for Managing Advanced HIV Disease and Rapid Initiation of Antiretroviral Therapy. 2017. http://apps.who.int/iris/bitstream/10665/255884/1/9789241550062-eng.pdf?ua=1. [PubMed] [Google Scholar]
- 12. Ministry of Health U . Uganda Clinical Guidelines. 2016. https://www.health.go.ug/sites/default/files/Uganda%20Clinical%20Guidelines%202016_FINAL.pdf (accessed March 23, 2020). [Google Scholar]
- 13. Meya DB, Kiragga AN, Nalintya Eet al. Reflexive laboratory-based cryptococcal antigen screening and pre-emptive fluconazole therapy for cryptococcal antigenemia in HIV-Infected individuals with CD4 <100 Cells/microl: a stepped-wedge, cluster-randomized trial. J Acquir Immune Defic Syndr 2019; 80: 182–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Wake RM, Govender NP, Omar Tet al. Cryptococcal-related mortality despite fluconazole pre-emptive treatment in a cryptococcal antigen Screen-and-Treat program. Clin Infect Dis 2020; 70: 1683–1690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Sundar S, Singh A, Rai M, Chakravarty J.. Single-dose indigenous liposomal amphotericin B in the treatment of Indian visceral leishmaniasis: a phase 2 study. Am J Trop Med Hyg 2015; 92: 513–517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Jarvis JN, Leeme TB, Molefi Met al. Short-course high-dose liposomal amphotericin B for human immunodeficiency virus-associated Cryptococcal meningitis: a phase 2 randomized controlled trial. Clin Infect Dis 2019; 68: 393–401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. World Health Organization . Rapid advice: Diagnosis, prevention and management of cryptococcal disease in HIV-infected adults, adolescents and children. Geneva; 2011. p. 44. [PubMed] [Google Scholar]
- 18. Loyse A, Burry J, Cohn Jet al. Leave no one behind: response to new evidence and guidelines for the management of cryptococcal meningitis in low-income and middle-income countries. Lancet Infect Dis 2019; 19: e143–e7. [DOI] [PubMed] [Google Scholar]
- 19. Loyse A, Thangaraj H, Easterbrook Pet al. Cryptococcal meningitis: improving access to essential antifungal medicines in resource-poor countries. Lancet Infect Dis 2013; 13: 629–637. [DOI] [PubMed] [Google Scholar]
- 20. Rajasingham R, Meya DB, Greene GSet al. Evaluation of a national cryptococcal antigen screening program for HIV-infected patients in Uganda: a cost-effectiveness modeling analysis. PLoS One 2019; 14: e0210105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Boulware DR, Rolfes MA, Rajasingham Ret al. Multisite validation of cryptococcal antigen lateral flow assay and quantification by laser thermal contrast. Emerg Infect Dis 2014; 20: 45–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Carlson RD, Rolfes MA, Birkenkamp KEet al. Predictors of neurocognitive outcomes on antiretroviral therapy after cryptococcal meningitis: a prospective cohort study. Metab Brain Dis 2014; 29: 269–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. World Health Organization. Global Health Observatory data. 2018(accessed April 30, 2018). [Google Scholar]
- 24. Uganda Revenue Authority. 2018. www.ura.go.ug (accessed July 31, 2016). [Google Scholar]
- 25. International Medical Products Price Guide . 2016. http://mshpriceguide.org/en/home/ (accessed May 7 2019). [Google Scholar]
- 26. Meyer-Rath G, Brennan AT, Fox MPet al. Rates and cost of hospitalization before and after initiation of antiretroviral therapy in urban and rural settings in South Africa. J Acquir Immune Defic Syndr 2013; 62: 322–328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. WHO . Making choices in health: WHO guide to cost-effectiveness analysis. 2003. http://www.who.int/choice/book/en (accessed May 31, 2012). [Google Scholar]
- 28. Marseille E, Larson B, Kazi DS, Kahn JG, Rosen S. Thresholds for the cost-effectiveness of interventions: alternative approaches. Bulletin of the World Health Organization 2015; 93: 118–124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Woods B, Revill P, Sculpher M, Claxton K.. Country-Level cost-effectiveness thresholds: initial estimates and the need for further research. Value Health 2016; 19: 929–935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Molloy SF, Kanyama C, Heyderman RSet al. Antifungal combinations for treatment of Cryptococcal meningitis in Africa. N Engl J Med 2018; 378: 1004–1017. [DOI] [PubMed] [Google Scholar]
- 31. Rajasingham R, Rolfes MA, Birkenkamp KE, Meya DB, Boulware DR.. Cryptococcal meningitis treatment strategies in resource-limited settings: a cost-effectiveness analysis. PLoS Med 2012; 9: e1001316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Pastick KA, Nalintya E, Tugume Let al. Cryptococcosis in pregnancy and the postpartum period: case series and systematic review with recommendations for management. Med Mycol 2020; 58: 282–292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Nasuuna E, Tenforde MW, Muganzi A, Jarvis JN, Manabe YC, Kigozi J.. Reduction in baseline CD4 count testing following human immunodeficiency virus “Treat all” adoption in Uganda. Clin Infect Dis 2020; 71: 2497–2499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Siedner MJ, Ng CK, Bassett IV, Katz IT, Bangsberg DR, Tsai AC.. Trends in CD4 count at presentation to care and treatment initiation in sub-Saharan Africa, 2002-2013: a meta-analysis. Clin Infect Dis 2015; 60: 1120–1127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Avila D, Althoff KN, Mugglin Cet al. Immunodeficiency at the start of combination antiretroviral therapy in low-, middle-, and high-income countries. J Acquir Immune Defic Syndr 2014; 65: e8–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Nalintya E, Meya DB, Lofgren S, Huppler Hullsiek K, Boulware DR, Rajasingham R. A prospective evaluation of a multisite cryptococcal screening and treatment program in HIV clinics in Uganda. J Acquir Immune Defic Syndr 2018; 78: 231–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Cassim N, Coetzee LM, Schnippel K, Glencross DK.. Estimating the cost-per-result of a national reflexed cryptococcal antigenaemia screening program: forecasting the impact of potential HIV guideline changes and treatment goals. PLoS One 2017; 12: e0182154. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.