Rezafungin Versus Caspofungin for the Treatment of Candidemia and Invasive Candidiasis: Results from the Double-blind, Randomized, Phase 3 ReSTORE Trial Including the China Extension Study

Abstract

Background

Rezafungin is approved for use in adults with candidemia and/or invasive candidiasis (IC) based on data from the ReSTORE trial (NCT03667690), which demonstrated noninferior efficacy to caspofungin. For regulatory purposes, an additional cohort of patients from China was recruited to ReSTORE. Here, we compared rezafungin and caspofungin in patients with candidemia and/or IC through analysis of the ReSTORE global data plus the China extension study.

Methods

Adults with candidemia/IC were randomized (1:1) to receive weekly rezafungin (400/200 mg) or daily caspofungin (70/50 mg) for ≤28 days. Noninferiority was concluded for primary efficacy endpoints if the upper bound of the 95% confidence interval (CI) was below 20% for Day 30 all-cause mortality and if the weighted lower bound was above −20% for Day 14 global cure. Additional efficacy outcomes and safety were evaluated.

Results

Overall, 246 patients were randomized (122 rezafungin and 124 caspofungin). Noninferiority was demonstrated for both primary endpoints. Day 30 all-cause mortality was 25.2% and 24.8% (treatment difference 0.4%; 95% CI −10.8, 11.6) and Day 14 global cure was 56.5% and 57.3% (weighted treatment difference −1.0%; 95% CI −13.5, 11.6) with rezafungin versus caspofungin, respectively. Day 5 mycological eradication was numerically higher (68.7% vs 63.2%) and time to negative blood culture was numerically shorter (median 26.5 vs 38.8 h). Safety was comparable between groups; 53.3% (64/120; rezafungin) and 53.7% (66/123; caspofungin) of patients experienced serious adverse events.

Conclusions

This analysis confirmed the overall efficacy and safety of rezafungin demonstrated in ReSTORE, with early efficacy related to front-loaded exposure.

This analysis of the ReSTORE global data plus the China extension study confirms the efficacy and safety of rezafungin for the treatment of candidemia and invasive candidiasis in adults, with early efficacy related to front-loaded exposure.

Despite advances in antifungal therapies, candidemia and invasive candidiasis (IC) continue to have high morbidity and mortality rates [1–3]. The global incidence of candidemia and IC is estimated to be ∼3–5 cases per 100 000 persons, occurring in 1%–2% of all patients admitted to intensive care units (ICUs), although there are regional variations [2, 4, 5]. In the USA and Europe, the incidence of candidemia is 4–9 cases per 100 000 inhabitants [1, 6–8], while in China it is 26 per 100 000 admissions [9]. Among hospital inpatients, IC incidence is 90 per 100 000 admissions in the USA [10]; in Europe, the cumulative incidence is 700 per 100 000 ICU admissions [11].

Although treatment strategies for candidemia and IC can vary across the USA, Europe, and China [12–15], and ICU models differ systemically [16], the approach to first-line treatment of candidemia and IC is consistent. Echinocandins, antifungal drugs targeting fungal cell wall biosynthesis [17], are recommended in all 3 regions, including in critically ill patients and for targeted candidemia treatment [12–15]. The ongoing global rise in treatment-resistant, multidrug-resistant, non-albicans Candida species, particularly C. glabrata and C. auris, however, presents a growing healthcare concern that highlights the demand for new antifungal agents [2, 4, 18–20].

Rezafungin is a next-generation echinocandin approved for use in adults with candidemia and/or IC by regulatory authorities, including the US Food and Drug Administration and the European Medicines Agency [21, 22]. Rezafungin exhibits potent activity against azole- and some echinocandin-resistant Candida species in vitro [23]. Its stability and pharmacokinetic properties allow once-weekly intravenous dosing and high front-loaded exposures [24–26].

The efficacy and safety of rezafungin have been evaluated in patients with candidemia and/or IC in the Phase 2 STRIVE trial (NCT02734862) [27] and the Phase 3 ReSTORE trial (NCT03667690) [28]. ReSTORE demonstrated that rezafungin was noninferior to caspofungin for Day 30 all-cause mortality (ACM) and Day 14 global cure, with a similar safety profile [28]. Pooled analysis of STRIVE and ReSTORE [29] further supported the efficacy and safety of rezafungin [21, 22]. The trials also suggested rezafungin demonstrated faster mycological eradication than caspofungin, potentially due to its front-loaded dosing [27, 28].

To meet the regulatory requirements of the Chinese Center for Drug Evaluation of National Medical Products Administration, a cohort of patients from China was recruited to the ReSTORE trial (ReSTORE China) [30], facilitating analysis of rezafungin in a larger, more diverse population. A separate analysis of this ReSTORE China extension study combined with patients from China enrolled in the original ReSTORE trial demonstrated similar efficacy and safety for rezafungin versus caspofungin, in line with the findings of the primary ReSTORE analysis [30]. This analysis of ReSTORE plus the ReSTORE China extension study aimed to evaluate the efficacy and safety of rezafungin in the final complete, combined data set of all ReSTORE patients, representing the largest available sample from this trial. In a separate analysis, conducted as part of the regulatory assessment, the pharmacokinetics of rezafungin in a China versus a non-China cohort of ReSTORE patients were also evaluated.

METHODS

Patients and Study Design

Detailed methodology for the randomized, double-blind, noninferiority, Phase 3 ReSTORE study has been reported previously [28, 30]. Briefly, adults with candidemia and/or IC, determined by systemic signs and mycological evidence (from blood or a normally sterile site ≤96 h before randomization) of Candida infection, were randomized (1:1) to receive rezafungin or caspofungin intravenous treatment. Randomization used a random allocation sequence and was stratified by diagnosis (candidemia only vs IC) and a combination of modified Acute Physiology and Chronic Health Evaluation (APACHE) II score and absolute neutrophil count (ANC) at screening (APACHE II score ≥20 or ANC <500 cells/μL vs APACHE II score <20 and ANC ≥500 cells/μL).

Rezafungin was administered on Day 1 (400 mg) and Day 8 (200 mg), with optional 200 mg doses on Days 15 and 22 (investigator's decision). Caspofungin was administered on Day 1 (70 mg), followed by 50 mg daily for 3–28 days. If step-down criteria were met (ie, the Candida species was susceptible to fluconazole, symptoms had resolved, and blood culture was negative for Candida), patients began oral step-down therapy after ≥3 days of intravenous treatment (or the minimum duration of intravenous therapy advised by the site's national/regional/local guidelines, whichever was greater). Participants receiving weekly rezafungin stepped down to daily placebo (to avoid double dosing), while participants receiving daily caspofungin stepped down to daily fluconazole. Intravenous treatment with rezafungin or caspofungin was initiated on an inpatient basis; patients receiving rezafungin could switch to outpatient once-weekly intravenous rezafungin treatment at the investigator's discretion.

Patient Consent Statement

All participants, or their legally authorized representatives, provided written informed consent for study participation. The trial was carried out in accordance with all relevant regulations, the International Conference on Harmonisation Good Clinical Practice guidelines, and the Declaration of Helsinki. The protocol, including any amendments, was approved by independent ethics committees or institutional review boards at the trial sites.

Endpoints and Assessments

The primary endpoints for this analysis were ACM at Day 30, defined as death regardless of cause or unknown survival status (ie, missing data) at Day 30 (or within 2 days before Day 30), and global cure at Day 14, defined as clinical cure assessed by the investigator, radiological cure (for participants with IC), and mycological eradication at Day 14 (or within 1 day either side), as confirmed for all 3 by an independent, blinded data review committee (DRC). Missing data for global cure, clinical cure, radiological cure, and mycological eradication resulted in classification as nonresponse.

Key secondary endpoints were global cure at Day 5, mycological eradication at Days 5 and 14, and clinical cure at Days 5 and 14. Percentages of patients with negative blood cultures at 24 and 48 h after the first dose of study drug and time to negative blood culture (TTNBC; time from the first study drug dose to first negative Candida culture, with no subsequent positive cultures after 24 and 48 h) were exploratory outcomes. ICU and hospital admissions were also evaluated.

Clinical symptoms and physical findings were assessed at screening, on Days 5, 14, and 28–30, at the end of treatment, and at follow-up (Days 52–59), with repeated Candida cultures performed daily, or every other day, until a negative result was obtained. Fungal isolates underwent species identification and susceptibility testing (central laboratories in North Liberty, IA, USA, and Shanghai, China). Isolates were identified using matrix-assisted laser desorption ionization–time of flight (MALDI–TOF) using the MALDI Biotyper (Bruker Daltonics, Billerica, MA, USA) or the VITEK2 COMPACT system (bioMérieux, Marcy-l’Étoile, France; trial isolates from China only). Susceptibility testing was carried out using Clinical and Laboratory Standards Institute (CLSI) broth microdilution (M27 Ed4) methods, per quality control guidelines (M27M44S Ed3).

Safety was assessed through evaluation of treatment-emergent adverse events (TEAEs), vital signs, physical examination, electrocardiogram, and clinical laboratory data. TEAEs were coded using the Medical Dictionary for Regulatory Activities version 23.0, and severity was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 5.0.

Exposure parameters for rezafungin following a single 400 mg dose were compared in a China and non-China cohort of patients using population pharmacokinetic modeling. A population pharmacokinetic model for rezafungin was developed previously, based on data from healthy individuals, people with hepatic impairment, and people with candidemia and/or IC, including 6 patients from China who participated in the global ReSTORE trial [31]. Pharmacokinetic sampling was conducted in 22 patients in the ReSTORE China extension study and the model was updated to include these data. This updated model was then used to simulate rezafungin exposure data for the 28 patients in the China cohort (comprising the 6 patients from the global ReSTORE trial plus the 22 patients from the ReSTORE China extension study) and 161 patients in a non-China comparator cohort. Area under the concentration–time curve from time 0 to 168 h (AUC_0–168h), maximum drug concentration (C_max), and minimum drug concentration (C_min) were derived for both China and non-China cohort patients following a 400 mg dose. C_min was defined as the simulated plasma concentration at 168 h after the 400 mg dose, while C_max was defined as the simulated maximum plasma concentration after the 400 mg dose.

The probability of achieving nonclinical pharmacokinetic/pharmacodynamic targets for net fungal stasis and 1-log drop in colony-forming units, across the range of minimum inhibitory concentration (MIC) values for 6 Candida species, based on CLSI and European Committee on Antimicrobial Susceptibility Testing (EUCAST) methods, was estimated for a virtual population of patients from China. Using the updated population pharmacokinetic model (described above) and distributions of demographic covariates (determined previously to be significant predictors of rezafungin pharmacokinetics [31]) of patients in the China cohort, plasma concentration–time profiles were simulated for a virtual population of 100 000 patients from China following a 400 mg dose of rezafungin, from which the AUC_0–168h was derived. These derived exposures were adjusted for unbound “free” drug and used to calculate the free-drug (f) AUC_0–168h/MIC ratios over the MIC distribution ranges for the 6 Candida species, to be used in the probability of target attainment analyses (the methods for which have been described previously [31]).

Statistical Analysis

Sample size calculations were reported previously [28, 30]. Noninferiority testing was performed for the primary endpoints as in ReSTORE [28]. Two-sided 95% confidence intervals (CIs) were calculated using Miettinen–Nurminen methodology (Day 30 ACM: unweighted; Day 14 global cure: weighted by randomization strata). Rezafungin was considered noninferior to caspofungin if the upper bound of the CI was below 20% for Day 30 ACM and if the lower bound of the weighted CI was above −20% for Day 14 global cure. Patients with an indeterminate response were considered not to have global cure (ie, were treatment failures). For the exploratory outcome of TTNBC, data were summarized using Kaplan–Meier methods. ICU and hospital admissions, pharmacokinetic data, and safety data were analyzed descriptively. Efficacy analyses were conducted in the modified intent-to-treat (mITT) population, comprising patients with documented Candida infection within 96 h before randomization who received ≥1 dose of study drug. Safety analyses were conducted in the safety population, comprising patients who received any study drug. Statistical analyses used SAS version 9.4 (SAS Institute, Cary, NC, USA).

RESULTS

Demographics and Baseline Characteristics

Of 271 individuals screened, 246 patients were randomized (intent-to-treat [ITT] population): 122 and 124 in rezafungin and caspofungin groups, respectively (Supplementary Figure 1). Three patients did not receive study drug, thus the safety population comprised 120 and 123 patients in rezafungin and caspofungin groups, respectively; correspondingly, the mITT population included 115 and 117 patients.

Demographics and baseline characteristics of the combined ReSTORE and ReSTORE China extension studies were generally well balanced and similar between treatment groups, although the proportion of male participants was numerically higher in the rezafungin group (69.7% [85/122] vs 56.5% [70/124] in the caspofungin group) (ITT population; Table 1). Patients from China were distributed equally between groups (rezafungin: 25.4% [31/122]; caspofungin: 25.0% [31/124]).

Table 1.

Demographics and Baseline Characteristics (ITT Population)

Characteristic	Rezafungin 400/200 mg (n = 122)	Caspofungin 70/50 mg (n = 124)
Median age, years (min, max)	58.5 (18, 89)	62.0 (19, 91)
Age category, n (%)
<65 y	73 (59.8)	73 (58.9)
≥65 y	49 (40.2)	51 (41.1)
Sex, n (%)
Male	85 (69.7)	70 (56.5)
Female	37 (30.3)	54 (43.5)
Race, n (%)
White	61 (50.0)	60 (48.4)
Asian	49 (40.2)	56 (45.2)
Black or African American	5 (4.1)	4 (3.2)
American Indian or Alaska Native	1 (0.8)	1 (0.8)
Other or not reported	6 (4.9)	3 (2.4)
Geographical region, n (%)
Europe and Israel	39 (32.0)	40 (32.3)
China and Taiwan	31 (25.4)	31 (25.0)
Asia–Pacific (excluding China and Taiwan)	25 (20.5)	28 (22.6)
USA	26 (21.3)	25 (20.2)
South America	1 (0.8)	0
Diagnosis at randomization, n (%)
Candidemia only	94 (77.0)	92 (74.2)
IC	28 (23.0)	32 (25.8)
Final diagnosis, n (%)
Candidemia only	90 (73.8)	90 (72.6)
ICa	32 (26.2)	34 (27.4)
Modified APACHE II scoreb, n (%)c
≥20	18 (14.8)	23 (18.5)
<20	103 (84.4)	101 (81.5)
Mean body mass index, kg/m2 (SD)c	24.8 (6.8)	24.0 (6.1)
ANC, n (%)c
<500 cells/µL	11 (9.0)	11 (8.9)
≥500 cells/µL	108 (88.5)	113 (91.1)
Mean estimated creatinine clearance, mL/min (SD)c	90.6 (103.1)	79.8 (57.9)
Child-Pugh score category, n (%)
<7	0	0
≥7	3 (2.5)	7 (5.6)
No history of liver disease (or not calculatedd)	119 (97.5)	117 (94.4)
≥1 Candida risk factor in the last 3 m, n (%)e	122 (100)	124 (100)
Central venous catheter	65 (53.3)	69 (55.6)
Peripherally inserted central catheter	31 (25.4)	24 (19.4)
Active malignancy	28 (23.0)	33 (26.6)
Broad-spectrum antibiotic therapy	91 (74.6)	83 (66.9)
Diabetes mellitus	29 (23.8)	36 (29.0)
Immunosuppression	22 (18.0)	23 (18.5)
Major surgery	37 (30.3)	40 (32.3)
Total parenteral nutrition	25 (20.5)	28 (22.6)
Transplant recipient	11 (9.0)	9 (7.3)
Trauma	9 (7.4)	4 (3.2)
End-stage renal disease/dialysis	11 (9.0)	12 (9.7)
Burns	1 (0.8)	2 (1.6)
Pancreatitis	11 (9.0)	7 (5.6)
Other	24 (19.7)	26 (21.0)
In ICU at time of randomization, n (%)	39 (32.0)	51 (41.1)
Current mechanical ventilation, n (%)	22 (18.0)	40 (32.3)
Dialysis within the past 3 d, n (%)	13 (10.7)	15 (12.1)
Pancreatitis within the past 10 d, n (%)	9 (7.4)	8 (6.5)
Catheter placement
Yes, n (%)	76 (62.3)	68 (54.8)
Removed within 48 h of diagnosis, n/N (%)	11/76 (14.5)	15/68 (22.1)

Abbreviations: ANC, absolute neutrophil count; APACHE, Acute Physiology and Chronic Health Evaluation; GCS, Glasgow Coma Score; IC, invasive candidiasis; ICU, intensive care unit; ITT, intent-to-treat; max, maximum; min, minimum; SD, standard deviation.

^aFinal diagnosis of IC was determined based on the investigator's response on the tissue/fluid culture assessment and radiological test case report form pages.

^bCombined APACHE II and GCS, calculated as APACHE II + (15 minus GCS).

^cReported for participants with available data.

^d n = 3.

^ePercentages are not mutually exclusive.

At baseline, treatment groups had a similar distribution of Candida species in blood and normally sterile culture sites (mITT population; Supplementary Table 1). C. albicans was the most commonly isolated species (in >40% of patients), followed by C. glabrata and C. tropicalis (≥22%). Two patients (both in the rezafungin group) had baseline isolates with echinocandin nonsusceptible MIC values according to CLSI breakpoints [32]. One had a C. dubliniensis isolate that was nonsusceptible to rezafungin (MIC 0.25 µg/mL), and the other had a C. glabrata isolate classified as intermediate for caspofungin (MIC 0.25 µg/mL) but susceptible to rezafungin (MIC 0.5 µg/mL).

Efficacy

Day 30 ACM was comparable between rezafungin and caspofungin, with 25.2% (29/115) and 24.8% (29/117) of patients, respectively, either having died or with unknown survival status (Table 2). Noninferiority of rezafungin was demonstrated, with a treatment difference of 0.4% (95% CI −10.8, 11.6). In participants with a final candidemia-only diagnosis, Day 30 ACM rates were 29.8% (25/84) and 29.5% (26/88) with rezafungin and caspofungin, respectively; correspondingly, for IC, rates were 12.9% (4/31) and 10.3% (3/29). In an exploratory analysis, most subgroups by baseline Candida species were too small for meaningful comparisons between treatments (Supplementary Table 2). In the largest subgroup, C. albicans at baseline, rates were comparable between rezafungin (31.9% [15/47]) and caspofungin (27.7% [13/47]); corresponding rates for C. parapsilosis, C. glabrata, C. krusei, and C. tropicalis combined were 21.9% (14/64) and 20.8% (15/72). Day 30 ACM rates by sex are presented in Supplementary Table 3.

Table 2.

All-cause Mortality at Day 30 and Global Response at Day 14 (mITT Population)

Endpoint	Rezafungin 400/200 mg (n = 115)	Caspofungin 70/50 mg (n = 117)	Treatment Difference (95% CI)
ACM at Day 30a, n (%)b
Died or unknown survival	29 (25.2)	29 (24.8)	0.4% (−10.8, 11.6)e
Known to have died	22 (19.1)	23 (19.7)
Unknown survival status	7 (6.1)	6 (5.1)
ACM at Day 30a by diagnosis, n/N (%)
Candidemia only	25/84 (29.8)	26/88 (29.5)	0.2% (−13.4, 13.9)e
IC	4/31 (12.9)	3/29 (10.3)	2.6% (−15.7, 20.5)e
Global response at Day 14c, as assessed by DRC, n (%)d
Cure	65 (56.5)	67 (57.3)	−1.0% (−13.5, 11.6)f
Failure	37 (32.2)	39 (33.3)
Indeterminate	13 (11.3)	11 (9.4)
Global response at Day 14 by diagnosis, as assessed by DRC, n/N (%)
Candidemia only
Cure	49/84 (58.3)	53/88 (60.2)	−1.9% (−16.5, 12.7)e
Failure	28/84 (33.3)	27/88 (30.7)
Indeterminate	7/84 (8.3)	8/88 (9.1)
IC
Cure	16/31 (51.6)	14/29 (48.3)	3.3% (−21.6, 27.9)e
Failure	9/31 (29.0)	12/29 (41.4)
Indeterminate	6/31 (19.4)	3/29 (10.3)

Abbreviations: ACM, all-cause mortality; ANC, absolute neutrophil count; APACHE, Acute Physiology and Chronic Health Evaluation; CI, confidence interval; DRC, data review committee; IC, invasive candidiasis; mITT, modified intent-to-treat.

^aAssessed on Day 28–30.

^bPrimary endpoint for the US Food and Drug Administration.

^c±1 d.

^dPrimary endpoint for the European Medicines Agency.

^eTwo-sided 95% CI for the observed difference (%), rezafungin − caspofungin group.

^fTwo-sided 95% CI for the observed differences, rezafungin − caspofungin group, adjusted for the 2 randomization strata (diagnosis [candidemia only vs IC] and APACHE II score/ANC [APACHE II score ≥20 or ANC <500 cells/μL vs APACHE II score <20 and ANC ≥500 cells/μL] at screening).

Day 14 global cure rates were comparable between groups overall (rezafungin: 56.5% [65/115]; caspofungin: 57.3% [67/117]), and noninferiority was demonstrated for rezafungin to caspofungin with a weighted treatment difference of −1.0% (95% CI −13.5, 11.6) (Table 2). Results were also comparable between groups within each diagnosis. Day 14 global cure rates by sex are shown in Supplementary Table 3.

At Day 5, mycological eradication rates were 68.7% (79/115) and 63.2% (74/117) with rezafungin and caspofungin, respectively; corresponding rates in participants with a candidemia-only diagnosis were 77.4% (65/84) and 68.2% (60/88) (Table 3). By Day 14, mycological eradication rates and other key secondary endpoints were comparable between groups. In an exploratory analysis of subgroups stratified by baseline Candida species, rates of mycological eradication were broadly comparable between rezafungin and caspofungin at Days 5 and 14, considering the relatively small subgroup sizes (Supplementary Table 4).

Table 3.

Key Secondary Endpoints (mITT Population)

Endpoint	Rezafungin 400/200 mg (n = 115)	Caspofungin 70/50 mg (n = 117)	Treatment Difference (95% CI)a
Day 5 endpoints, n (%)
Global cure, as assessed by DRC	60 (52.2)	57 (48.7)	3.5% (−9.4, 16.2)
Mycological eradicationb	79 (68.7)	74 (63.2)	5.4% (−6.8, 17.5)
Candidemia only, n/N (%)	65/84 (77.4)	60/88 (68.2)	9.2% (−4.2, 22.3)
Investigator assessment of clinical cure	65 (56.5)	73 (62.4)	−5.9% (−18.3, 6.8)
Day 14 endpoints, n (%)
Mycological eradicationb	76 (66.1)	77 (65.8)	0.3% (−11.9, 12.4)
Candidemia only, n/N (%)	59/84 (70.2)	60/88 (68.2)	2.1% (−11.8, 15.8)
Investigator assessment of clinical cure	71 (61.7)	75 (64.1)	−2.4% (−14.7, 10.0)

Abbreviations: CI, confidence interval; DRC, data review committee; mITT, modified intent-to-treat.

^aTwo-sided 95% CI for the observed difference (%), rezafungin − caspofungin group, unless otherwise stated.

^bProgrammatically derived; includes both documented and presumed eradication.

Proportions of patients with negative blood cultures were higher with rezafungin than caspofungin at 24 h (46.5% [95% CI 35.7, 57.6; 40/86] vs 36.0% [95% CI 26.0, 47.1; 31/86]) and 48 h (72.3% [95% CI 61.4, 81.6; 60/83] vs 59.5% [95% CI 48.3, 70.1; 50/84]) after the first dose (Figure 1). Median (95% CI) TTNBC was numerically shorter with rezafungin versus caspofungin (26.5 h [22.8, 41.2] vs 38.8 h [26.8, 57.4]), indicating a potentially faster mycological response with rezafungin. In an additional analysis of patients with a positive blood culture proximal to randomization (determined from blood sampled within 12 h prior to or 72 h after randomization, reflecting the clinical situation more closely), median (95% CI) TTNBC was 41.2 h (15.5, 90.3) with rezafungin and 62.0 h (41.6, 111.3) with caspofungin (Figure 2).

Figure 1.

Patients with negative blood culture^a at 24 and 48 h (mITT population with positive blood culture before randomization). This figure shows the percentage of patients with negative blood culture by treatment group at 24 and 48 h after the first dose in the mITT population with a positive blood culture before randomization. ^aParticipants censored prior to 24 and 48 h were excluded from the denominator for 24 and 48 h, respectively. Abbreviation: mITT, modified intent-to-treat.

Figure 2.

Time to first negative blood culture^a (mITT population with a positive blood culture sampled within 12 h prior to or 72 h after randomization). This figure shows the time to first negative blood culture by treatment group in the mITT population with a positive blood culture sampled within 12 h prior to or 72 h after randomization. The data were summarized using Kaplan–Meier methods. ^aTime to first negative blood culture, without subsequent positive culture, from a sample drawn following the first dose of study drug. Crosses indicate censored patients; patients were censored if they received an alternative antifungal (ie, other than study drug) for the treatment of the candidemia, died, or were lost to follow-up prior to having the negative blood culture. The P-value cannot be interpreted as hypothesis-testing because this was an exploratory outcome within a subgroup of patients. Abbreviation: mITT, modified intent-to-treat.

The proportions of ICU admissions and mean length of hospital admission were comparable between groups (Supplementary Results and Supplementary Table 5). Investigator perspectives on the potential for earlier patient discharge are included in the Supplementary Results.

Safety

Following intravenous study drug treatment, 23.3% (28/120) and 32.5% (40/123) of patients stepped down to oral therapy from blinded rezafungin and caspofungin, respectively. Switching most frequently occurred on Days 4–6 (rezafungin: 57.1% [16/28]; caspofungin: 45.0% [18/40]). Median (interquartile range [IQR]) study drug exposure was 14.0 (6.0–14.0) days and 14.0 (8.0–15.0) days for rezafungin and caspofungin, respectively.

TEAEs occurred in 91.7% (110/120) and 87.0% (107/123) of patients in the rezafungin and caspofungin groups, respectively (Table 4). The most common TEAEs overall were hypokalemia, pyrexia, anemia, pneumonia, and septic shock. Study drug-related TEAEs occurred in 17.5% (21/120) and 12.2% (15/123) of patients in the rezafungin and caspofungin groups, respectively. The most frequent study drug-related TEAE in the rezafungin group was infusion-related reaction (2.5% [3/120] of patients), while in the caspofungin group it was diarrhea (2.4% [3/123]). Similar proportions of patients in each group interrupted study drug, discontinued study drug, and discontinued the study owing to TEAEs (Table 4).

Table 4.

Overview of Treatment-emergent Adverse Events (Safety Population)

Participants with ≥1 TEAE, n (%)	Rezafungin 400/200 mg (n = 120)	Caspofungin 70/50 mg (n = 123)
Any TEAE	110 (91.7)	107 (87.0)
Any drug-related TEAE	21 (17.5)	15 (12.2)
Grade 4/5 TEAE	46 (38.3)	50 (40.7)
SAE	64 (53.3)	66 (53.7)
SAE leading to death	33 (27.5)	35 (28.5)
Drug-related SAE	2 (1.7)	3 (2.4)
Drug-related SAE leading to death	0	0
TEAE leading to interruption of study drug	3 (2.5)	1 (0.8)
TEAE leading to discontinuation of study drug	14 (11.7)	12 (9.8)
TEAE leading to study discontinuation	22 (18.3)	19 (15.4)
TEAEs occurring in ≥5% of patients in either arm
Hypokalemia	17 (14.2)	14 (11.4)
Pyrexia	15 (12.5)	7 (5.7)
Anemia	13 (10.8)	11 (8.9)
Pneumonia	12 (10.0)	7 (5.7)
Septic shock	12 (10.0)	12 (9.8)
Diarrhea	9 (7.5)	8 (6.5)
Abdominal pain	7 (5.8)	4 (3.3)
Hypomagnesemia	7 (5.8)	5 (4.1)
Sepsis	7 (5.8)	5 (4.1)
Vomiting	7 (5.8)	3 (2.4)
Hyponatremia	6 (5.0)	3 (2.4)
Multiple organ dysfunction syndrome	6 (5.0)	3 (2.4)
Nausea	6 (5.0)	3 (2.4)
Hypotension	5 (4.2)	7 (5.7)
Urinary tract infection	5 (4.2)	7 (5.7)
Acute kidney injury	4 (3.3)	9 (7.3)
Hypocalcemia	4 (3.3)	7 (5.7)
Respiratory failure	2 (1.7)	7 (5.7)

Abbreviations: SAE, serious adverse event; TEAE, treatment-emergent adverse event.

There was a similar incidence of serious adverse events between treatment groups (rezafungin: 53.3% [64/120]; caspofungin: 53.7% [66/123]) (Table 4). Study drug-related serious adverse events occurred in 1.7% (2/120) of patients in the rezafungin group (n = 1 each for infusion-related reaction and urticaria) and 2.4% (3/123) of patients in the caspofungin group (n = 1 each for hypertransaminasemia, liver injury, and anaphylactic shock).

Pharmacokinetics

Rezafungin exposure following a 400 mg dose was compared between the China and the non-China cohorts of patients. As presented in the Supplementary Results and Supplementary Figure 2, overall, there were no clinically meaningful differences in exposure between these groups.

The probability of target attainment analysis for a virtual population of patients from China after a 400 mg dose of rezafungin demonstrated coverage for the majority of MIC distributions for 6 Candida species, by both CLSI and EUCAST methodologies (Supplementary Figure 3).

DISCUSSION

As a relatively new drug in its class, rezafungin differs from other echinocandins in that its stability and pharmacokinetic properties permit less frequent, once-weekly (as opposed to once-daily) intravenous dosing and high front-loaded exposures [24–26]. It also appears to be suitable for use in various populations without the need for dose modifications, including in patients with organ dysfunction or obesity, and older and critically ill patients [33]. This analysis of the final complete, combined data from the Phase 3 ReSTORE trial and an extension study of patients from China, representing a more diverse population of all patients who participated in ReSTORE, demonstrated that weekly dosing of rezafungin was noninferior to daily dosing of caspofungin with respect to Day 30 ACM and Day 14 global cure. Day 30 ACM rates were comparable between rezafungin and caspofungin, including within subgroups by baseline Candida diagnosis (C. albicans vs non-C. albicans), with the caveat that any efficacy inferences from the subgroup analyses can only be tentative due to the limited sample sizes. Mycological eradication and clinical cure rates were also comparable between treatments, while there was a trend for earlier mycological clearance with rezafungin. The safety profile of rezafungin was comparable to that of caspofungin and consistent with existing reports [21, 22, 27–29, 34]. There were no clinically meaningful differences in rezafungin exposure between China and non-China cohort patients; furthermore, probability of target attainment simulations demonstrated that a 400 mg dose of rezafungin provided coverage for most MIC distributions for 6 Candida species in a virtual population of patients from China.

Efficacy and safety outcomes from this combined analysis of ReSTORE and the ReSTORE China extension study were consistent with the primary analysis of ReSTORE [28]. In terms of efficacy, Day 30 ACM rates were 25.2% and 24% for rezafungin and 24.8% and 21% for caspofungin in the current report and the primary analysis, respectively; correspondingly, Day 14 global cure rates were 56.5% and 59% for rezafungin and 57.3% and 61% for caspofungin [28]. The proportion of patients experiencing serious adverse events was similar between this analysis (rezafungin: 53.3%; caspofungin: 53.7%) and the previous report (56% and 53%, respectively), and the most frequent TEAEs were also comparable between reports [28].

Rezafungin was associated with numerically higher rates of negative blood culture at 24 and 48 h, and a numerically shorter TTNBC, than caspofungin. Although a nominal statistical difference was not demonstrated in this exploratory analysis, the data are consistent with the findings of the primary ReSTORE [28] and pooled STRIVE/ReSTORE [29] analyses. Earlier initiation of antifungal treatment is crucial for lowering the morbidity and mortality of candidemia and IC [35]. The low clearance and prolonged half-life of rezafungin (∼133 h) facilitate front-loaded, extended-interval dosing [27, 36]. This may rapidly reduce fungal load [37] and reduce the risk of underdosing—a frequent challenge in antifungal therapy, as seen with anidulafungin and caspofungin, where standard doses may not achieve adequate target attainment in critically ill patients [38–40]—potentially avoiding development of resistance mutations in Candida species [37]. The potential ability of rezafungin to rapidly reduce fungal load at the start of treatment, and its clinical implications, require further investigation.

This analysis suggested average ICU admission duration may be shorter for patients treated with rezafungin rather than caspofungin, consistent with an analysis of STRIVE [41] and a pooled STRIVE/ReSTORE analysis [42]. In the STRIVE/ReSTORE pooled analysis, patients in the ICU at randomization who received rezafungin had a numerically shorter length of ICU stay versus those who received caspofungin in all analyses [42]. If confirmed in future studies, the shorter ICU stay with rezafungin could have economic implications, as both candidemia and IC are associated with high hospitalization costs [43], particularly driven by resource-intensive ICU admissions [44]. In support of this, a cost-of-illness study of patients with Candida infections, based on real-world data from a German hospital, reported that the shorter ICU stay seen with rezafungin versus caspofungin in the STRIVE trial is associated with substantial cost savings [44]. Further, a preliminary (unpublished) cost-minimization/cost-utility analysis from the UK healthcare perspective has suggested that once-weekly rezafungin is both a cost-saving and a cost-effective treatment option in candidemia and IC compared with daily caspofungin, anidulafungin, or micafungin, with cost savings driven by ICU stay and treatment response [45]. It is possible, however, that confounding factors, such as differences in infection severity and hospital practices, could explain some of the observed difference in the duration of ICU stay.

This combined analysis permitted evaluation of rezafungin in a larger, more diverse patient population than the individual ReSTORE and ReSTORE China analyses [28, 30], and patients from China were distributed equally between treatment groups. Population pharmacokinetic analyses have not shown sex to be a statistically significant covariate impacting rezafungin pharmacokinetics [46], so the observed difference in the proportion of males in the rezafungin versus caspofungin group is unlikely to have had a major impact on the findings. In a subgroup analysis, variability in the data did not permit a meaningful comparison of the primary outcomes by sex between the 2 groups, and thus the marginal treatment differences by sex are unlikely to be clinically relevant. ReSTORE incorporated APACHE II scores into randomization strata, in line with the common use of APACHE II for illness severity assessment in clinical trials, although more recent prognostic models are available [47]. In addition, pediatric patients and patients with forms of IC requiring long courses of antifungal treatment were excluded from ReSTORE, and no patients had infections due to C. auris [28]. However, collectively, several case reports and case series suggest that rezafungin administered as a 400 mg loading dose followed by 200 mg weekly may be effective in chronic, difficult-to-treat forms of IC, such as endocarditis (in 1 case series of 6 adults, for example, the authors concluded that treatment was generally effective and well tolerated [48]). Further, robust evaluation of rezafungin is warranted in these specific populations. Importantly, studies of rezafungin in two additional settings are ongoing: to treat chronic pulmonary aspergillosis (NCT06794554) and as prophylaxis for invasive fungal diseases in adults undergoing allogeneic bone marrow or stem cell transplantation (NCT04368559).

CONCLUSIONS

Overall, this analysis of data from the global ReSTORE study plus the China extension study confirms the efficacy and safety of rezafungin in the treatment of candidemia and IC in adults. This analysis also supports recently published global practice guidelines recommending echinocandins, including rezafungin, as first-line treatment for candidemia and all forms of IC, except for central nervous system and ocular infections, due to their broad activity and safety profile [15]. With drug choice to be guided by patient-specific pharmacokinetic considerations [15], rezafungin may be an appropriate choice for many patients given its lack of drug–drug interactions and no requirement for dose modifications [33]. Findings also suggest that rezafungin may provide earlier mycological eradication and reduce ICU admission duration compared with caspofungin. Population pharmacokinetic analysis showed there is no clinically meaningful difference in rezafungin exposure between China and non-China cohort patients, and supports the appropriateness of the 400/200 mg rezafungin dose in patients from China.

Supplementary Data

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.