Rapid spread of Candida auris in China after COVID-19

Dear Editor

A previous report in this journal reported the emergence of drug-resistant Candida auris in Beijing, China.¹ In the current study, we performed a nationwide surveillance of C. auris infections in China and found a dramatic increase in case numbers after the COVID-19 pandemic. As an emerging multidrug-resistant fungal pathogen, C. auris has rapidly spread worldwide and become a global threat to public health since its first description in Japan in 2009.^2,3 C. auris causes not only superficial infections but also life-threatening deep-seated or bloodstream infections.³ More importantly, because of its enhanced abilities to persist in the environment, colonize human skin, and resist common disinfectants, C. auris can easily spread in healthcare settings, thus often leading to hospital-associated outbreaks.³

More than 300 clinical cases have been described in ten province-level regions throughout China by the end of 2023. ^4,5 In this study, we provide an update on the dynamics of its epidemiology after the COVID-19 pandemic. We collected C. auris case information from three major sources: 1) published cases, 2) health departments of local governments, and 3) unpublished cases recorded by our collaborative hospital network. In total, we describe 1846 C. auris clinical cases (including infection and colonization cases) in China through December 2024. These cases were associated with 95 hospitals or aging care institutions in 22 province-level regions of China (Fig. 1). In 2023 and 2024, the numbers of both clinical cases and geographic areas reporting C. auris infections dramatically increased, coinciding with changes in the COVID-19 policy that occurred at the end of 2022 in China (Fig. 1B and C). Annual case counts in China increased by 305% from 130 in 2022 to 527 in 2023, and further increased by 64% to 864 in 2024. The increase in number of cases was most dramatic in mainland China, where annual case counts increased by 697% from 38 in 2022 to 303 in 2023, and subsequently increased by 140% to 727 in 2024 (Fig. 1C). The case counts and numbers of hospitals reporting C. auris cases in provinces throughout China are shown in Fig. 1D. At least six province-level regions reported outbreaks during this timeframe, some of which are experiencing ongoing outbreaks.

Fig. 1.

Number and geographical distribution of C. auris cases in China between 2016–2024. (A) Geographical distribution of C. auris cases through December, 2024. (B and C) Cumulative number of cases (black shadow) and annual increased number of cases (pink) of C. auris in China (B) or in mainland China (C). The total case numbers throughout the year and annual increased case numbers are shown on the top of the corresponding column. The annual increased numbers are indicated with pink font in the brackets. Data from Hong Kong and Taiwan were not included in panel C. (D) Reported numbers of C. auris cases and involved hospitals in different province-level regions: AH, Anhui; BJ, Beijing; CQ, Chongqing; FJ, Fujian; GD, Guangdong; GX, Guangxi; GZ, Guizhou; HaiN, Hainan; HeB, Hebei; HK, Hong Kong; HuB, Hubei; HuN, Hunan; JS, Jiangsu; JX, Jiangxi; LN, Liaoning; SC, Sichuan; SD, Shangdong; SH, Shanghai; SX, Shaanxi; TW, Taiwan; YN, Yunnan; and ZJ, Zhejiang.

Most cases were collected from four regions, namely Guangdong (GD, 838/1846, 45.4%), Hong Kong (HK, 678/1846, 36.7%), Liaoning (LN, 164/1846, 8.9%), and Beijing (BJ, 69/1846, 3.7%). Once infection or colonization cases were identified, active screening for C. auris was performed at the healthcare facilities in HK but not at most hospitals of mainland China. Based on ITS-D1/D2 or genomic sequence analyses of 329 strains, we identified three genetic clades (I, II, and III) of C. auris in China. The majority of these cases were caused by clade I (34.4%) and clade III (64.7%) isolates, while only 3 cases (0.9%) were associated with clade II isolates. GD, BJ, and Shanghai (SH) regions reported C. auris strains encompassing all three genetic clades. Moreover, mixed strain infections caused by clades I and III strains were identified in two hospitals in GD, further complicating accurate diagnosis and treatment.

The ratios of male and female patients infected or colonized with C. auris were 67.8% and 32.2%, respectively. Most infected patients were ICU patients or patients with serious underlying diseases. Patient age ranged from 1 to 96 years old, and the majority (77.7%) of patients were older than 50 years old. The number of bloodstream or deep-seated infection cases were relatively low (9.0% for bloodstream, 3.6% for venous catheter-related, and 5.3% for drainage-related infections), while superficial infection or colonization cases were dominant in this collection (26.5% for urine, 25.5% for sputum and oral swabs, 7.1% for feces and rectal swabs, and 5.3% for bronchoalveolar lavage fluid (BALF)-associated samples).

To reveal the genetic population structure and transmission features of C. auris in China, we analyzed the genomic sequences of 81 representative C. auris strains from different regions. Of these 81 C. auris strains, whole genome sequences for 52 strains were retrieved from published materials, while 29 strains were sequenced in this study (Fig. 2). Based on the genomic epidemiology analysis using strains from China and strains collected internationally (Figs. 2, S1, S2 and S3), several major findings were made. First, multiple introductions of C. auris from other countries into China and multiple local transmission events were observed. Five major clusters of clade I isolates and two clusters of clade III isolates were found (Figs. S1 and S2). The first C. auris case identified in GD likely acquired the pathogen abroad since the patient had at the time, recently traveled internationally (Fig. S3). The LN outbreak strains (clade III) were phylogenetically closely related to the strains from Australia, and could be among the earliest strains introduced to China (Fig. S2). Three clusters of clade I strains were found in Sichuan (SC), suggesting these strains were introduced from different sources. One patient in SC could acquire the C. auris strain during medical treatment in Egypt (Fig. S3). Second, multiple inter-hospital and inter-province transmissions of C. auris occurred. Inter-hospital patient transfers could be a major contributor to trans-regional transmissions. For example, the clade I strains isolated in a hospital in SH were associated with strains identified in Anhui (AH) and Jiangsu (JS) provinces (Fig. S2), whereas the clade III strains from the same hospital exhibited close phylogenetic relationships with the GD strains (Fig. S1). Third, the clade II strains had not circulated in China since the three sporadic cases were isolated from different regions and were genetically unrelated (Fig. 2). All the three clade II strain cases were associated with ear infections and no epidemiological links were found among the three cases.

Fig. 2.

Maximum-likelihood phylogenetic tree constructed based on genomic sequences of representative C. auris isolates from different regions of China. A total of 79,471 high-confidence SNPs across 81 isolates were used for phylogenetic reconstruction. The transversion substitution model (TVM) was statistically validated as the optimal nucleotide replacement pattern through Bayesian Information Criterion (BIC) comparisons in ModelTest-NG. Topological robustness was assessed via 1000 nonparametric bootstrap replicates. Three genetic clades (I, II, and III) were identified in China. Strains from different provinces are indicated in different colors. Three strains shown in black were isolated from FJ, GD, and JX. AH, Anhui; BJ, Beijing; FJ, Fujian; GD, Guangdong; HK, Hong Kong; JS, Jiangsu; JX, Jiangxi; LN, Liaoning; SC, Sichuan; SH, Shanghai; and ZJ, Zhejiang.

Since C. auris clinical strains are often resistant to multiple antifungal drugs, we next analyzed the drug susceptibilities of 507 representative isolates across 13 provinces in China (Table S1). Based on the published data and analysis of these representative C. auris clinical isolates, we found that the vast majority (98.4%) were resistant to fluconazole (MIC ≥ 32 mg/L). A subset of strains showed a high MIC value to voriconazole (7.9%, MIC ≥ 4 mg/L), amphotericin B (39.1%, MIC ≥ 2 mg/L) and echinocandins (4.1%, MIC ≥ 4 mg/L). All clade I strains from AH and JS were resistant to amphotericin B, suggesting the spread of resistant lineages in these regions. The ratio of amphotericin B-resistant or echinocandin-resistant cases remarkably increased in the past two years. Genomic sequencing analysis indicated that fluconazole resistance of C. auris strains was often associated with the Y132F and VF125AL “hotspot” mutations in ERG11, which encodes a lanosterol 14-α-demethylase, the target of azoles. The echinocandin-resistant isolates often carried the mutations S639F, S639Y, S639P, or W691L in the gene FKS1, which encodes a β-1,3-glucan synthase, the target of echinocandins. These hotspot mutations have been previously reported in the literature.⁶ Moreover, we identified several other mutations in the genes TAC1B, CDR1, and MDR1 that could also contribute to antifungal resistance in clinical strains of C. auris.

In summary, the present study provides an update on the prevalence and genomic epidemiology of C. auris in China. We found that this emerging fungal pathogen has rapidly spread in China and we observed a marked increase in infection cases particularly after the COVID-19 pandemic. The increase in international traveling, changes made to disinfection protocols, and increased patient risk factors could have all contributed to the epidemiological dynamics we observed for C. auris infections in China in this timeframe. Consistent with our previous report, most cases were identified in the eastern regions of China. Multiple outbreaks occurred and are still ongoing in several provinces. Since active screening was not performed in most healthcare facilities (especially in mainland China), the actual case numbers could be heavily underestimated.

Our study has some notable limitations. First, many large hospitals were not included in our surveillance. Second, active screening was not performed in most hospitals and some cases were not included due to delays in annual reports. Third, in some instances, it was difficult to distinguish infection cases from colonization cases in clinical settings. Given the dramatic increase in C. auris infection cases globally in hospital settings over the past two years, our study highlights the need for rapid detection and continued surveillance in healthcare facilities. To mitigate further rapid transmission and to prevent additional hospital outbreaks, infection control measures such as the early and accurate identification of C. auris through screening measures, and enhanced disinfection protocols must be implemented.

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jinf.2025.106476.