Patients with prolonged, profound neutropenia and allogeneic hematopoietic cell transplant (HCT) recipients are at increased risk of invasive mould diseases (IMD), including invasive aspergillosis (IA) and invasive mucormycosis (IM). In the absence of a microbiological finding (e.g., a positive culture), these two fungal entities are often undistinguishable. Despite the recent availability of new broad-spectrum antifungal drugs and substantial improvements in diagnosing IMD over the last two decades, their management remains complex. At present, the all-cause mortality rate of IM ranges between 40% to 80%, largely depending on underlying conditions and sites of infection. The poorest outcome is observed in patients with hematological malignancies and HCT recipients (Cornely et al., 2019). Disseminated disease, especially involving the central nervous system (CNS), is often associated with mortality rates exceeding 80%. In addition, co-infection with other fungi also increases mortality (Marty et al., 2018). Suspected mucormycosis requires urgent intervention, especially given the rapidly progressive and destructive nature of the infection (Katragkou et al., 2014). Delayed initiation of appropriate therapy is associated with increased mortality.

As with other fungal diseases, the overall survival can be significantly improved by an earlier diagnosis. In addition, and specifically for IM, a multidisciplinary approach involving (in many cases) surgical debridement, control of the underlying disease and the use of polyene antifungals is of paramount importance. However, current diagnostic tools, such as culture, lack sensitivity (Skiada et al., 2013; Cornely et al., 2019). Hence, mucormycosis is often not diagnosed premortem or suspected (too) late during the course of the disease, e.g. when a patient with suspected invasive aspergillosis is not improving while receiving voriconazole therapy (Bergantim et al., 2013).

Recently, several PCR assays for the detection of Mucorales DNA have been developed (Millon et al., 2016; Mercier et al., 2019; Guegan et al., 2020; Skiada et al., 2020). Mucorales PCR detection in plasma or serum samples could be an interesting non-invasive test for screening purposes to trigger early intervention (Guegan et al., 2020).

This study aims to evaluate the performance of a commercially available Mucorales PCR assay (MucorGenius^®, PathoNostics, Maastricht, The Netherlands) as a screening tool on prospectively collected and stored sera from a cohort of at-risk haematology patients. If detection of circulating Mucorales DNA proves to be a good and fast diagnostic test that precedes the final clinical diagnosis, this test could be especially useful as a non-invasive screening tool in patients with unresponsive or relapsing fever or in cases in which sampling for culture or histopathology is not feasible. With an IA co-infection rate of up to 25%, we want to evaluate if there is an underestimated incidence of Mucorales infections in patients with aspergillosis (Millon et al., 2016; Scherer et al., 2018; Guegan et al., 2020).

Therefore, the objectives of this study were to assess the performance of Mucorales-specific PCR testing in high-risk hematology patients, to assess the incidence of IM in these patients, specifically the rate of missed diagnosis, and the difference of IM incidence in patients with IA versus in patients without IA. As a second objective we wanted to compare the survival difference in patients with IA with and without positive Mucorales PCR in the diagnostic sample. Third, in patients with positive Mucorales PCR, we wanted to investigate the kinetics of the Mucorales PCR with or without adequate treatment.

Since early diagnosis of IM has been shown to improve survival, an early diagnostic test may potentially result in a survival benefit. In addition, early confirmation may reduce the number of unnecessary diagnostic test and antimicrobial drugs, thereby reducing hospital cost. Finally, a better understanding of the incidence of Aspergillus and Mucorales co-infections will lead to a better approach towards a patient with an invasive mould disease, with a smaller delay to adequate treatment.

The aim of our study was primarily to assess the incidence of IM as well as the rate of missed cases – or delayed diagnoses – in a large cohort of at-risk hematology patients. In addition, we wanted to assess the incidence of co-infections between IA and IM, and to assess if underlying mucormycosis could explain a possible deterioration under the initial treatment for aspergillosis.

Predisposing factors for both pulmonary mucormycosis and IA in haematology patients are similar and co-infections have been reported in 20-25% of mucormycosis cases (Millon et al., 2016; Scherer et al., 2018; Skiada et al., 2020). It is therefore not surprising that a positive Mucorales PCR is found more frequently in cases of IA than in the control group of high-risk hematology patients. Our results were in line with findings by Guegan et al. where six of 63 patients (9.5%) diagnosed with probable IA (alone), according to EORTC/MSGERC criteria, had at least one positive mucorales PCR when performed on pulmonary samples (BAL, tracheal aspiration, sputum, pleural fluid and lung biopsy) (Guegan et al., 2020).

As per current EORTC/MSGERC definitions (Donnelly et al., 2020), the cases of IA with positive Mucurales PCR would not be defined as cases of IM because culture and/or biopsy was negative. If Mucorales PCR would have been an accepted mycological criterium for diagnosis of probable IM, the four patients would have had the disease. However, its clinical significance remains unclear since mortality was not different between the cases of IA with and those without positive Mucorales PCR. However, these four patients were treated with either Amphotericin B, isavuconazole or posaconazole, drugs with well-known activity against Mucorales (Cornely et al., 2019). In addition, PCR was only positive once. Do we have to see this as an unsignificant ‘bleb’, or does this mean detectable Mucorales DNA which is cleared immediately after start of adequate treatment?

In the retrospective study by Millon et al. time to negativity was 7 (3-19) days (Millon et al., 2016), but this timing seemed to depend more on the availability of samples (no daily sampling available for all of the cases). Moreover, in the cases in which PCR became negative, the Ct-value of the second sample after the start of therapy was usually close the 40 cycles. Cq-values in the study of Millon et al. are similar our results.

Guegan et al. used serum PCR in patients with a positive PCR on pulmonary samples to consider the samples as truly positive (Guegan et al., 2020). PCR was not performed on pulmonary samples of our patients, but in a study by Scherer et al. out of 24 patients with Mucorales PCR positive on BALf, 17 patients had a positive serum PCR and in 15/17 serum was the earliest to become positive (Scherer et al., 2018). An important difference between our results and the results of Guegan et al. is that in their cohort survival between PCR positive and PCR negative patients was significantly different.

False positive PCRs are rare. False positive Aspergillus PCR has been described in up to 5.9% of serum samples, coming from contamination of the sample during processing or from translocation of fungal components from the gut (White et al., 2015; Millon et al., 2019; Egger et al., 2020). The results of the Mucorales PCR in general exceed the Aspergillus PCR results because a higher number of repeats targeted, presence of multiple nuclei providing more fungal DNA and more vascular invasion in IM (Millon et al., 2016). Given the risk profile of our patients, clinicians might consider these positive Mucorales PCR results as indicative of IM missed by conventional cultures, or they might consider this a trigger for starting with an antifungal drug that also covers Mucorales species.

While the EORTC/MSGERC microbiological criteria do not include the results of Mucorales PCR, Aspergillus PCR is included, but it has its limitations. It is proposed to be combined with other antigen-based biomarkers such as GM (Egger et al., 2020). This would not be possible for Mucorales infections. Also, combining positive PCR on BALf would not help to increase specificity.

Chest CT from 3 of the 4 Mucorales PCR positive cases showed more than 10 noduli, which has previously been suggested to be more frequent in IM (Skiada et al., 2013). Other radiological findings that could be more frequent in IM are the reversed halo sign, present in none of the 4 patients [this often presents early in the disease and would probably have already triggered the clinician to cover Mucorales (White et al., 2015)], and pleural effusion, present in 2 out of 4 patients. In addition, previous prophylaxis with voriconazole could prone to IM, however, none of the patients in our cohort received anti-mould prophylaxis or anti-mould treatment ≥2 days before sampling. Also, sinus involvement is more frequent, which was the case in none of the PCR positive patients, but we know in haematology patients lung involvement remains the most frequent presentation. Time between bronchoscopy and chest CT is around 1-2 days in our cohort. Pathology was not performed on these patients, except from cytology on BALf of 1 patient which was normal.

This study probably does not provide enough evidence to implement Mucorales PCR screening. However, with a significant difference of Mucorales infections in cases of IA versus controls (an odds ratio of 17), our findings emphasize once more the importance of considering the possibility of dual mould infections, even in patients with a positive serum or BAL galactomannan detection (Cornely et al., 2019).

This study has its limitations. Other mechanisms that increase the risk of Mucorales infections in patients with IA, apart from the same underlying risk factors, have been described. Breakthrough infections occur under treatment with voriconazole and echinocandins probably because of prolonged treatment, increased survival in these patients, and because voriconazole could increase the virulence of Mucorales (Skiada et al., 2020). Because we only analysed Mucorales PCR at the for IA diagnostic sample, late co-infections could be missed. Breakthrough infections under posaconazole are also possible (Auberger et al., 2012). Analysing samples in a later stage of IA could be interesting, however this timing probably falls outside the scope of an interesting screening window.

To use Mucorales PCR as a screening tool, a high sensitivity and negative predictive value is necessary. This study does not provide sufficient information to assess this. Because our analysis starts from patients diagnosed with IA, we cannot conclude that treatment for IM could have been started earlier, or a treatment switch would be beneficial after the use of Mucorales PCR. Since results from a comparative study on Aspergillus PCR suggest that sensitivity may be higher in plasma than in serum, it is possible that our study underestimated the performance of the Mucorales qPCR, but testing in serum could be more specific (White et al., 2015; Mercier et al., 2019). Finally, the retrospective nature of the study. Performing a prospective project in which patients are monitored with daily or biweekly PCR would provide more information.

Mucorales PCR could be implemented in combination with Aspergillus screening within the global management of targeted patients with a high risk of invasive mold infections. If a patient with IA improves under initial therapy together with a single negative blood Mucorales PCR test result, changing to or adding therapy covering Mucorales can be safely withheld, resulting in a reduction of unnecessary use of antifungal agents and therefore a lower number of patients exposed to potential drug toxicity. However, more information is necessary to decide on the value of this approach.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

This study involving human participants is reviewed and approved by Ethics Committee Research UZ/KU Leuven: S61979 / S65530. The patients/participants provided their written informed consent to participate in this study.

Each author has made substantial contributions to the conception or acquisition of the study or to the analysis or interpretation of the data and substantively revised the work. We describe contributions to the paper using the CRediT taxonomy (Brand et al., 2015). Writing and Original Draft: RA and JM; Review and Editing: AS and KL; Conceptualization: JM; Investigation: RA, SB, and KB; Methodology: RA, KL, and JM; Formal Analysis: RA and JM. All authors contributed to the article and approved the submitted version.

Financial support to perform the PCR testing was provided by Pfizer Inc. However, no conflicts of interest are in place. There is no involvement of the company during the sample collection, measurements and the analysis itself. Also, the necessary test material is not coming from the supporting company and they have no financial benefit in completing the study successfully.

KL received consultancy fees from Pfizer, Abbott, MSD; and SMB Laboratoires Brussels, received travel support from Pfizer and MSD; and received speaker fees from Gilead, MSD, Roche, Abbott. JM received research grants from Merck/MSD, Gilead Sciences and Pfizer; is a consultant to Astellas, Basilea, Bio-Rad, Merck/MSD, Pfizer, Schering-Plough, F2G, Gilead Sciences, Cidara, Scynexis, Amplyx, and Luminex; and served on the speaker’s bureau of Astellas, Gilead Sciences, Bio-Rad, Merck/MSD, Pfizer, Schering-Plough, Basilea, and Viropharma/Shire.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.