Burden of Seasonal Human Coronavirus Infections in Hematopoietic Cell Transplant Recipients.

Introduction

Community respiratory viruses (CRV) are common infectious complications among hematopoietic cell transplant (HCT) recipients and include influenza, parainfluenza virus (PIV), respiratory syncytial virus (RSV), rhinovirus, and human coronavirus (HCoV) [ 1 ]. Infections related to these viruses are associated with grave outcomes such as progression from upper respiratory tract infection (URI) to lower respiratory tract infection (LRI), increased airway resistance, increased oxygen use, and mortality [ 2 , 3 ].

Advances in molecular testing improved the detection of HCoV and other respiratory viruses more consistently [ 4 ]. HCoV has been described as a common cause of respiratory viral infection in HCT recipients and patients with hematologic malignancies with a reported seasonal incidence up to 17% [ 4 , 5 , 6 ]. Seasonal HCoV consists of strains such as OC43, NL63, 229E, and HKU1. The risk of HCoV progression from URI to LRI in HCT recipients can occur in up to 30% [ 4 ]; however, the mortality rate associated with HCoV LRI in HCT recipients was high, around 50% among allogeneic HCT recipients in one study [ 6 ], almost comparable to the reported mortality in HCT recipients with coronavirus disease 2019 (COVID‐19) prior to mass vaccination and access to different treatment options [ 7 , 8 ]. On the other hand, therapeutic options for patients with HCoV URI or LRI are limited to supportive care as no targeted therapies are available.

Numerous host factors have been described as predictors of poor outcomes in HCT recipients with RSV, influenza, or rhinovirus infections [ 9 , 10 , 11 , 12 ]. An immunodeficiency scoring index (ISI) for HCT recipients with RSV infection was derived from host related risk factors for progression to LRI and mortality [ 9 ]. A moderate or high ISI correlated with poor clinical outcomes in HCT recipients infected with other respiratory viruses, such as influenza and parainfluenza [ 13 , 14 , 15 , 16 ] signifying the possible versatility of this scoring index in HCT recipients. Based on the ISI, guidance for management of RSV‐infected HCT recipients who are at high risk for complications have been proposed [ 14 , 17 ]. Furthermore, the ISI could be used to risk‐stratify HCT recipients diagnosed with HCoV to predict progression to severe disease and mortality.

Classifying immunocompromised patients who are at high risk for complications and mortality from HCoV infections is imperative for future therapeutic trials. In this study, we report the clinical characteristics and outcomes of HCT recipients with HCoV infections, including progression to LRI and mortality after risk stratification based on the ISI.

Methods

We performed a retrospective cohort analysis of all consecutive HCT recipients at The University of Texas MD Anderson Cancer Center (Houston, Texas, USA) with laboratory diagnosed HCoV infections between September 1, 2015, and August 31, 2017. We included all adult patients (aged ≥ 18 years), with any type or source of HCT. Baseline characteristics, clinical, laboratory, and outcomes data were collected from the electronic medical records for each episode of HCoV infection. An ISI was calculated for all included patients, using data at the time of HCoV diagnosis (ISI components displayed in Table S1 ). Based on the ISI score, episodes were stratified as low (0–2), moderate (3–7), or high (> 7) risk for worse outcomes. The primary variables of interest were HCoV URI and HCoV LRI (including possible, probable, and progression). Other variables included hospitalization, length of hopsital stay, intensive care unit (ICU) admission, supplemental oxygen use, mechanical ventilation, and 30‐day mortality. Definitions of the study terminology are depicted in Table 1 .

Statistical Analysis

We used medians with interquartile ranges and frequencies with percentages to describe continuous and categorical data, respectively. Categorical variables were compared using chi‐square or Fisher's exact test when appropriate and continuous variables compared using Wilcoxon rank sum test. For the statistical analysis, HCoV LRI included episodes of possible HCoV LRI, probable HCoV LRI, and progression to HCoV LRI. Multivariate logistic regression analysis was used to identify the independent risk factors for LRI HCoV.

To determine an optimum ISI cutoff, we generated a receiver operating characteristic (ROC) curve using LRI as the outcome of interest. A second multivariate analysis was performed using this cutoff, replacing variables that are components of the ISI. A similar analysis was used to determine risk factors for mortality. All tests were two‐sided tests with a significance level of 0.05. We also compared outcomes, such as oxygen use rates, admission rates, length of hospital and ICU stay, and mechanical ventilation rates using Fisher's exact test or Wilcoxon test when appropriate. The statistical analyses were performed using SAS version 9.3 and JMP Pro Version 15 (SAS Institute Inc., Cary, NC).

Ethical Considerations

This study was approved by the University of Texas MD Anderson's institutional review board, and a waiver of patient's consent was granted. All patients' data were stored in an encrypted file and de‐identified for use in the analysis.

Results

Cohort Characteristics

We identified 138 HCT recipients with laboratory confirmed HCoV infections resulting in 164 episodes. Overall, 1466 HCTs were performed at our institution during the study period and the incidence of HCoV infections was 9%. Most episodes occurred during the Fall Winter season (Figure S1 ).

Briefly, most of the patients were White (59%), the most common comorbidities were lung diseases (19%), diabetes mellitus (16%), and heart failure (4%), and 30% of patients had a history of smoking (Table 2 ). The most common indications for HCT were acute myelogenous leukemia (29%) and acute lymphoblastic leukemia (17%). Most patients underwent allogeneic HCT (77%) from a matched donor (56%), and most had myeloablative conditioning regimens (91%) prior to transplantation. The median number of days from HCT to HCoV infection was 329 days (IQR = 92–798). A large proportion of patients were on immunosuppressive therapy (63%) or steroid therapy (41%) at the time of or within 2 weeks of being infected. Other baseline characteristics are displayed in Table 2 .

The most common HCoV serotype was OC43 (34%), followed by NL63 (28%) and 229E (27%); 11 (7%) episodes included 2 serotypes. Coinfection with a second respiratory virus occurred in 48 (29%) episodes. The most common coinfecting viruses were rhinovirus (10%) and RSV (7%). Other infection‐related characteristics are depicted in Table 2 .

Risk Factors for HCoV LRI

Of the 164 episodes of HCoV infections, 129 (79%) and 35 (21%) were at the URI and the LRI stage, respectively. We included six patients who progressed to LRI after initially presenting as URIs in the LRI cohort. On univariate analysis, infections with HCoV prior to engraftment, time from HCT, steroids use within 2 weeks of HCoV infection, neutropenia, lymphopenia, hypoalbuminemia, coinfections with other respiratory viruses, and nosocomially acquired infections were associated with LRI (Table 2 ). No specific serotype was associated with LRI (Figure S2 ).

Validation of the ISI

Of the 164 episodes of HCoV infections, 119 (72%) were classified as moderate, 31 (19%) as low, and 14 (9%) as high risk based on the ISI (Table 2 ). Based on a ROC curve (Figure S3 ), we determined that an ISI score of ≥ 5 was the optimal cut off value to predict LRI (sensitivity of 54%, specificity of 79%, positive predictive value of 40%, and negative predictive value of 87%). When comparing ISI in patients with LRI or URI; more patients with LRI had high ISI (20% vs. 5%, p = 0.0248) (Table 2 ).

We performed a logistic regression analysis to identify independent predictors of HCoV LRI (Table 3 ). In model 1, we included the main risk factors for progression to LRI as separate variables and included neutropenia, lymphopenia, use of steroids, age, time from transplant, time from engraftment, use of myeloablative conditioning, respiratory viral coinfections, hypoalbuminemia, and nosocomial infections. Respiratory viral coinfections and hypoalbuminemia were independent predictors of LRI (Table 3 , model 1). In model 2, we replaced all ISI components with an ISI cutoff of 5. ISI ≥ 5, hypoalbuminemia, and respiratory viral coinfections were independently associated with LRI (Table 3 , model 2).

Outcomes Associated With LRI

Only six patients (4%) died within 30 days of HCoV infections (Table 4 , Figure 1 ). All deaths were due to respiratory failure in HCT recipients who had HCoV LRI. Due to the low number of deaths, further analysis was not possible. In addition, HCoV LRI was associated with worse outcomes in our HCT recipients than in patients with HCoV URI. These outcomes included higher rates of hypoxia that required supplemental oxygen use (30% vs. 3%, p < 0.0001), admission rates (80% vs. 21%, p < 0.0001), ICU admission (37% vs. 2%, p < 0.0001), longer ICU stay (6 days vs. 1 day, p = 0.0024), mechanical ventilation (23% vs. 1%, p < 0.0001), and mortality (17% vs. 0%, p < 0.0001) (Table 4 ).

Discussion

In this large cohort study, we demonstrated that a high ISI (> or equal to 5) is independently associated with HCoV LRI in HCT recipients. We also identified independent predictors of HCoV LRI such as low albumin at the time of infection, and coinfections with other respiratory viruses. In addition, HCoV LRI was associated with several adverse outcomes such as ICU admission, oxygen requirement, mechanical ventilation, and mortality. The utility of the ISI in identifying high‐risk HCT recipients with HCoV infections that may potentially benefit from targeted therapy need to be determined in future clinical trials.

Multiple risk factors for progression to LRI in HCoV‐infected HCT recipients were identified in prior studies [ 4 , 18 ]. Pinana et al. reported that lymphopenia or ALC < 500 cell/mL, active graft‐versus‐host disease, HCoV infection occurring after the first year of transplant, and fever at time of presentation were independent risk factors for HCoV LRI in allogeneic HCT recipients [ 18 ]. Ogimi et al. recently reported that male gender, hypoalbuminemia (< 3 g/dL), hyperglycemia (> 150 mg/dL), presence of respiratory co‐pathogens and high ISI (7–12) as risk factors for occurrence of LRI [ 16 ]. In comparison, we identified similar but fewer risk factors for HCoV LRI in our study and we used an ISI cutoff of ≥ 5 instead of analyzing ISI as a continuous variable as Ogimi et al. [ 16 ]. We also reported a higher mortality rate due to HCoV LRI and respiratory failure when compared to Pinana et al. [ 18 ] (17% vs. 7.4%), but lower to that reported by Ogimi et al. [ 6 ] (17% for 30‐day mortality and 30% for 90‐day mortality).

HCoV LRI has been associated with poor outcomes, besides mortality, in both immunocompetent and immunocompromised patients. HCT recipients and patients with hematologic malignancies with HCoV infections have high rates of mechanical ventilation leading to longer hospital and ICU stays [ 6 ]. We found similar poor outcomes in our cohort with higher hospital and ICU admission rates and higher mechanical ventilation rates. These findings underscore the unmet need when it comes to direct‐antiviral therapy for vulnerable patient population.

In this current study, not all of the ISI components were associated with HCoV LRI; instead, the combination of these risk factors with an ISI of ≥ 5 was a predictor of LRI, especially when it comes to its negative predictive value (87%). This makes the ISI a versatile tool that can be used to stratify HCT recipients based on the combination of multiple risk factors of worse outcomes and with any type of respiratory viral infections for which sparse data are available, such as SARS‐CoV‐2. Prospective studies validating the use of the ISI to predict progression to LRI is needed to verify its clinical utility.

Seasonal HCoV was reported to be a common cause of acute respiratory viral infections in both children and adults prior to the spread of SARS‐CoV‐2 [ 19 ]. HCoV infections can lead to severe infections in older patients with underlying cardiopulmonary disease or immunocompromised conditions [ 5 , 20 ]. In comparison, patients with COVID‐19 presented with a wide spectrum of illness [ 7 , 8 ]. In retrospective studies of cancer patients with COVID‐19, risk factors for hospitalizations included non‐White patients, patients with hematologic malignancies, chronic lymphopenia, steroids use, cardiac comorbidities, and treatment with immune checkpoint inhibitors [ 21 , 22 ]. Among HCT recipients with COVID‐19, attributable mortality is up to 17% and 14% in allogeneic and autologous HCT recipients, respectively, based on pooled analysis [ 8 ]. In comparison, in our cohort of HCT recipients with HCoV infections, 30‐day mortality rate was only 4% overall, with an in‐hospital mortality rate of 11% (6 deaths from 56 admissions).

Our study has few limitations. It is a retrospective study, which makes it difficult to draw exact causal relationships between HCoV LRI and the studied risk factors. There was a limited number of LRI episodes and few cases of documented progression to LRI with subsequent wide confidence intervals noted in our multivariate analysis for LRI in HCoV infected patients. Due to limited number of deaths, we did not perform further analysis to identify independent risk factors for mortality. In addition, this study from a large comprehensive single cancer center may limit the generalizability of the results. However, our results confirm those reported by other centers.

In summary, we identified multiple risk factors for HCoV LRI in HCT recipients and the use of the ISI may determine the patients at risk who may benefit the most from targeted therapy.

Conflicts of Interest

Fareed Khawaja: Research funding paid to his institution from MERCK, Eurofins Viracor, and Symbio Pharma. May Daher and the University of Texas MD Anderson Cancer Center have an institutional financial conflict of interest with Takeda Pharmaceutical. May Daher participates on the scientific advisory board of Cellsbin and Aurigene. Roy F. Chemaly: Consultant/Speaker/Advisor for ADMA Biologics, Janssen, Merck/MSD, Partner Therapeutics, Takeda, Shinogi, AiCuris, Roche/Genentech, Astellas, Tether, Oxford Immunotec, Karius, Moderna, Pfizer, Invivyd, Eurofins‐Viracor, and Ansun Pharmaceuticals and received Research grants paid to his institution from Merck/MSD, Karius, AiCuris, Ansun Pharmaceuticals, Takeda, Genentech, Oxford Immunotec, and Eurofins‐Viracor. The other authors declare no conflicts of interest.

Supporting information

Supplemental Table 1 : Immunodeficiency index components.

Supplemental Figure 1 : Seasonal Distribution of Human Coronavirus Serotypes among Hematopoietic Cell Transplant Recipients between September 2015 and August 2017.

Supplemental Figure 2 : Percentage of Lower (LRI) and Upper respiratory Tract Infections (URI) stratified by Human Coronavirus Serotypes in Hematopoietic Cell Transplant Recipients.

Supplemental Figure 3 : ROC Curve to Determine the Optimal Immunodeficiency Scoring Index Cut‐Off to Predict Lower Respiratory Infection in Hematopoietic Cell Transplant Recipients.

Acknowledgments

We thank Ann Sutton in the Department of Scientific Publications at University of Texas MD Anderson Cancer Center for editorial support.

Khawaja F., Shigle T. L., Yaghi L., et al. “Burden of Seasonal Human Coronavirus Infections in Hematopoietic Cell Transplant Recipients.” Transplant Infectious Disease 27, no. 5 (2025): e70094. 10.1111/tid.70094

Funding : This study was supported in part by the NIH/NCI under award number P30 CA016672 and used Cancer Center Support Grant resources to edit this article.

Contributor Information

Fareed Khawaja, Email: fkhawaja@mdanderson.org.

Roy F. Chemaly, Email: rfchemaly@mdanderson.org.

References

1. Wolfromm A., Porcher R., Legoff J., et al., “Viral Respiratory Infections Diagnosed by Multiplex PCR After Allogeneic Hematopoietic Stem Cell Transplantation: Long‐Term Incidence and Outcome,” Biology of Blood and Marrow Transplantation 20 (2014): 1238–1241.

2. Kim Y.‐J., Boeckh M., and Englund J., “Community Respiratory Virus Infections in Immunocompromised Patients: Hematopoietic Stem Cell and Solid Organ Transplant Recipients, and Individuals With Human Immunodeficiency Virus Infection,” Seminars in Respiratory and Critical Care Medicine 28 (2007): 222–242.

3. Peck A. J., Englund J. A., Kuypers J., et al., “Respiratory Virus Infection Among Hematopoietic Cell Transplant Recipients: Evidence for Asymptomatic Parainfluenza Virus Infection,” Blood 110 (2007): 1681–1688.

4. Eichenberger E. M., Soave R., Zappetti D., et al., “Incidence, Significance, and Persistence of Human Coronavirus Infection in Hematopoietic Stem Cell Transplant Recipients,” Bone Marrow Transplantation 54 (2018): 1058–1066.

5. Ogimi C., Englund J. A., Bradford M. C., Qin X., Boeckh M., and Waghmare A., “Characteristics and Outcomes of Coronavirus Infection in Children: The Role of Viral Factors and an Immunocompromised State,” Journal of the Pediatric Infectious Diseases Society 8 (2019): 21–28.

6. Ogimi C., Waghmare A. A., Kuypers J. M., et al., “Clinical Significance of Human Coronavirus in Bronchoalveolar Lavage Samples From Hematopoietic Cell Transplant Recipients and Patients With Hematologic Malignancies,” Clinical Infectious Diseases 64 (2017): 1532–1539.

7. Sharma A., Bhatt N. S., St Martin A., et al., “Clinical Characteristics and Outcomes of COVID‐19 in Haematopoietic Stem‐Cell Transplantation Recipients: An Observational Cohort Study,” Lancet Haematology 8 (2021): e185–e193.

8. Randi B. A., Higashino H. R., Da Silva V. P., Xavier E. M., Rocha V., and Costa S. F., “COVID‐19 in Hematopoietic Stem‐Cell Transplant Recipients: A Systematic Review and Meta‐Analysis of Clinical Characteristics and Outcomes,” Reviews in Medical Virology 33 (2023): e2483.

9. Shah D. P., Ghantoji S. S., Ariza‐Heredia E. J., et al., “Immunodeficiency Scoring Index to Predict Poor Outcomes in Hematopoietic Cell Transplant Recipients With RSV Infections,” Blood 123 (2014): 3263–3268.

10. Vilar‐Compte D., Shah D. P., Vanichanan J., et al., “Influenza in Patients With Hematological Malignancies: Experience at Two Comprehensive Cancer Centers,” Journal of Medical Virology 90 (2018): 50–60.

11. Kmeid J., Vanichanan J., Shah D. P., et al., “Outcomes of Influenza Infections in Hematopoietic Cell Transplant Recipients: Application of an Immunodeficiency Scoring Index,” Biology of Blood and Marrow Transplantation 22 (2016): 542–548.

12. Piñana J. L., Hernández‐Boluda J. C., Calabuig M., et al., “A Risk‐Adapted Approach to Treating Respiratory Syncytial Virus and Human Parainfluenza Virus in Allogeneic Stem Cell Transplantation Recipients With Oral Ribavirin Therapy: A Pilot Study,” Transplant Infectious Disease 19 (2017): e12729.

13. Damlaj M., Bartoo G., Cartin‐Ceba R., et al., “Corticosteroid Use as Adjunct Therapy for Respiratory Syncytial Virus Infection in Adult Allogeneic Stem Cell Transplant Recipients,” Transplant Infectious Disease 18 (2016): 216–226.

14. Khawaja F. and Chemaly R. F., “Respiratory Syncytial Virus in Hematopoietic Cell Transplant Recipients and Patients With Hematologic Malignancies,” Haematologica 104 (2019): 1322–1331.

15. Wang L., Allen J., Diong C., et al., “Respiratory Virus Infection After Allogeneic Hematopoietic Stem Cell Transplant in a Tropical Center: Predictive Value of the Immunodeficiency Scoring Index,” Transplant Infectious Disease 19 (2017): e12693.

16. Ogimi C., Xie H., Waghmare A., et al., “Risk Factors for Seasonal Human Coronavirus Lower Respiratory Tract Infection After Hematopoietic Cell Transplantation,” Blood Advances 5 (2021): 1903–1914.

17. El Chaer F., Kaul D. R., Englund J. A., et al., “American Society of Transplantation and Cellular Therapy Series: #7 ‐ Management of Respiratory Syncytial Virus Infections in Hematopoietic Cell Transplant Recipients,” Transplantation and Cellular Therapy 29 (2023): 730–738.

18. Pinana J. L., Xhaard A., Tridello G., et al., “Seasonal Human Coronaviruses Respiratory Tract Infection in Recipients of Allogeneic Hematopoietic Stem Cell Transplantation,” Journal of Infectious Diseases 223 (2020): 1564–1575.

19. Killerby M. E., Biggs H. M., Haynes A., et al., “Human Coronavirus Circulation in the United States 2014–2017,” Journal of Clinical Virology 101 (2018): 52–56.

20. Trombetta H., Faggion H. Z., Leotte J., Nogueira M. B., Vidal L. R. R., and Raboni S. M., “Human Coronavirus and Severe Acute Respiratory Infection in Southern Brazil,” Pathogens and Global Health 110 (2016): 113–118.

21. Robilotti E. V., Babady N. E., Mead P. A., et al., “Determinants of COVID‐19 Disease Severity in Patients With Cancer,” Nature Medicine 26 (2020): 1218–1223.

22. Khawaja F., Angelidakis G., Feldman A., et al., “COVID ‐19 in Cancer Patients: The Impact of Vaccination on Outcomes Early in the Pandemic,” Cancer Medicine 12 (2023): 22006–22022.

Associated Data

Supplementary Materials

Supplemental Table 1 : Immunodeficiency index components.

Supplemental Figure 1 : Seasonal Distribution of Human Coronavirus Serotypes among Hematopoietic Cell Transplant Recipients between September 2015 and August 2017.

Supplemental Figure 2 : Percentage of Lower (LRI) and Upper respiratory Tract Infections (URI) stratified by Human Coronavirus Serotypes in Hematopoietic Cell Transplant Recipients.

Supplemental Figure 3 : ROC Curve to Determine the Optimal Immunodeficiency Scoring Index Cut‐Off to Predict Lower Respiratory Infection in Hematopoietic Cell Transplant Recipients.