Infectious Disease Alert Updates
June 1, 2023
Related Articles
-
Infectious Disease Updates
-
Noninferiority of Seven vs. 14 Days of Antibiotic Therapy for Bloodstream Infections
-
Parvovirus and Increasing Danger in Pregnancy and Sickle Cell Disease
-
Oseltamivir for Adults Hospitalized with Influenza: Earlier Is Better
-
Usefulness of Pyuria to Diagnose UTI in Children
By Carol A. Kemper, MD, FIDSA
Medical Director, Infection Prevention El Camino Hospital, Palo Alto Medical Foundation
Immunocompromised Patients with Persistent COVID-19
SOURCE: Snell LB, Alcolea-Medina A, Charalampous T, et al. Real-time whole genomic sequencing to guide patient-tailored therapy of severe acute respiratory syndrome coronavirus 2 infection. Clin Infect Dis 2023;76:1125-1127.
The treatment of immuncompromised patients with persistent or relapsing COVID-19 infection remains uncertain. Data for immunocompromised hosts (ICHs) with COVID simply are not available since most such patients were excluded from clinical trials. Formal guidelines recommend the use of antiviral drugs and immunotherapies at the dosages and durations recommended for the general population, which is clearly inadequate for some patients. For example, reports suggest remdesivir (RDV) can suppress but not always eliminate viral replication in ICHs, and RDV resistance has been observed in transplant patients with persistent COVID-19 infection. Anecdotal reports describe the use of combined antiviral and immunotherapies, as well as extended courses of remdesivir for 10-30 days for some ICHs with persistent infection, but infectious disease specialists are, to a large degree, functioning in the dark with such cases.
What is clear is that such patients exist, and as the COVID pandemic has “settled” into a more routine infection for most healthy vaccinated individuals, a greater proportion of those requiring hospitalization are ICHs with suspected persistent/relapsing infection and respiratory failure. Patients receiving therapies that suppress B and T cell function appear to be at the greatest risk for inability to clear replicating SARS-CoV-2 virus.
In my experience, those agents that seem to be more frequently involved in such cases especially include the anti-CD20 biologics (e.g., rituximab, and ocrelizumab for multiple sclerosis), but also tumor necrosis factor (TNF) inhibitors (e.g., etanercept, adalimumab, infliximab); and immunosuppressives used in transplantation, especially mycophenolate and cyclosporine when used in combination with prednisone. General immunosuppressants, such as prednisone, methotrexate, and azathioprine, when used as single agents, seem to confer a lesser risk for persistent COVID-19 infection. Importantly, the effect of rituximab may last six to 12 months following treatment.
These authors described the benefits of real-time application of whole genomic sequencing (WGS) for guiding treatment decisions in a series of complex patients with SARS-CoV-2 infection — including the use of real time WGS for detecting resistant variants. Four different scenarios were described:
1. Persistent infection. A 59-year-old man with kidney transplantation on anti-CD20 treatment presented in December 2020 with SARS-CoV-2 infection — and again in February 2021 — and again in January 2022. On each occasion, WGS demonstrated the same B.1.177.18 strain, which had ultimately “migrated” by 18 single nucleotide polymorphisms (SNPs) from the original strain over a period of 13 months, consistent with the occurrence of approximately one spontaneous mutation every two weeks. He was treated with REGN-COV2 (casirivimab/imdevimab) because this strain was known to be susceptible to this agent and cleared his infection. In a second case, a 50-year-old woman with a thymoma developed recurrent COVID infection in February 2022, requiring treatment with the usual course of remdesivir and dexamethasone. She continued to require oxygen and subsequently was rehospitalized for suspected lung fibrosis in April 2022. Repeat positive nasopharyngeal polymerase chain reaction (PCR) tests were thought to be due to non-viable shedding. However, when she presented again in June 2022, both the first and third samples, obtained more than four months apart, were confirmed Omicron BA.1.1 sublineage, strongly suggesting chronic infection. She was treated with sotrovimab and Paxlovid, cleared her infection, and no longer required oxygen.
2. Recurrent infection with differing strains. A 44-year-old man with combined immunodeficiency developed SARS-CoV-2 infection in March 2022 and was treated with sotrovimab. Two months later, he remained PCR positive, although lateral flow testing was negative. He then developed recurrent symptoms with fever and cough, and lateral flow testing became positive. Although his physicians initially planned to treat him more aggressively for chronic infection, WGS confirmed the strains from March and June were different Omicron sublineages (BA.2 and BA.5.2), indicating two separate infections. He received sotrovimab and recovered.
3. Detection of SARS-CoV-2 spike mutations. A 47-year-old woman with primary sclerosing cholangitis awaiting liver transplantation developed COVID-19 infection. WGS identified BA.2.12.1, which has diminished susceptibility to monoclonal therapies. Drug interactions precluded the use of Paxlovid, and she instead received a double dose of sotrovimab (1,000 mg) and recovered. In another interesting case, a 45-year-old woman with advanced human immunodeficiency virus (HIV) infection had chronic, asymptomatic SARS-CoV-2 infection. She was treated with REGN-COV2 (casirivimab/imdevimab). While WGS of a pre-treatment sample showed no spike mutations, sequences obtained 19 days later demonstrated a spike G476D substitution, which confers reduced sensitivity to casirivimab by > 1,000-fold, as well as a spike V445A mutation, conferring reduced susceptibility to imdevimab. As a result, she was treated with Paxlovid and HIV therapy and became PCR-negative.
4. Detection of other resistance mutations. A 60-year-old man receiving obinutuzumab for chronic lymphocytic leukemia developed respiratory failure with SARS-CoV-2 infection in April 2022. He later was admitted and received Paxlovid and double dose sotrovimab, and recovered sufficiently to be discharged, although PCRs remained positive for four months. He relapsed with respiratory failure, and WGS demonstrated that two separate samples, taken 10 weeks apart, showed the same BA.2.3 infection with a 7 SNP difference. There was evidence of a new spike K356R substitution that has been observed following sotrovimab therapy without affecting its apparent efficacy. WGS found no evidence of Paxlovid resistance on the nsp5 gene, or evidence of remdesivir resistance in the ribonucleic acid (RNA)-dependent RNA-polymerase gene, indicating that either Paxlovid or remdesivir could reasonably be used to treat this ICH with chronic infection.
Real-time WGS-directed therapy provides a wealth of information, allowing smart decisions in the selection of COVID therapeutics. Of course, the first problem for many of us clinicians laboring in the community becomes the identification of such patients, and how to distinguish those with chronic infection from lung fibrosis or other infections — or from a new COVID-19 infection. This report describes one patient with chronic infection for 13 months, whereas another ICH with apparent persisting infection for three months actually had a second infection, the opposite of what you might guess without the benefit of strain typing. How helpful this information would be when weighing treatment options. Improved methods for the rapid detection of replicating virus and remdesivir resistance are urgently required. We also have observed patients with repeatedly (and deceptively) negative COVID-19 nasopharyngeal PCR testing — but bronchoalveolar lavage samples — or autopsy specimens — have demonstrated persistent virus in the lungs and other organs.
Endemic Mycoses Are No Longer ‘-Demic’
SOURCE: Mazi PB, Sahrmann JM, Olsen MA, et al. The geographic distribution of dimorphic mycoses in the United States for the modern era. Clin Infect Dis 2023;76:1295-1301.
I remember my friend and fellow Stanford infectious disease fellow, the late Richard M. Tucker, MD, saying he saw more coccidioidomycosis after moving to Wenatchee, WA, than he ever did in San Jose, CA. Both his “snow bunny” patients escaped Washington winters by vacationing in Arizona and returned home with cocci infection and migrant workers and their families spent their spring and summers picking strawberries and lettuce in California’s Central Valley, only to develop acute cocci infection as they arrived to pick apples in central Washington in the fall. These imported infections were acquired in their known areas of endemicity, although cocci now has been identified as far north as eastern Utah and south central Washington, possibly due to the effects of climate change.
The root “-demic” is Greek for “of or belonging to a place.” However, the endemic mycoses (coccidioidomycosis, blastomycosis, and histoplasmosis) are no longer steady and predictable in their distribution as people with these infections now are being diagnosed with increasing frequency outside of the usually defined areas of risk. Such cases often are the result of travel but also may occur as the result of reactivation, latent infection in immunocompromised hosts, or exposure to fomites. There also is an increasing body of evidence that these three dimorphic mycoses are occurring far outside their known endemicity.1
These authors performed a retrospective analysis of more than 45 million Medicare fee-for-service patients, searching for a diagnosis of one of the three “endemic mycoses” (based on ICD-9/10 coding) between January 2007 and December 2016. During this nine-year period, there were 79,749 unique diagnoses of histoplasmosis, 37,726 unique diagnoses of coccidioidomycosis, and 6,109 unique diagnoses of blastomycosis. The primary outcome was the incidence of each of the three mycoses for each county in the United States. A clinically meaningful threshold was defined as at least 100 diagnoses/100,000 person years (PY) for histoplasmosis and coccidioidomycosis and 50 diagnoses/100,000 PY for blastomycosis.
The resulting revised geographic distributions for each of these infections is startling. Including Washington, DC, 48/51 (94%) of the states had one or more counties that met the primary endpoint for histoplasmosis; 35/51 (69%) of the states had at least one or more counties that met the primary endpoint for coccidioidomycosis, and even more surprising, 40/51 (78%) of the states had at least one county that met the endpoint for blastomycosis. The resulting maps of affected counties throughout the United States look like continental checkerboards, especially when compared with the similar maps from 1955. Of the three dimorphic fungi, histoplasmosis has always been less geographically restricted, occurring in other parts of the world, probably wherever the organism favorably finds a niche in soil containing bat or bird droppings.
While still dominant in states bordering the Ohio and Mississippi River valleys, and around the Great Lakes and Saint Lawrence River, histoplasmosis now commonly occurs throughout the eastern half of the United States, from the most northern New England counties to the Southeastern seaboard, and especially Florida. But what is more surprising is the number of Western states with five to 250 cases of histoplasmosis/100,000 PY, including counties in Washington, Oregon, and California, and with scattered counties throughout the Southwest, especially in Arizona and New Mexico. Perhaps clinical investigations for cocci in these states will uncover histoplasmosis instead. Documentation of histoplasmosis in domestic and wild animal species acquired far outside the common disease-endemic area confirms this broader geographic distribution of infection.
In contrast, while the distribution of coccidioidomycosis cases in 1955 was virtually limited to central California and the Sonora life zone extending into Arizona, New Mexico, and the southwestern border of Texas, cocci diagnoses in 2007-2016 extend well beyond these historical areas of risk. While the densest distribution of cocci diagnoses remains southern/central California and Arizona, cases now can be found in many Western, Midwestern, and Eastern states, including many counties bordering the Atlantic. The area around Duluth, MN, as well as a few counties in Montana and Colorado have upwards of 100-250 cases/100,000 PY. Northern Minnesota might be explained if the physicians there are looking more intently for blastomycosis — or perhaps they are simply more experienced in recognizing clinical syndromes of endemic mycoses in winter travelers.
Blastomycosis, also, now is diagnosed almost as frequently in California, Arizona, and New Mexico in 2007-2016 as it was in historically endemic areas — including many East Coast counties and counties bordering the Atlantic, extending its range all the way from Maine to Florida, and to the southern tip of Texas.
The authors contend that the word “endemic” to delineate specific areas of contagion for these three infections is misleading — and leads directly to delays in diagnosis and missed diagnoses. Data suggest that 46% of cocci diagnoses and 40% of blastomycosis diagnoses are delayed by more than one month. According to the authors, the most common cause of delayed diagnosis is failure to consider these infections in the differential. How or why these three infections are being diagnosed outside of their historically defined areas can be debated, but, to a large degree, is irrelevant. It is not where you acquired infection (sometimes decades earlier), but where you are when you present for medical care.
These data indicate we need to reframe our thinking and learn to consider the possibility of these three infections really anywhere in the United States and to recognize their clinical presentations, especially in patients with common risk factors, such as diabetes, chronic renal and pulmonary disease, and immunosuppression.
Limitations to this study include the focus on Medicare patients, who are older and at greater risk for these three infections. Also, the county of residence was used, rather than the county of diagnosis, which may have skewed the data toward nonendemic counties.
REFERENCE
- Benedict K, Thompson GR, Deresinski S et al. Mycotic infections acquired outside areas of known endemicity, United States. Emerg Infect Dis 2015;21:1935-1941.
Immunocompromised Patients with Persistent COVID-19; Endemic Mycoses Are No Longer ‘-Demic’
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.