March 2019: Onc and transplant ID

Invasive pulmonary aspergillosis is common in patients with a new diagnosis of acute myeloid leukemia (AML).  The authors prospectively studied adults with AML admitted to their hospital for induction chemotherapy, adults with relapsed AML admitted for salvage therapy, and adults admitted for allogenic stem cell transplant over a 3-year period.  These patients received a baseline chest CT, and those who had radiographic findings suggesting invasive fungal disease (e.g. pulmonary nodules, alveolar or ground glass opacities) underwent bronchoscopy with BAL to identify a potential pathogen.  BAL testing included fungal stain and culture, galactomannan testing, and Aspergillus PCR; invasive pulmonary aspergillosis (IPA) was defined by Mycoses Study Group criteria.

Of the 285 patients included in the study, 91 had abnormal baseline pulmonary imaging, 49 underwent bronchoscopy and BAL for suspicion of IPA, and 15 (5%) were diagnosed with IPA on admission.  A subsequent 29 patients (10%) were diagnosed during their hospitalization, meaning that IPA complicated 15% of all admissions in this population. 

By subgroup, IPA occurred most often in patients with a new diagnosis of AML (20%), then those with relapsed AML (17%), then those undergoing SCT (4%).  Patients with a new diagnosis of AML and IPA at baseline had had a median 14 days of symptoms related to their cancer prior to the admission, and 4/11 had had fever in the week before admission.  The median neutrophil count on admission was 900; only 3/11 were neutropenic using an ANC cutoff of 500 cells/ul.  Patients with IPA diagnosed at baseline versus during their hospitalization were more likely to have a newly diagnosed malignancy (73% vs 31%; p=0.008), to be older (median 64 vs 51 years; p=0.004), and to be admitted for chemotherapy rather than transplant (80% vs 45%; p=0.07).  That patients with hematologic cancer might have IPA at baseline didn’t surprise me – people have AML before their AML is diagnosed, so you ought to see invasive fungal infection in a newly diagnosed AML patient from time to time in the same way that we simultaneously diagnose people with HIV and opportunistic infections.  But this circumstance was much more common than I’d expected.  30855077

The MMR vaccine is safe in liver transplant recipients who are minimally immunosuppressed at least one year after transplantation.  The ID and transplant communities have traditionally been reluctant to give live vaccines to transplant patients on account of concern that clinically significant infections might develop given the patients’ immunosuppression.  While not an unreasonable concern, we have to balance that risk against the (ever-increasing) risk of the patient developing a vaccine-preventable disease.

This finding comes from a retrospective cohort study performed in pediatric liver transplant recipients from Switzerland.  The authors examined the outcome of MMR vaccination in children who had undergone liver transplantation and who had prevaccination measles IgG levels <150 IU/ml (the threshold for seroprotection).  The patients had been transplanted a mean 6.3 years (IQR 4 to 11 years) prior, and were revaccinated up to 2 times if they failed to produce serologic responses or subsequently lost those responses after the initial vaccine.  Vaccine responses occurred in 98%, with 62% achieving seroprotection by 1 year and 89% by 3 years.  All vaccinations were well-tolerated and no serious adverse events were attributed to vaccination. The authors conclude that a subset of pediatric liver transplant patients who are at least a year out from transplant and who are receiving minimal immune suppression can safely receive and will likely benefit from the MMR. 30171797

Cryptococcosis occurs in immunosuppressed people without HIV, often involves the CNS in such cases, and is associated with substantial sequela. The authors performed a prospective observational study in which they longitudinally followed patients without HIV who developed cryptococcal disease across 25 medical centers.  Cryptococcal infection was diagnosed by either antigen in blood or CSF or organism detection in culture or on biopsy.  At 3mo and then 6mo intervals for up to 2 years, patient took a health survey, underwent cognitive testing, and had blood, urine, and saliva samples collected for further testing.

Of 152 patients who consented to the study, 121 completed longitudinal followup.  Most patients were male, lived in the Southern US, and either engaged in some sort of outdoor work or hobby (e.g. landscaping, family, hiking, or horseback riding) or had exposure to birds.  The majority had immune suppression due to solid organ transplantation (34%), autoimmune disease (16%), hematologic cancer (12%), or cirrhosis (10%); of those receiving immunosuppressives, steroids (48%) and cytotoxic chemotherapy (41%) were most common. 

Fever was an uncommon presenting manifestation, present in only 28% of patients. While pulmonary disease was the most common manifestation of cryptococcus in this cohort, half of patients had CNS disease, and many of these had no radiographic evidence of pulmonary involvement.  Among patients with CNS disease, gait disorders, headache, and hearing deficits were the most common presenting complaints.  Five patients had infection due to C. gatti (three from the Pacific Northwest – where this organism was once alleged to be confined – and two from the American southeast) and four of these patients had no underlying disease.

Longitudinal surveys indicated a poor quality of life for immunosuppressed patients after cryptococcal infection, particularly in those with CNS disease.  MoCA testing showed that CNS cryptococcus was associated with impaired cognition at diagnosis and that moderate to severe cognitive impairment was associated with need for multiple LPs and a poor long-term prognosis. On the other hand, patients who presented with mild or no cognitive impairments had few neurologic sequela.  Of patients with CNS disease, 35% needed multiple LPs to manage intracranial pressure, and 11% ultimate required VP shunting or ventriculostomy. Median antifungal therapy in this population was 197 days for patients without CNS disease and 261 days for patients with CNS disease.  More than half of patients had multiple cryptococcal antigen tests sent and some had positive test results more than 3 months into therapy; I thought this was interesting, because as far as I am aware the data do not support any prognostic benefit of trending cryptococcal antigens.

Higher mortality risk within the cohort was most associated with pretreatment serum antigen titers >1:128 (HR 2.3; p=0.078) and CNS disease (HR 1.8; p=0.09). The main protective factor was prior SOT (HR 0.22; p=0.004), which makes sense if you consider that the patient’s primary underlying problem leading to cryptococcal infection -  cirrhosis – has just been fixed with an aftermarket part.  The Kaplan-Meier curves show that most deaths in the cohort occurred within the first 60 days, and nearly all within the first six months.  30855688