Lymphomatous involvement of the larynx is a rare entity. We present a case of atypical laryngotracheitis as the initial manifestation of non-Hodgkin’s lymphoma in a pediatric patient. The diagnosis was aided through the use of microbial cell-free DNA (mcfDNA) testing, which detected the presence of Epstein-Barr virus in the patient’s plasma. This enabled the consideration of an Epstein-Barr virus–related lymphoproliferative process, leading to additional workup and the final diagnosis of lymphoma. To our knowledge, this is the first case of mcfDNA testing leading not simply to an infectious organism, but further to a new oncologic diagnosis. Plasma mcfDNA testing has the potential to inform clinical practice beyond classic infectious disease manifestations. In this article, we review both the possible future applications and the areas of further investigation that remain.
Lymphoma is the third most common pediatric neoplasm in children, with non-Hodgkin’s lymphoma encompassing nearly half of the newly diagnosed cases in the United States each year.1 However, laryngeal lymphoma is a rare occurrence, with nonepidermoid cancer of the larynx overall representing no more than 5% to 7% of laryngeal neoplasms.2
Here, we present a case of atypical laryngotracheitis as the initial manifestation of diffuse large B-cell lymphoma (DLBCL), a subtype of non-Hodgkin’s lymphoma, in a pediatric patient. The final diagnosis was aided through the use of next-generation sequencing of microbial cell-free DNA (mcfDNA) from the patient’s plasma. This test led to the identification of Epstein-Barr virus (EBV) in the patient’s plasma and subsequent modification of the differential diagnosis to include an EBV-related lymphoproliferative process. This helped the final pathologic diagnosis of EBV-positive DLBCL.
In recent times, mcfDNA testing (in one form commercially available as the Karius test) has enabled the diagnosis of various unrecognized infections, particularly in immunocompromised or otherwise medically complex hosts, through the DNA detection of pathogens in the bloodstream. To our knowledge, this is the first case of mcfDNA leading not simply to an infectious organism but further to a new oncologic diagnosis. We discuss herein the implications of the rapid availability of such genomic and metagenomic data for the practicing clinician.
Case Description
A 7-year-old boy with a history of posterior fossa and spinal medulloblastoma status post resection and chemoradiation, presented ∼1 month after the completion of the fourth cycle of chemotherapy with hoarseness, barky cough, and fever. The patient tested positive for adenovirus and rhinovirus and was conservatively managed for a presumed diagnosis of croup by the general pediatrics service. Otolaryngology was subsequently consulted in the setting of increasing stridor and work of breathing despite initial measures. On the initial flexible laryngoscopy, there was evidence of supraglottic exudate and glottic edema and debris (Fig 1). He was diagnosed with bacterial laryngotracheitis, which was thought to be secondary to the initial viral upper respiratory infection. His symptoms did not improve despite intravenous antibiotics, for which reason he required multiple rounds of operative laryngoscopy and bronchoscopy for debridement. A pathologic review of the operative specimens predominantly revealed tissue necrosis. Cultures from each debridement were polymicrobial, and he was treated with intravenous antibiotics.
Flexible fiberoptic laryngoscopy in the patient with stridor and increased work of breathing revealed supraglottic exudate and glottic edema and debris, initially suggestive of bacterial laryngotracheitis.
Flexible fiberoptic laryngoscopy in the patient with stridor and increased work of breathing revealed supraglottic exudate and glottic edema and debris, initially suggestive of bacterial laryngotracheitis.
Approximately 1 month into his course, he was noted to have a tender right cervical mass. Contrast-enhanced computed tomography revealed a rim-enhancing fluid collection involving the right tonsillar soft tissues extending posteriorly to involve the right pharyngeal and retropharyngeal space, with abutment of the carotid space as well as narrowing of the right internal jugular vein (Fig 2A). There were also enlarged right cervical lymph nodes, predominantly in right level II (Fig 2B). Furthermore, additional imaging revealed multiple nodules in the brain, lungs, and liver. He underwent biopsies of the oropharynx and nasopharynx by otolaryngology. Initially the surgical pathology resulted as tissue necrosis with mixed inflammation.
Contrast-enhanced computed tomography scan of the soft tissues of the neck revealed the following findings. A, A rim-enhancing fluid collection involving the right tonsillar soft tissues extending posteriorly to involve the right pharyngeal and retropharyngeal space, with abutment of the carotid space as well as narrowing of the right internal jugular vein; swelling of the glottis and supraglottic soft tissues with patent airway. B. Enlarged right cervical lymph nodes, predominantly in right level II.
Contrast-enhanced computed tomography scan of the soft tissues of the neck revealed the following findings. A, A rim-enhancing fluid collection involving the right tonsillar soft tissues extending posteriorly to involve the right pharyngeal and retropharyngeal space, with abutment of the carotid space as well as narrowing of the right internal jugular vein; swelling of the glottis and supraglottic soft tissues with patent airway. B. Enlarged right cervical lymph nodes, predominantly in right level II.
The patient did not have a classic presentation for bacterial laryngotracheitis, especially considering the minimal response to antibiotics. In addition, given the presence of multiple lesions, concern remained for a complicated infection versus other systemic processes, such as malignancy. In consultation with immunology, the patient was tested comprehensively for immunologic deficiencies and tumor syndromes, but no pathogenic variants were found. The decision to send plasma for mcfDNA testing was made by the interdisciplinary team. The turnaround time for test results was <1 business day. Test results returned positive for, in descending order by DNA molecules per microliter, cytomegalovirus (CMV), EBV, Stenotrophomonas maltophilia, BK polyomavirus, human adenovirus C, Staphylococcus epidermidis, and Escherichia coli (Table 1). Despite the CMV viremia, there was no evidence of end-organ damage related to CMV. The patient was already known to carry both adenovirus and multidrug-resistant Stenotrophomonas, for which he had been previously treated. The coagulase-negative Staphylococcus was most likely a skin contaminant. The significance of BK polyomavirus and E coli was unknown, but the presence of neither pathogen was congruent with the clinical picture. The most salient finding was the unusually high signal for EBV. Subsequent confirmatory EBV polymerase chain reaction testing confirmed the presence of 431 600 copies in the plasma, which increased suspicion for an EBV-related lymphoproliferative process.
Report of mcfDNA Testing
Microorganism Name . | DNA, Molecules per μL . | Reference Interval, Molecules per μL . |
---|---|---|
CMV | 28 575 | <10 |
EBV | 4307 | <10 |
S maltophilia | 4060 | <63 |
BK polyomavirus | 1007 | <10 |
Human adenovirus C | 249 | <10 |
S epidermidis (coagulase-negative Staphylococcus) | 184 | <17 |
E coli | 149 | <15 |
Microorganism Name . | DNA, Molecules per μL . | Reference Interval, Molecules per μL . |
---|---|---|
CMV | 28 575 | <10 |
EBV | 4307 | <10 |
S maltophilia | 4060 | <63 |
BK polyomavirus | 1007 | <10 |
Human adenovirus C | 249 | <10 |
S epidermidis (coagulase-negative Staphylococcus) | 184 | <17 |
E coli | 149 | <15 |
On account of this new finding, the histopathology of the oropharyngeal and nasopharyngeal biopsies was reviewed by pathology. Tissue from the peritonsillar region revealed the presence of atypical monomorphic large cells that stained positive for EBV by in situ hybridization and CD20 (a B-cell–specific marker) by immunohistochemistry. A subsequent liver biopsy then confirmed the diagnosis of EBV-positive DLBCL. Figure 3 reveals the pathologic specimens corresponding to peritonsillar skeletal muscle and liver, revealing the lymphomatous involvement and presence of EBV.
Peritonsillar skeletal muscle and liver involved by EBV-positive DLBCL. A, Skeletal muscle fibers infiltrated by a dense infiltrate of enlarged and irregular lymphoid cells and tissue necrosis (H&E, ×200). The inset is a high-power view of skeletal muscle and lymphoid cells (H&E, ×400). B, Liver needle core biopsy specimen with necrosis and enlarged and irregular lymphoid cells morphologically analogous to those present within the peritonsillar skeletal muscle (H&E, ×200). The inset is a high-power view of lymphoid cells within the liver (H&E, ×400). C, EBV in situ hybridization staining of atypical lymphoid cells within the liver (×200). H&E, hematoxylin and eosin.
Peritonsillar skeletal muscle and liver involved by EBV-positive DLBCL. A, Skeletal muscle fibers infiltrated by a dense infiltrate of enlarged and irregular lymphoid cells and tissue necrosis (H&E, ×200). The inset is a high-power view of skeletal muscle and lymphoid cells (H&E, ×400). B, Liver needle core biopsy specimen with necrosis and enlarged and irregular lymphoid cells morphologically analogous to those present within the peritonsillar skeletal muscle (H&E, ×200). The inset is a high-power view of lymphoid cells within the liver (H&E, ×400). C, EBV in situ hybridization staining of atypical lymphoid cells within the liver (×200). H&E, hematoxylin and eosin.
His EBV-positive DLBCL and florid EBV and CMV viremia was concerning for underlying primary immunodeficiency, especially in the setting of lymphopenia with no B cells and reduced immunoglobulins (immunoglobulin G level, nadir 264 mg/dL). Results of the primary immunodeficiency genetic panel were negative for any pathologic variants, including SH2DIA, XIAP, and CD27, which ruled out X-linked lymphoproliferative disease 1 and 2 and CD27 deficiency, which are associated with EBV lymphoproliferation. His T-cell receptor repertoire was largely normal. Additionally, his T cells had a normal response to mitogens. Although he had no B cells, he did have protective titers to diphtheria and tetanus, which suggested that he had B cells before. This, in addition to the presence of tonsillar tissue, made X-linked agammaglobulinemia unlikely.
The previous presence of medulloblastoma status post radiotherapy, with such rapid occurrence of a secondary malignancy, raised concern for possible radiation-sensitive immunodeficiency, such as ataxia telangiectasia; however, his α-fetoprotein level was within normal limits. Significant EBV viremia granted evaluation for possible hemophagocytic lymphohistiocytosis. A bone marrow biopsy specimen was without evidence of hemophagocytosis. Oncologic management for EBV-positive DLBCL was initiated. The patient died some months later from complications related to the lymphoma.
Discussion
EBV-positive DLBCL is a rare entity in the pediatric population, especially in nonimmunocompromised hosts3 and particularly in the Western world.4 Furthermore, the most commonly affected extranodal sites are the gastrointestinal tract, bone, and skin.5 Thus, considering the rarity of this condition in pediatric patients, as well as the intersection of the presenting airway and constitutional symptoms with a wide range of other pathologies, this case of early lymphomatous involvement of the larynx posed a significant diagnostic challenge. mcfDNA testing enabled the interdisciplinary team to rapidly identify a number of possible etiologic agents to narrow the differential diagnosis in a medically complex patient.
Generally, cell-free DNA refers to circulating fragmented DNA in blood, serum, plasma, synovial fluid, cerebrospinal fluid, stool, saliva, urine, and other bodily fluids. mcfDNA is released through apoptosis or necrosis and is typically found as double-stranded fragments <200 bp in length.6 mcfDNA testing has long been in common use for prenatal testing7 and selected other applications. In recent times, mcfDNA has also become available for the identification of pathogenic DNA in human plasma, in one form available commercially as the Karius test. It relies on next-generation sequencing, which involves rapid DNA sequencing much faster than traditional Sanger sequencing, on account of the ability to sequence many millions of DNA reads in parallel without the need for cloning.8 When sequencing is performed not strictly on the genome of a single organism but on a sample taken directly from the environment, this is referred to as “metagenomics.”9,10 The environment in question can be the natural world or, relevantly to the clinician, the environment of the human body, as described above, in plasma, urine, and other bodily fluids. Next-generation sequencing of mcfDNA relies on the premise that infectious agents leave genetic traces in the plasma of infected hosts, which can be recovered as mcfDNA and sequenced by next-generation sequencing. At present, this type of testing can identify >1000 pathogens (spanning bacteria, viruses, and fungi) from 5 mL of blood. Typical applications include infections in immunocompromised hosts, complicated pneumonia, and endocarditis. Turnaround time for 85% of specimens is estimated at 1 day from receipt of the sample.11
There are several advantages of mcfDNA testing in selected individuals. Firstly, this test is agnostic (ie, no hypothesis is required as to the nature of the underlying infectious agent). It is additionally noninvasive compared with a biopsy and is not site specific, with a possible cost savings associated with the ability to avoid a procedure.12 The wait time is significantly shorter than that required for culture, with higher yields, and it has the ability to identify multiple organisms at once.11 The downsides include the possibility of identifying organisms without clear pathogenic significance, which may complicate interpretation and subsequent clinical decisions. Additionally, there is no in-built antibiotic sensitivity information, as would be the case with culture results, although genetic determination of antibiotic resistance is on the horizon.13 Further work remains to be done in optimizing the relatively low specificity, decreasing turnaround time for serious acute infection, and detecting resistance determinants.12 Overall, it is evident that mcfDNA testing is gaining traction as a diagnostic aid for clinicians in an increasingly complex medical landscape.
Plasma mcfDNA represents a unique tool in the ever-growing armamentarium of diagnostic options for the modern physician. This case highlights that metagenomic data may have further reaching consequences in the clinical setting than would be evident at first glance because infectious agents often have noncanonical effects for which scientific understanding continues to grow. Various pathogens have been linked to the development of cancers,14 neurodegenerative diseases,15 psychiatric conditions,16 cardiovascular diseases,17 and more.
When considering the implications of metagenomics and mcfDNA on clinical diagnostics, there are 3 overarching features to consider: the human genetic material, the microbial genetic material, and the interaction between these two. This is most evident, as in this case report, in virally mediated cancers. For example, reads of human and tumor DNA may be useful in assessing the complex and often heterogeneous genomic backgrounds of the tumor populations within a patient as well as in identifying novel therapeutic targets.18,19 Meanwhile, the microbial reads offer the opportunity to determine the underlying oncogenic pathways, as seen in this case. And finally, the interplay of host and microbe, such as in the human host response, is an avenue of active exploration as well.13
In that same vein, a particularly challenging but important question is the signal level, dependent to a degree on the ratio between microbial and human mcfDNA, that implies infection rather than simply commensal, colonizing, or contaminating organisms.11 Although current reference intervals for mcfDNA are derived from healthy adults,11 the intervals are yet-to-determined in pediatric patients. Furthermore, the degree to which signal levels may correlate with clinical presentation is not fully known. There is evidence to suggest that, for example, the viral load of EBV is an indicator of disease activity in Hodgkin’s lymphoma20 and a prognostic indicator in nasopharyngeal carcinoma.21 However, the precise relationships of mcfDNA signal levels and clinical presentations for various diseases are not fully elucidated. From the human genomic standpoint, several benign and pathologic conditions, including exercise, inflammatory conditions, cardiovascular disease, can increase human mcfDNA in circulation,19,22 and this, too, may create additional challenges in the “needle in a haystack” problem of mcfDNA because the vast majority of signal is human in origin.
Conclusions
This case is an example of the possibilities that may lie ahead, foreshadowing a growing role for next-generation sequencing and metagenomic data in accelerating time-to-diagnosis for challenging pathologies even outside the realm of traditional infectious disease manifestations.
Dr Munjal performed the literature review, drafted the initial manuscript, and reviewed and revised the manuscript; Dr Vukkadala advised on the case description and reviewed and revised the manuscript; Dr Hazard provided pathologic images, advised on the description of the pathology, and reviewed and revised the manuscript; Dr Meister managed this patient’s care, conceptualized the report, and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: Dr Munjal acknowledges the generous support of the National Institutes of Health National Institute of Deafness and Other Communications Disorders training grant T32DC015209. Funded by the National Institutes of Health (NIH).
References
Competing Interests
POTENTIAL CONFLICT OF INTEREST: Dr Munjal serves as consultant for Spiral Therapeutics for work unrelated to this study; Drs Vukkadala, Hazard, and Meister have indicated they have no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: Dr Munjal serves as consultant for Spiral Therapeutics for work unrelated to this study; Drs Vukkadala, Hazard, and Meister have indicated they have no financial relationships relevant to this article to disclose.
Comments