Copyright American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
The cause of follicular spicules in multiple myeloma (MM) is not known.
We present a case of follicular spicules in a patient with MM, which is very reminiscent of trichodysplasia spinulosa caused by a polyomavirus. No trichodysplasia spinulosa–associated polyomavirus could be isolated from the skin lesions; however, the spicules were positive for Merkel cell carcinoma virus, which is also a polyomavirus.
Conclusions and Relevance
Follicular spicules in MM are probably not caused by the trichodysplasia spinulosa–associated virus. Merkel cell polyomavirus could contribute to the origin of this dermatosis.
Multiple myeloma (MM) has many cutaneous manifestations. Follicular spicules are relatively rare among them.1 Previous articles on this dermatosis categorized it as paraneoplastic in origin. No specific treatment is available. The clinical picture is very reminiscent of trichodysplasia spinulosa (TS), which is known to be caused by the TS–associated polyomavirus (TSV). Herein, we describe a case of follicular spicules in a patient with MM. We attempted to identify a possible viral cause and to classify this enigmatic entity.
A man in his 70s consulted us regarding coarse outgrowths on his face that had been present for several months (Figure 1A). These lesions were asymptomatic, but the patient was concerned about his appearance. He had been treated with mid-potency topical corticosteroids for several weeks with no apparent effect. The patient had no medical history, but he complained of fatigue and malaise. Examination of his skin revealed small, folliculocentric, coarse, yellowish spicules located mainly on his face but also on his chest and lower arms. Ulcerated papules covered with hemorrhagic crusts were also visible on his torso. All his fingernail plates were white (Figure 1B).
A, Folliculocentric spicules on the face of the patient. B, A white discoloration appeared on the fingernails. C, Cidofovir treatment cleared the spicules from the face.
We considered the differential diagnosis of facial folliculocentric spicules comprising TS and follicular spicules of MM (FSMM). The first is a well-characterized infection with TSV in immunosuppressed patients. The latter is seen in patients with MM, but its pathophysiologic role is not understood. Conventional hematoxylin-eosin staining of a skin biopsy specimen taken from the forehead showed a widened hair follicle with a protruding spicule with hyperparakeratosis surrounded by multiple eosinophilic bodies (Figure 2A). Electron microscopy of a follicular keratinocyte revealed multiple round structures of approximately 1 μm in diameter arranged in a colonlike pattern (Figure 2B). Further magnification showed these structures to consist of fibrillar bodies arranged in a paracrystalline configuration. No viral particles could be found. These findings are reminiscent of both TS and FSMM, making differentiation based on these investigations alone impossible. Further serological testing showed an increased serum protein level of 10.1 g/dL (to convert to grams per liter, multiply by 10), which was primarily composed of γ-globulins (5.85 g/dL). No antibodies against the human immunodeficiency virus were found, and leukocyte levels were normal. A bone marrow biopsy revealed 50% infiltration by monoclonal plasmacytoid, CD138-positive, and IgG-producing cells. Fluorescence in situ hybridization revealed an extra copy of chromosome 9 in 41% of the analyzed cells, fitting the diagnosis of MM. With this information, we diagnosed FSMM in this patient.
A, Histopathologic findings in a spicule specimen. A widened hair follicle with a protruding spicule with dyskeratosis of follicular epithelium and excess inner root sheathlike cells surrounded by multiple eosinophilic bodies (hematoxylin-eosin, original magnification ×40). The interfollicular epidermis displays acanthosis (inset: magnification of eosinophilic bodies in follicular epithelium of the infundibulum [hematoxylin-eosin, original magnification ×100]). B, Electron microscopy of an affected follicle. Inside follicular keratinocyte fibrillar structures arranged in a paracrystalline configuration (arrowheads). The inset shows these structures further magnified.
Because the etiology of FSMM is unknown, specific treatment is lacking. However, the clinical and histopathological findings in FSMM are reminiscent of TS. Identifying TSV in the lesions of our patient could provide a basis for treatment. We therefore endeavored to identify a possible viral causative agent for FSMM. Trichodysplasia spinulosa has been well characterized as a viral dermatosis caused by a polyomavirus, but in the case of FSMM, no publications record attempts to identify the causative agent. A TSV-specific polymerase chain reaction2 did not reveal TSV in the spicules. To rule out another polyomavirus, rolling circle amplification was undertaken; however, no polyomavirus was detected using this technique.
Having ruled out TSV as the causative agent, we turned to our random amplification virus discovery approach to detect viral sequences. In this protocol, we enriched for virus particles and used a combined random amplification and 454 deep-sequencing approach to search for both RNA and DNA viruses.3 We used a spicule from a hair follicle to search for viral presence. The protocol for the discovery of RNA viruses and DNA viruses yielded 42 641 and 152 082 reads, respectively. After quality trimming, reads were assembled into contigs. Contigs and reads that were not assembled into contigs (so-called singletons) were submitted to the Basic Local Alignment Search Tool for nucleotides (BLASTn). This algorithm finds regions of local similarity between the submitted sequence and known sequences in a nucleotide database and makes calculations on statistical significance of possible matches. Sequences were classified into viruses, bacteria, and/or eukaryotes based on the taxonomic origin of the best-hit sequence using MEGAN (metagenome analyzer) software.4 The contigs from both the RNA and DNA virus discovery protocols and the singletons derived from the RNA virus discovery protocol did not result in any viral hits in the nucleotide BLAST. However, the singletons resulting from the DNA virus discovery protocol, showed 10 hits that were assigned to the polyomavirus family. All hits had the highest score to Merkel cell polyomavirus (MCV). A translated protein BLAST search based on the nucleotide sequences did not result in any additional viral hits. Using different BLAST parameters and MCV genome as a reference, another 4 reads were found to be assigned to MCV. The 14 sequence reads ranged in size from 109 to 462 nucleotides. The sequence identity of these hits to MCV ranged from 95% to 100%. Together, the 14 reads covered 1943 nucleotides of the viral genome. However, the reads that aligned to the genome were not equally distributed (P > .05). Instead, the first approximately 1800 nucleotides of the genome were not covered by any reads. A polymerase chain reaction directed against the MCV large-T (tumor) gene5 confirmed the presence of MCV and revealed a low mean copy number of 0.1 per cell.
With a possible viral causative agent, treatment was commenced with cidofovir gel, 1%, once daily for the patient’s face only. This therapy had shown success in cases of TS.6 Within a week this lead to hemorrhagic crusts on his face, and within 3 weeks it cleared his face almost completely of spicules while no effect was seen on untreated skin (Figure 1C). Two weeks after the commencement of cidofovir treatment, combination therapy with melphalan, prednisone, and bortezomib was started for the treatment of the patient's MM.
Polyomaviruses are ubiquitous viruses capable of infecting humans and many animal species. Sequences from human polyomaviruses have been isolated from stool, respiratory secretions, blood, central nervous fluid, urine, and skin.7 Because of new technologies, their involvement in various human diseases is easier to establish. Our findings suggest a pathophysiologic role for a polyomavirus but not TSV in FSMM.
A recent article reported absence of TSV in a case of follicular spicules in a MM confirming our findings.8 Satta and colleagues9 suggested a role for paraprotein deposition in FSMM after finding the same electrophoresis pattern in a patient's spicules and serum paraprotein, although we found no immunoglobulin depositions in our patients’ skin.
Although MCV cannot be appointed as the etiological agent of FSMM, several aspects suggest a role for this virus in this dermatosis. First, the patient has MM, a disease that predisposes to opportunistic infections. Second, the clinical picture resembled TS, caused by a related polyomavirus TSV. Third, the treatment with cidofovir, a viral DNA-polymerase inhibitor, cleared the spicules. This effect is also seen in TS and human papillomavirus–infected skin, although polyomaviruses are not known to encode any polymerase. Although the patient was started on chemotherapy 1 week before the full clearing of spicules, the confinement of the effect to the treated area is suggestive of some cidofovir effect. In addition, the patient had a notable response to cidofovir within 1 week, which was prior to the initiation of chemotherapy. And finally, MCV was discovered inside the affected tissue. Regarding the latter, and the fact that MCV spread from healthy skin cannot be ruled out because asymptomatic carriage is highly prevalent in older individuals,10 further study into the observed association between MCV and FSMM as well as why cidofovir would have an effect on the lesions is warranted.
We have found evidence implicating a possible viral etiology instead of a paraneoplastic origin for FSMM. The differential diagnosis of TS was rejected owing to the absence of TSV by polymerase chain reaction. However, further research is necessary to confirm our findings.
Accepted for Publication: June 1, 2014.
Corresponding Author: Sander van Boheemen, MSc, Department of Viroscience, Erasmus Medical Centre, Dr Molewaterplein 50, 3015GE Rotterdam, the Netherlands (email@example.com).
Published Online: September 3, 2014. doi:10.1001/jamadermatol.2014.1616.
Author Contributions: Mr Van Boheemen and Dr Hajdarbegovic had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Van Boheemen, Feltkamp, Fouchier, Hajdarbegovic.
Acquisition, analysis, or interpretation of data: Van Boheemen, Jones, Muhlemann, Feltkamp, Hajdarbegovic.
Drafting of the manuscript: Van Boheemen, Hajdarbegovic.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Van Boheemen.
Obtained funding: Fouchier.
Administrative, technical, or material support: Van Boheemen, Feltkamp, Fouchier.
Study supervision: Van Boheemen, Fouchier, Hajdarbegovic.
Programming and Visualization: Jones.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported in part by the European Union Seventh Framework Programme (FP7) under EMPERIE grant agreement No. 223498, SILVER grant agreement No. 260644, and ANTIGONE grant agreement No. 278976.
Role of the Funder/Sponsor: EMPERIE and ANTIGONE had a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript. The study sponsors had no role in the decision to submit the manuscript for publication.
Additional Contributions: Theo Bestebroer, BSc, and Siamaque Kazem, MSc, provided technical assistance, and Derek Smith, PhD, and David Burke, PhD, provided essential discussions. No financial compensation was provided for their contributions.
van Boheemen S, Jones T, Muhlemann B, Feltkamp MC, Fouchier RAM, Hajdarbegovic E. Cidofovir Gel as Treatment of Follicular Spicules in Multiple Myeloma. JAMA Dermatol. 2015;151(1):82-84. doi:10.1001/jamadermatol.2014.1616