Abstract
POEMS (polyneuropathy, organomegaly, endocrinopathy, monoclonal protein and skin changes) syndrome is a rare paraneoplastic syndrome due to an underlying plasma cell disorder. The chronically evolving nature of this syndrome and diverse clinical manifestations make it a challenging diagnosis, especially in the early stages of the disease. We present a case that illustrates well how these confounding factors can delay recognition of the disorder. We describe the presentation of some of the classical symptoms, and discuss how investigations can be optimised in order to fit together the clinical picture. Recent advances in the understanding of nerve conduction studies and electromyography in POEMS may help improve the early identification of this disease. This is of great importance, as early recognition and the initiation of appropriate treatment can reduce morbidity and mortality in POEMS.
Background
POEMS syndrome is a rare paraneoplastic syndrome due to an underlying plasma cell disorder. It is named for the constellation of symptoms originally used to define it: polyneuropathy, organomegaly, endocrinopathy, monoclonal (M) protein and skin changes.1 It remains a challenging diagnosis for a number of reasons. First, few clinicians will see a confirmed case during their career. Second, although the acronym is memorable, not all features it describes are required for diagnosis. Moreover, there are a number of important additional features, some of which may occur more frequently in POEMS cases than those described in the acronym.2 Third, the different symptoms may evolve slowly, and may appear at different times. This staggered time course, together with the diversity of the clinical manifestations of the disease, means that the cardinal symptoms are often not at first considered as a collective entity.3 Lastly, the treatments initiated for relief of individual symptoms at the time they first manifest may have effects that mask the development of additional symptoms, or affect laboratory test results.
A series of 99 confirmed POEMS cases reported that 85% were initially misdiagnosed, and median time from symptom onset to diagnosis was 18 months.4 Minimising this delay in diagnosis is increasingly important as more specific therapies for POEMS are now available, and early initiation of these is critical to reducing morbidity and mortality.3 We describe a case that illustrates the complexities involved in recognising this unusual disease in a timely fashion.
Case presentation
A 47-year-old African-American man was referred to haematology in September 2015, following a 15-month history of significant unintentional weight loss of approximately 23 kg; CT and positron emission tomography/CT (PET/CT) scans had revealed extensive retroperitoneal and pelvic lymphadenopathy (figure 1). The CT scan had also shown hepatosplenomegaly, with a liver craniocaudal measurement of 21.8 cm and a spleen length of 14.5 cm. The PET/CT scan additionally showed multiple hypermetabolic foci in the axial and appendicular skeleton (figure 1). Most of these had a sclerotic appearance and one had a lytic appearance.
Figure 1.
Images from a PET/CT scan showing both lymphadenopathy and a sclerotic bone lesion. Four corresponding images are shown from level L5. Yellow arrows indicate an enlarged hypermetabolic left-sided presacral lymph node. Blue arrowheads indicate an example of a hypermetabolic bone lesion with a sclerotic appearance in the right iliac bone. PET/CT, positron emission tomography/CT.
The patient had an existing diagnosis of chronic inflammatory demyelinating polyneuropathy (CIDP), diagnosed 4 years previously in May 2011, which had steadily progressed despite treatment with prednisolone and intravenous immunoglobulin (IVIG) since July 2011. His medical history was also of note for lumbar vertebral debridement and fusion in February 2009 for damage following a presumed spinal osteomyelitis.
Physical examination revealed a chronically ill appearance with severe motor and sensory impairment of the upper and lower limbs. The lower extremities were more severely affected, with loss of all reflexes and complete loss of motor function, vibration sensation and proprioception below the knee. In the upper limb, motor function was reduced, with distal muscles more greatly affected than proximal muscles, and visible wasting of the intrinsic muscles of the hand. Vibration sensation in the fingers was absent, but proprioception was retained. Triceps and biceps reflexes were reduced, and more distal reflexes were absent. Cranial nerves remained intact. Of additional note, there was marked pitting oedema of the lower limbs, which was reported as persistent for at least the past 2 years; the abdomen was distended with a 1 cm palpable spleen edge; there was mild finger clubbing; and bilateral gynaecomastia was present. There were no discernible skin changes.
The combination of lymphadenopathy, refractory CIDP and osteosclerotic lesions prompted the consideration of POEMS syndrome as a possible underlying diagnosis.
Further in-depth review of the patient's medical notes revealed a number of additional features in support of this diagnosis, including the following active problems indicative of endocrine dysfunction: erectile dysfunction since January 2011; gynaecomastia since September 2011; glucose intolerance since September 2013; and subclinical hypothyroidism since June 2014.
Plasma electrophoresis studies at the time of diagnosis of CIDP in May 2011 had shown a trace IgA λ monoclonal gammopathy of undetermined clinical significance. Repeat tests in October and December 2011 failed to detect the trace protein. Of note, however, these repeats were performed after start of IVIG therapy in July 2011.
Furthermore, retrospective inspection of the patient's weight chart indicated a previous period of rapid weight loss of 18 kg between July 2011 and January 2012, followed by several months of steady weight gain. We speculate that this initial weight loss was masked by a compensatory weight gain, likely attributable to the high dose of prednisolone (80 mg daily) that the patient was being treated with at the time, and may also have been due to the development of extravascular fluid overload.
Investigations
Lymph node biopsy showed large sheets of plasma cells positive for CD138, with several small foci of residual small B and T lymphocytes (figure 2). Human herpesvirus 8 (the aetiological agent in most cases of Castleman's lymphadenopathy, plasma-cell type) and cytokeratin AE1/AE3 (a marker used to detect epithelial micrometastases) stains were negative. Immunohistochemistry for κ and λ light chains was inconclusive, but in situ hybridisation stains for κ and λ showed a predominance of λ-positive plasma cells (figure 2). This is typical of plasma cell proliferates in POEMS, which are λ-restricted in more than 95% of cases.
Figure 2.
Lymph node biopsy showing presence of a lambda-restricted clonal plasma cell proliferate. H&E stain at 40× magnification demonstrates the presence of large sheets of plasma cells (A). Immunostain for the plasma cell marker CD138, confirms the identity of the plasma cells (B). In situ hybridisation (ISH) staining for κ-restricted light chains (C). In situ hybridisation staining for lambda-restricted light chains (D).
Plasma electrophoresis showed a trace IgA λ monoclonal band.
Plasma vascular endothelial growth factor (VEGF) was elevated at 192 pg/mL (reference range: 31–86). The best VEGF cut-off for POEMS diagnosis is considered to be 146 pg/mL; however, a cut-off of 200 pg/mL has a specificity of 95%, with a sensitivity of 68% in support of a POEMS diagnosis.5
Bone marrow biopsy was also performed. No marrow was aspirable, but a trephine core biopsy showed normocellular marrow with trilineal haematopoiesis, including mild relative erythroid hyperplasia. Plasma cells comprised approximately 5% of marrow cellularity in the core biopsy, typical of POEMS, and a distinguishing feature from multiple myeloma, in which a much higher plasma cell marrow content is seen.
Differential diagnosis
The diversity of clinical manifestations of POEMS means that it has a broad differential. As in this case, patients presenting with a demyelinating polyneuropathy as their dominant symptom are frequently misdiagnosed with idiopathic CIDP. Patients with this diagnosis who fail to respond to IVIG therapy should be re-evaluated for possible POEMS.
Patients who present with a monoclonal plasma cell disorder need to be distinguished from differentials including monoclonal gammopathy of undetermined significance, multiple myeloma, amyloidosis and plasmacytoma, as, similarly, the treatment of these disorders is different to the treatment of POEMS.
Table 1 summarises the diagnostic criteria for POEMS, and the features seen in our patient that enabled the diagnosis.
Table 1.
Diagnostic criteria for POEMS syndrome
| Criteria | Per cent affected | Features seen in our patient |
|---|---|---|
| Mandatory criteria | ||
| 1. Polyneuropathy (usually demyelinating) | 100 | Present since May 2011 |
| 2. Monoclonal plasma cell proliferative disorder (>95% lambda-restricted) | 100 | Demonstrated by lymph node biopsy, bone marrow biopsy and plasma electrophoresis |
| Major criteria | ||
| 3. Castleman's disease | 11–25 | |
| 4. Sclerotic bone lesions | 27–97 | Multiple axial and appendicular lesions seen on PET/CT scan |
| 5. VEGF elevation | 86 | Plasma VEGF elevated at 192 pg/mL |
| Minor criteria | ||
| 6. Organomegaly (splenomegaly, hepatomegaly or lymphadenopathy) | 45–85 | Splenomegaly, hepatomegaly and lymphadenopathy seen on CT scan |
| 7. Extravascular volume overload (oedema, pleural effusion or ascites) | 29–87 | Lower extremity oedema present since at least 2013 |
| 8. Endocrinopathy (adrenal, thyroid*, pituitary, gonadal, parathyroid, pancreatic*) | 67–84 | Erectile dysfunction since 2011; gynaecomastia since 2011; glucose intolerance* since 2013; and subclinical hypothyroidism* since 2014 |
| 9. Skin changes (hyperpigmentation, hypertrichosis, glomeruloid haemangiomata, plethora, acrocyanosis, flushing, white nails) | 68–89 | |
| 10. Papilloedema | 29–64 | |
| 11. Thrombocytosis or polycythaemia | 12–88 | |
| Other symptoms and signs | ||
| Clubbing, weight loss, hyperhidrosis, pulmonary hypertension/restrictive lung disease, thrombotic diathesis, diarrhoea, low vitamin B12 | Two episodes of significant unintentional weight loss in 2011 and 2014–2015 Mild finger clubbing |
|
Both mandatory criteria must be met, along with at least one other major criterion and one minor criterion.2 Table also shows the frequency in which each of these features are seen in POEMS cases.2 4 The features that were seen in our patient are summarised.
*Indicates: although thyroid and pancreatic endocrine disorders are recognised features of POEMS, the presence of these alone are insufficient to fulfil the criteria of endocrinopathy, due to their high prevalence in the general population.
PET/CT, positron emission tomography/CT; POEMS, polyneuropathy, organomegaly, endocrinopathy, monoclonal protein and skin changes; VEGF, vascular endothelial growth factor elevation.
Treatment
More research is needed into effective treatments for POEMS. For disease with disseminated bone marrow involvement, the current main treatment strategies are dexamethasone with either melphalan, lenalidomide, thalidomide or cyclophosphamide. This can be used either as primary treatment or prior to an autologous stem cell transplant, depending on patient fitness.
Our patient was started on dexamethasone and lenalidomide, with the intention of completing three cycles of treatment, and then reassessing for response and consideration of suitability for autologous stem cell transplant.
Discussion
In recognition of the ambiguity that existed over the features necessary to establish a diagnosis of POEMS syndrome, Dispenzieri et al6 proposed, in 2003, the first diagnostic criteria based on their retrospective analysis of a series of 99 cases. These criteria have been adapted and updated according to advances in understanding of the syndrome, with the latest update published in September 2015. The criteria are summarised in table 1.
While these criteria enable the suspicious clinician to definitively confirm or refute a possible diagnosis of POEMS, Dispenzieri warns that: ‘the diagnosis will be missed if it is not considered’.2 In a recent review, Li and Zhou3 observed that while the diagnosis of advanced POEMS syndrome with a combination of characteristic manifestations may not be difficult, the challenge remains in recognising the disease at an early stage, when only one or two initial symptoms may be present.
In the majority of POEMS cases, the primary presenting complaint is polyneuropathy, which is frequently misdiagnosed as idiopathic CIDP, as in the case presented. Distinguishing idiopathic CIDP from CIDP due to secondary causes including POEMS can be difficult, particularly in the absence of any other overt systemic symptoms. The clinical pattern of the neuropathy can be indicative, with particular features of POEMS being: an early predominance of sensory symptoms, especially severe lower limb pain; muscle atrophy; and a distal dominant muscle weakness pattern.7 With the benefit of retrospective vision, case notes from our patient indicate that his neuropathy did indeed fit this pattern in the early stages. Relatively recently, and regrettably postdating the initial work up and diagnosis of our patient, two large cohort studies from Japan and the USA have defined distinguishing features in nerve conduction and electromyography studies that can differentiate POEMS from idiopathic CIDP, even before the onset of systemic features of POEMS.8 9
In our case, the patient presented with polyneuropathy without overt evidence of the second mandatory POEMS criterion, a monoclonal plasma cell proliferative disorder. Although serum plasma electrophoresis was performed, this initially showed as trace, and two later repeats were negative. It is important to note that POEMS is described as a ‘low tumour burden disease’, and the monoclonal protein is usually less than 2 g/dL, and thus plasma electrophoresis, even with immunofixation, is negative in 10–15% of confirmed cases.6 7 Furthermore, in this case, the repeat tests were performed several months after the initiation of monthly IVIG therapy, which may have influenced the results by overwhelming the detection of a trace paraprotein. In cases where plasma electrophoresis is negative, a biopsy of bone lesions is recommended, or in rare cases without bone lesions, a blind iliac crest bone marrow biopsy, in order to reveal the presence of a clone.10
Sclerotic bone lesions are seen in ∼95% of POEMS cases, making it a more frequent finding than most of the features defined by the acronym.2 In the case we present here, the presence of osteosclerotic lesions was complicated by the patient's history of presumed lumbar spine osteomyelitis, treated surgically in 2009. It is unclear from his medical records whether this diagnosis was definitively proven by histopathology or culture, and it seems possible that this could have instead been an early manifestation of POEMS. More importantly, however, it is worth noting that having the prior diagnosis of osteomyelitis seems to have been a distracting factor in the interpretation of additional bone lesions seen on subsequent CT scans.
PET scan proved a useful investigation in this case, revealing the sclerotic bone lesions as PET-avid, as well as showing the presence of hypermetabolic lymphadenopathy. The role of PET in the diagnosis and evaluation of POEMS is a relatively new concept, but one that is showing much promise.11
With regard to endocrinopathy, erectile dysfunction is the most common feature in men (reports of up to 89%), and is frequently the first or second symptom.4 10 As in our case, and others reported in the literature,12 patients may present with polyneuropathy or sclerotic bone lesions, with a pre-existing diagnosis of erectile dysfunction of unknown origin, and this can act as a signal to prompt investigation for POEMS.
Altogether, we feel this case illustrates well the barriers to recognising features of POEMS syndrome in the early stages of this disease, and highlights some ways in which recent advances in the understanding of the disease can aid investigation and diagnosis. In particular, it emphasises the importance of re-evaluating for POEMS in any painful polyneuropathy not responding to steroids or IVIG, and the need to go beyond simple plasma electrophoresis in the hunt for the presence of a monoclonal plasma cell disorder, especially if IVIG therapy has already begun. We refer readers to the latest guidelines for further in-depth details of recommended minimum testing in order not to miss this challenging diagnosis.2
Learning points.
POEMS (polyneuropathy, organomegaly, endocrinopathy, monoclonal protein and skin changes) syndrome is a rare disease that is not completely defined by its acronym; clinicians need a high index of suspicion, and an awareness of the other important clinical manifestations that may indicate this diagnosis.
The individual symptoms of POEMS may evolve at different stages, necessitating a complete review of the patient in order for the constellation of symptoms to be recognised. This is particularly important in the context of a demyelinating polyneuropathy not responding to intravenous immunoglobulin treatment.
Recent advances in the understanding of the nerve conduction study and electromyography attributes of POEMS can now help to distinguish it from other demyelinating polyneuropathies early in the disease process.
Evidence of the presence of a monoclonal plasma cell disorder can be difficult to obtain. Up to 15% of cases will have a negative plasma electrophoresis, even with immunofixation. In these cases, biopsy of bone lesions or iliac crest bone marrow is indicated. PET scans may be helpful in revealing bone lesions.
Erectile dysfunction is a common first or second symptom in males, and should prompt investigation for POEMS in patients presenting with demyelinating polyneuropathy or osteosclerotic lesions.
Footnotes
Acknowledgements: The authors thank Claudia Baumann, Turang Behbahani, Zacchary Chinn and Shawn Carter for their help in the care of our patient and the preparation of the manuscript. Dr Robey thanks the following institutions for supporting her elective placement at Emory University: Bloodwise, The British Society of Haematology, The Medical Women’s Federation and the King’s College London Vandervall Fund.
Contributors: RCR, BR and MS were involved in the patient's consult, investigation, diagnosis work up and subsequent care. RCR drafted the manuscript and BR and MS revised the drafted manuscript. CC interpreted the patient's pathology, provided the pathology figures, and revised the pathology sections and figure legends for the manuscript.
Competing interests: None declared.
Patient consent: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1.Bardwick PA, Zvaifler NJ, Gill GN et al. Plasma cell dyscrasia with polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes: the POEMS syndrome. Report on two cases and a review of the literature. Medicine (Baltimore) 1980;59:311–22. 10.1097/00005792-198007000-00006 [DOI] [PubMed] [Google Scholar]
- 2.Dispenzieri A. POEMS syndrome: update on diagnosis, risk-stratification, and management. Am J Hematol 2015;90:951–62. 10.1002/ajh.24171 [DOI] [PubMed] [Google Scholar]
- 3.Li J, Zhou DB. New advances in the diagnosis and treatment of POEMS syndrome. Br J Haematol 2013;161:303–15. 10.1111/bjh.12236 [DOI] [PubMed] [Google Scholar]
- 4.Li J, Zhou DB, Huang Z et al. Clinical characteristics and long-term outcome of patients with POEMS syndrome in China. Ann Hematol 2011;90:819–26. 10.1007/s00277-010-1149-0 [DOI] [PubMed] [Google Scholar]
- 5.D'Souza A, Hayman SR, Buadi F et al. The utility of plasma vascular endothelial growth factor levels in the diagnosis and follow-up of patients with POEMS syndrome. Blood 2011;118:4663–5. 10.1182/blood-2011-06-362392 [DOI] [PubMed] [Google Scholar]
- 6.Dispenzieri A, Kyle RA, Lacy MQ et al. POEMS syndrome: definitions and long-term outcome. Blood 2003;101:2496–506. 10.1182/blood-2002-07-2299 [DOI] [PubMed] [Google Scholar]
- 7.Gupta M, Verma R, Garg RK et al. An unusual case of POEMS syndrome. BMJ Case Rep 2012;2012:pii: bcr0220125885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mauermann ML, Sorenson EJ, Dispenzieri A et al. Uniform demyelination and more severe axonal loss distinguish POEMS syndrome from CIDP. J Neurol Neurosurg Psychiatr 2012;83:480–6. 10.1136/jnnp-2011-301472 [DOI] [PubMed] [Google Scholar]
- 9.Nasu S, Misawa S, Sekiguchi Y et al. Different neurological and physiological profiles in POEMS syndrome and chronic inflammatory demyelinating polyneuropathy. J Neurol Neurosurg Psychiatr 2012;83:476–9. 10.1136/jnnp-2011-301706 [DOI] [PubMed] [Google Scholar]
- 10.Dispenzieri A, Buadi FK. A review of POEMS syndrome. Oncology (Williston Park) 2013;27:1242–50. [PubMed] [Google Scholar]
- 11.Pan Q, Li J, Li F et al. Characterizing POEMS Syndrome with 18F-FDG PET/CT. J Nucl Med 2015;56:1334–7. 10.2967/jnumed.115.160507 [DOI] [PubMed] [Google Scholar]
- 12.Livingston J, Cobiella C, Hall-Craggs MA. POEMS syndrome—a unique presentation of a rare paraneoplastic syndrome. BMJ Case Rep 2010;2010:pii: bcr0920103324. [DOI] [PMC free article] [PubMed] [Google Scholar]


