Solitary plasmacytoma (SP) is a rare plasma cell dyscrasia, accounting for less than 5% of plasma cell disorders [1, 2]. It is characterized by the presence of bone or extramedullary tumors, consisted of monoclonal plasma cells, without evidence of multiple myeloma (MM) . Local radiotherapy (R/T), with or without surgical excision, is the recommended treatment of choice . Several studies investigated the outcome of patients with SP and the possible prognostic factors of survival, plasmacytoma relapse or progression to MM [3, 5-9]. However, no definite prognostic factors have been identified because of the heterogeneity of the published studies and the limited number of SP patients included in those studies. Moreover, an appreciable follow-up is required to investigate prognosis and outcome in patients with such a rare neoplastic disorder .
The role of adjuvant chemotherapy (C/T) in the treatment of SP was explored in several published studies with inconclusive results [3, 11-13]. According to the current guidelines, additional C/T is not generally recommended unless there is suboptimal response to R/T, relapse, or progression to MM [4, 14]. However, patients with large masses and high-grade histology may have some benefit from the additional use of C/T, as supported in previous studies [4, 12, 13]. In the past decade, there have been major advances in the treatment of MM. The availability of the novel agents (NA) thalidomide, bortezomib, and lenalidomide has expanded treatment options and has improved the outcome of patients with MM . In addition, bortezomib-based regimens have proved effective in the management of extramedullary disease [16, 17]. The role of NA-based combinations in the treatment of SP has not been currently explored.
The purpose of our study was to describe the clinical features, the outcome and prognosis of a large consecutive cohort of patients with SP who were diagnosed, treated and followed for a long period of time in 12 Greek Myeloma Centers. In addition, we investigated the role of adjuvant C/T or NA-based combinations in the treatment of SP.
We retrospectively reviewed the medical records of 97 consecutive SP patients, 65 with solitary bone plasmacytomas (SBP), and 32 with solitary extramedullary plasmacytomas (SEP), diagnosed and treated in 12 Greek Myeloma Centers between 1991 and 2013. The participating physicians of each center were asked to fill in a questionnaire, which included—among others—the criteria for SP diagnosis, that is, biopsy proven SP, absence of clonal plasma cells in the bone marrow, normal results of skeletal survey including X-rays of the long bones, absence of anemia, hypercalcemia or renal impairment attributed to the plasma cell dyscrasia [4, 18]. Magnetic resonance imaging (MRI) of the thoracic and lumbar spine, which is generally performed as clinically indicated , was available in 50/65 patients with SBP (76.9%); the rest 15 patients underwent skeletal X-rays and CT scan of the involved area and fulfilled the aforementioned criteria of SP as previously described [4, 18]. MRI of the thoracic and lumbar spine was also available in 21/32 patients with SEP (65.6%); PET scan was performed in four patients with SBP after the termination of R/T. Treatment was given according to each center's policy. Complete response was defined as the disappearance of the SP confirmed by CT scan or MRI and the disappearance of M-Component, in case it was present at diagnosis.
Pearson's v2 square test, Mann–Whitney U test, and one-way ANOVA were used for the determination of significant differences and correlations of clinical or laboratory characteristics and response rates between groups. Cox regression likelihood ratio univariate and multivariate analysis were used to determine possible independent predictive factors of survival for the whole group of SP and for SBP and SEP patients, separately. Survival curves were estimated using the Kaplan–Meier method. Plasmacytoma relapse-free survival (PRFS) was defined as the time from SP diagnosis until relapse of plasmacytoma; patients without plasmacytoma relapse were censored. Multiple myeloma-free survival (MMFS) was defined as the time from SP diagnosis until progression to MM; patients without progression to MM were censored. Progression-free survival (PFS) was defined as the time from SP diagnosis until the time of plasmacytoma progression or progression to MM or death.
The results were considered statistically significant for values of P < 0.05. Data processing and analysis were performed with the software package SPSS v16.
The median age of the studied population was 61 years (range 17–85). The median size of the SP was bigger in patients with SBP compared to those with SEP (P = 0.003). Patients with SBP had more often serum or urinary monoclonal component at diagnosis, compared to patients with SEP (P < 0.001 for both parameters). There was no significant difference regarding the levels of the M-component in patients with detectable serum paraprotein between SBP and SEP patients (maximum monoclonal serum paraprotein in SBP and SEP: 22.4 g/L and 26.9 g/L respectively; NS). Patients' characteristics and comparisons between the two groups are shown in Table 1.
Table 1. Patients Characteristics
Abbreviations: SBP: solitary bone plasmacytoma; SEP: solitary extramedullary plasmacytoma; WBC: white blood count; PLT: platelets; CRP: C-reactive Protein; LDH: lactate dehydrogenase; M-Component: monoclonal component; ISS: international staging system; NS: non significant.
Median age (range)
Performance Status (ECOG)
Hemoglobin g/dL (range)
WBCx 103/μL (range)
PLTx 103/μL (range)
LDH U/L (range)
Albumin g/dL (range)
Creatinine mg/dL (range)
CRP mg/L (range)
Calcium mg/dL (range)
β2 microglobulin mg/L (range)
Size of plasmacytoma (cm)
Presence of M-Component
Presence of Bence Jones protein
Expression of CD56 antigen on plasma cells was more frequent in SBP biopsies (P = 0.03). Impairment of humoral immunity (immunoparesis) defined as a suppression of at least one uninvolved immunoglobulin (i.e., for IgG < 700 mg/dL, for IgA <70 mg/dL and for IgM <40 mg/dL), was observed in 24 patients: 16 with SBP and 8 with SEP (NS). Immunoparesis correlated marginally with the presence of M-component (P = 0.04), however, it did not show any correlation with the type of SP (SBP or SEP) with serum β2 microglobulin levels or with LDH. Baseline MRI of the thoracolumbar spine was performed in 17/24 patients (71%) exhibiting immunoparesis at diagnosis. With regard to SBP location, plasmacytoma was located in the vertebrae in 26/65 patients (40%); SEP was more frequently located in the upper respiratory tract (18/32 patients; 56.2%) (Table 2).
Table 2. Anatomic Location of Plasmacytomas
Abbreviations: SEP: solitary extramedullary plasmacytoma; SBP: solitary bone plasmacytoma
Treatment and response
Eighty patients (82.5%) received R/T, either alone or in combination with C/T or surgical excision. The median dose of the delivered radiation was 40Gy (range: 24–55Gy). Forty-seven patients (48.4%), 31 with SBP and 16 with SEP received C/T or NA-based regimens; 40 patients received C/T or NA-based regimens as an adjuvant treatment to R/T and/or surgical excision and seven patients as a single therapy. Twenty-seven patients received NA-based combinations. In particular 22/47 patients (46.8%) received bortezomib based combinations: nine patients received bortezomib, anthracycline, and dexamethasone (PAD), six patients received bortezomib and dexmethasone (VD), four patients received melphalan, bortezomib, and dexamethasone (VMP), and three patients received cyclophosphamide, bortezomib, and dexamethasone (CyBorD); 15/22 patients, received bortezomib based regimens, complementary to R/T. Five patients received immunomodulatory drug combinations, all of them in the adjuvant setting. In particular, three patients received cyclophosphamide, thalidomide, and dexamethasone (CDT) and two patients received lenalidomide and dexamethasone (RD). Thirteen patients underwent autologous transplantation after induction treatment. There was no statistical difference between patients with SBP and SEP, regarding the type of treatment administered. Of note, more patients who had a large plasmacytoma mass (≥5 cm) received C/T or NA-based regimens (72%), compared to patients with plasmacytoma mass <5 cm (50%; P = 0.08). Treatment details are shown in Table 3.
Objective response rate (≥PR; ORR) and complete response (CR) was 91.8% and 61.9%, respectively. The median time to response was 3 months (range 1–24 months) and did not differ between the two groups. In patients with SBP, ORR, and CR were 93.8% and 58.5%, respectively, while in patients with SEP, the respective rates were 87.7% and 68.8%. There was no significant difference in response rates between the two groups of SP. Achievement of CR did not correlate with the type or size of SP, the radiation dose, or the administration of adjuvant C/T or NA-based regimens. Disappearance of M-component was observed in 43/65 patients (66%) who had detectable paraprotein at diagnosis and it was more frequent in patients with SEP (10/13 patients with SEP and 33/52 patients with SBP), albeit without reaching a statistical significance (P = 0.1).
Overall, 38 (39.2%) patients relapsed: 14 patients had relapse of SP without MM progression, 16 patients progressed to MM without any evidence of plasmacytoma relapse, and 8 patients developed both MM progression and plasmacytoma relapse. Seven patients displayed local relapse and 15 patients had relapse in other locations. The median time to SP relapse was 29 months (range: 5–166 months). Nineteen out of 24 (79.1%) patients who progressed to MM had SBP and 5/24 (20.8%) had SEP (P = 0.1). The median time to progression to MM was 27.5 months (range: 7–110 months). Data regarding salvage treatment for plasmacytoma relapse or progression to MM were available in 30 patients: 21 patients received C/T, 4 received R/T, and 5 received combined modality treatment.
After a median follow-up of 60 months (range 3–264 months), 74 patients (76.3%) were alive, 13 patients (13.4%) had died, and 10 patients (10.3%) were lost to follow-up. Nine patients died due to MM progression, three patients due to pulmonary infection after C/T-induced neutropenia and one patient died from H1N1 virus infection. Overall, the 5- and 10-year overall survival (OS, PRFS, and PFS probability were 84% and 78%, 72% and 58%, and 58% and 43%, respectively (Fig. 1a–c). The 5- and 10-year MMFS probability overall, was 70 and 59%, respectively. The 5- and 10-year OS probability was longer in SEP compared to SBP (92% and 89% in SEP and 86% and 69% in SBP, respectively), but this difference was not statistically significant (Fig. 2a). The 5- and 10-year PFS probability for SBP vs. SEP was 44% and 40% vs. 71 and 50%, respectively (P = 0.1). The 5- and 10-year MMFS probability for patients with SBP vs. SEP was 58% and 50% vs. 90% and 70% respectively, (P = 0.054; Fig. 2b).
We performed a Cox regression analysis for the whole study population and separately for SBP and SEP group.
With regard to the whole population, age and PRFS were the only independent prognostic factors of OS, in the multivariate analysis (Table 4). The 10-year OS probability for patients <60 vs. ≥60 years was 85% vs. 60% (P = 0.02); patients with PRFS >24 vs. ≤24 months had a 10-year OS probability of 82% vs. 60% (P = 0.01) (Fig. 3a,b).
Immunoparesis at diagnosis was the only negative predictive factor of progression to MM (Table 4). Five- and 10-year MMFS probability for patients with immunoparesis vs. patients with intact uninvolved immunoglobulins was 48% vs. 76% and 30% vs. 50%, respectively (P = 0.03). Achievement of CR was the only positive predictive factor of PRFS for the whole population (HR: 2.4; 95% CI: 1.046–5.5; P = 0.03). Interestingly, treatment with CT or NA-based regimens was not an independent prognostic factor of survival parameters (i.e., OS, PRFS, PFS, and MMFS); 10-year OS probability for patients treated with RT ± surgical excision vs. those treated with combined therapies including CT± NA was 81% vs. 70%, respectively (P = 0.3).
In patients with SBP, age was the only prognostic factor of OS (HR: 1.07; 95% CI: 1.01–1.13; P = 0.017). No predictors of progression to MM were found in SBP patients. In the SEP group, no particular predictors of OS or MMFS were identified.
Toxicity data were available in 68 patients (70%). No significant acute or late radiation toxicity was reported. Six patients treated with local R/T for SP located at the upper respiratory tract had grade 1/2 mucositis/xerostomia; six patients that underwent R/T experienced grade 1 myelotoxicity. Overall, myelotoxicity, neurotoxicity, and neutropenic infections of any grade presented more frequently in patients treated with CT (P < 0.001, P = 0.01, P = 0.009, respectively). Myelotoxicity, GI toxicity, and neutropenic infections did not differ between patients that received NA-based combinations or C/T, whereas neurotoxicity of any grade was more commonly observed in patients treated with NA-based combinations (P = 0.001).
In this study, we presented data of a large consecutive cohort of SP patients diagnosed, treated, and followed for a median of 5 years, in 12 Greek Myeloma Centers. The number of patients and the appreciable median follow-up allowed us to draw quite reliable conclusions, with regard to outcome and prognosis, as well as to the role of adjuvant C/T and novel antimyeloma agent combinations in the treatment of SP. To diminish the possibility of misclassification or inclusion of “plasmacytoma plus” patients , we used appropriate questionnaires that included the diagnostic criteria of SP as previously described [4, 18]. Our results regarding clinical features were in accordance with previous studies [18, 20]. CD56 expression on plasma cells was more frequent in patients with SBP compared to patients with SEP. This is probably because SBP is located close to the bone marrow and therefore consists of plasma cells that require particular adhesion molecules, such as CD56, for their survival.
We have confirmed that patients with SP enjoy excellent response rates and prolonged OS, which is positively influenced by young age and sustained PRFS. Ozsahin et al. reported previously that young age and small size of SP were positive predictors of OS in a large cohort of SP patients . However, in other studies there is no report regarding predictors of OS [3, 6]. In our study, 5- and 10-year OS probability, although higher in SEP patients, did not statistically differ between the two groups. It is possible that a longer follow-up would allow us to show a significant survival advantage of SEP over SBP. Nevertheless, published data concerning this issue are inconsistent [3, 5, 7, 9]. PFS was marginally better in SEP patients. In agreement with previous studies , we did not find any statistical difference regarding PRFS, between SEP and SBP patients (data not shown). Achievement of CR was the only predictor of PRFS, indicating the important role of optimum initial response in SP. No significant predictors of PFS were identified in the current study. Nevertheless, negative predictors of SP relapse or PFS have not been consistent in published studies [4, 21]. In agreement with previous studies [3, 5, 13, 21], MMFS was better in patients with SEP compared to patients with SBP.
Immunoparesis at diagnosis was the most powerful prognostic factor of progression to MM. Previous studies have also supported the prognostic value of immunoparesis in SP [3, 22]; however, this has been questioned because the suppression of normal immunoglobulins may indicate the presence of occult MM. Moreover, immunoparesis has been used as a diagnostic criterion for MM . We have recently reported that 1527/1755 (87%) newly diagnosed MM patients had immunoparesis at diagnosis . This percentage was significantly higher compared to the 25% of SP patients presenting immunoparesis at diagnosis, reported in the current study (P < 0.001). Interestingly, Ozsahin et al. reported a very high (58%) percentage of SP patients exhibiting immunoparesis at diagnosis . A possible explanation for this discordance is that the limited use of MRI imaging (33% of patients) reported in the aforementioned study , may have led to misdiagnosis. In our study, the majority of patients exhibiting immunoparesis at diagnosis (71%) underwent MRI of the thoracic and lumbar spine during the initial diagnostic work-up, thus, diminishing the possibility of misclassification. The pathogenesis of immunoparesis in MM is unclear; possible implicated mechanisms include increased levels of transforming growth factor beta (TGF-β) and lack of B cell accessory signals from helper T cells . Of note, it has been previously suggested that angiogenesis is increased in SBP and is a significant predictor of progression to MM ; in that study, vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) were highly expressed in SBP. In another study, it has been shown that VEGF is increased in plasmacytomas . Recently, it has been demonstrated that the levels of VEGF were increased in breast cancer patients resistant to hormone therapy, and this correlated inversely with the levels of serum immunoglobulins , indicating that there is probably a relation between VEGF levels, angiogenesis and impairment of humoral immunity. All the aforementioned data support our finding that immunoparesis may accompany solitary plasma cell tumors and, therefore, it could be a useful parameter for predicting evolution to MM. Finally, we did not confirm the predictive value for progression to MM of parameters previously investigated, such as age, disappearance of M-component, SBP anatomic location and tumor size [2, 4, 18, 29].
The recommended treatment of choice for SP is definitely a moderate dose of R/T, which induces SP control in about 90% of patients [4, 5, 13, 22]. In several studies adjuvant C/T was administered in combination with R/T. Mayr et al. found that adjuvant chemotherapy may prevent progression to MM . Holland et al. demonstrated that C/T did not affect the incidence of conversion, although it did appear to delay progression to MM . A small randomized study suggested that there is benefit from additional melphalan and prednisone, administered for 3 years after radiotherapy . However, in most other series additional C/T had no beneficial effect [3, 11] and it may have been responsible for the development of secondary myelodysplastic syndrome . The role of NA-based combinations in the treatment of SP has not been previously explored, on one hand because in most centers SP is initially treated with R/T, according to recommended guidelines , and on the other hand because most of the published studies have been performed before the era of NA. In the current study, C/T and/or NA-based combinations were mostly applied complementary to R/T ± surgical excision, according to each center's policy. Interestingly, half of the patients that received additional therapy were treated with NA-based regimens, mostly bortezomib-based combinations. A possible reason for this preference is that bortezomib-based regimens have shown efficacy in the treatment of extramedullary myeloma [32, 33]. Our results suggest that the additional use of CT or NA-based combinations increases toxicity without offering any survival advantage, or any benefit in regard to inhibiting progression of SP to MM. This latter finding is important, because there are no published data regarding the role of NA-based combinations in the treatment of SP. We need to point out that the lack of additional efficacy of C/T and NA-based regimens was probably related to the fact that, in the majority of patients, those treatments were given in addition to R/T, which is highly effective in the treatment of SP. Therefore, we could argue that the use of C/T or NA-based therapy would be considered as “overtreatment” in this setting. Of note, patients that received only R/T had significantly better OS compared to patients that received only C/T or NA-based regimens (P = 0.01; data not shown). However, a larger number of patients is required in order to confirm this finding.
In summary, our study has demonstrated that R/T remains the best therapeutic option in patients suffering from SP, offering high response rates and excellent survival. The addition of C/T or NA-based combinations increases toxicity without improving outcomes and, therefore, it is not recommended as a therapeutic approach for the treatment of SP, even in patients with large tumor masses. We should underscore that our study included an appreciable number of SP patients diagnosed and treated after 2000, that is, after the introduction of NA in the treatment of plasma cell dyscrasias. Therefore, it is the first study that offers information regarding the role of NA-based combinations in the treatment of SP. Finally, according to our results, young age and prolonged PRFS were positive predictors of survival. Immunoparesis at diagnosis was the only negative predictor of progression to MM and achievement of CR was the only positive predictive factor for PRFS. We should stress that none of the published studies has managed to determine uniform and powerful prognostic factors of SP relapse, progression to MM and OS, mainly because of their heterogeneity in regard to study population, treatment administered and diagnostic criteria. Large prospective studies are needed in order to answer the unresolved queries concerning this rare plasma cell dyscrasia which carries a unique biology.