Annals of Oncology Advance Access originally published online on October 3, 2006
Annals of Oncology 2007 18(1):129-135; doi:10.1093/annonc/mdl329
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 European Society for Medical Oncology
hematologic malignancies |
A clinicopathological retrospective study of 131 patients with primary bone lymphoma: a population-based study of successively treated cohorts from the British Columbia Cancer Agency
1 Division of Hematology
2 Division of Medical Oncology
3 Division of Pathology, British Columbia Cancer Agency, Vancouver, British Columbia, Canada
* Correspondence to: Dr J. M. Connors, British Columbia Cancer Agency and the University of British Columbia, 600 West 10th Avenue, Vancouver, British Columbia V5Z 4E6, Canada. Tel: +1-604-877-6000 x 2746; Fax: +1-604-877-0585; E-mail: jconnors{at}bccancer.bc.ca
 |
Abstract |
|---|
 Top Abstract introductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Background: Primary bone lymphoma (PBL) is a distinct clinicopathological entity. Although PBL has been reviewed in several small studies, few reflect recent improvements in primary treatment.
Methods: We used the British Columbia Cancer Agency Lymphoid Cancer Database to identify all patients with PBL (1983–2005). All were staged in a uniform manner and treated with era-specific protocols.
Results: We identified 131 patients with a median age of 63 years (18–87). One third had disease in long bones and another one third had disease in the spine, of which half presented with spinal cord compression. Patients with diffuse large-cell lymphoma (DLCL) (n = 103, 79%) had 5- and 10-year overall survivals (OS) of 62% and 41%, respectively. Multivariate analysis identified three prognostic groups: age <60 with International Prognostic Index (IPI) 1–3 (n = 43), age 
60 with IPI 0–3 (n = 23) and age 60 with IPI 4–5 (n = 33), with markedly different 5-year OS of 90%, 61% and 25%, respectively (P < 0.0001). Neither primary site nor pathological fracture at presentation had an impact on OS. The 3-year progression-free survival in patients who received rituximab plus combination chemotherapy with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOPR) chemotherapy was 88% compared with 52% in those who received CHOP-like chemotherapy without rituximab (P = 0.005). The 10-year OS for those with advanced-stage disease who received irradiation plus chemotherapy was 25% versus 56% for those who received chemotherapy alone (P = 0.025). Patients received irradiation if spinal cord compression was present or residual disease at the end of chemotherapy was thought to require it.
Conclusions: PBL is usually of DLCL type and has an improved outcome with CHOPR. Younger patients with good IPI score have a favorable prognosis.
Key words: bone, lymphoma, presentation, treatment
|
introduction |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
Primary bone lymphoma (PBL) is a rare disease, accounting for approximately 3% of all primary bone malignancies [1, 2]. Bone involvement is not uncommon in advanced-stage lymphoma originating in other sites, but PBL constitutes <5% of extranodal lymphomas [3, 4] and <2% of all lymphomas in adults [5]. Disease may be complicated at presentation by pathological fracture or spinal cord compression. Depending on the definition used to distinguish primary from secondary lymphoma of the bone, most patients present with limited-stage disease [6]. As staging procedures have improved over time, especially with the combined use of computed tomography (CT), magnetic resonance imaging (MRI) and now positron emission tomography, a higher proportion of patients is diagnosed with stage IV disease [6, 7]. Most reports describe long bones as the main site of PBL, with a frequency ranging between 44% and 71% [5, 6, 8].
Diffuse large-B-cell lymphoma (DLBCL) accounts for the majority of cases [6, 9] of PBL. Appropriate treatment depends on the histologic type and stage. Traditionally, irradiation has been used alone or in combination with chemotherapy. Irradiation alone in limited-stage disease has resulted in a poor 5-year overall survival (OS) of around 45% [10] even when patients have been carefully chosen for apparently limited-stage disease. Over the last two decades, chemotherapy has become the standard treatment with or without consolidative irradiation.
In order to understand the presentation, natural history, response to treatment and long-term outcome of PBL, we examined all cases seen at the British Columbia Cancer Agency (BCCA) during the past two decades. The aim of this study was to describe the clinicopathological characteristics of PBL and identify prognostic factors that can affect the outcome of treatment.
|
materials and methods |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
patients
We identified all cases of PBL seen at the BCCA between January 1983 and March 2005, by searching the computerized Lymphoid Cancer Database. We only included patients who were diagnosed and treated at the BCCA. Our definition of PBL was a confirmed pathological diagnosis of lymphoma with bone involvement as the primary site of disease. We excluded cases positive for human immunodeficiency virus, or with a pathologic diagnosis of plasmacytoma, Castleman's disease or Hodgkin's lymphoma. We excluded all those with lymph node involvement on the other side of the diaphragm from the involved bone, those with distant-site bone marrow involvement and all patients with other extranodal involvement such as peripheral blood, lung, liver or other organs not intimately involved with the known site of bone involvement. Patients with brain involvement were included only if there was a biopsy-proven skull lesion and the brain involvement represented direct local extension of the lymphoma. The medical records of 131 patients identified were reviewed for presenting findings, pathological diagnosis, date of diagnosis, International Prognostic Index (IPI) score, lactate dehydrogenase (LDH) level, peripheral complete blood cell counts (CBC), tumor size, primary site of disease, stage, presence of a pathological fracture or spinal cord compression, treatment, date of last follow-up and, if deceased, cause of death. The University of British Columbia—BCCA Research Ethics Board—approved this retrospective study.
diagnosis, staging, IPI and treatment
All patients had a biopsy-proven diagnosis of lymphoma established by an experienced hematopathologist using the terminology of the current World Health Organisation classification [11]. Staging investigations included history and physical examination, CBC, LDH, serum bilirubin, creatinine and alkaline phosphatase, chest radiograph, CT scan of abdomen and pelvis and the site of presenting bone involvement and random bone marrow biopsy. Most patients also underwent radionucleotide technetium (Tc99) bone scanning and, since the mid-1990s, MRI. Stage was assigned according to Ann Arbor staging system. Stage IE consisted of a single localized bone lesion and stage IIE a single bony site and contiguous or closely associated lymph nodes. By definition, stage III disease was not found in our series because we excluded all patients with distant nodal disease. In order to exclude patients with a primary site other than bone, stage IV disease only included cases with either multiple sites of bone involvement or diffuse involvement of a single large bone such as the femur. We further grouped stages into either limited- or advanced-stage disease. Limited-stage disease was defined as stage IE or IIE in the absence of B symptoms and with no tumor mass 10 cm in maximum diameter. Advanced-stage disease consisted of all patients with stage IV disease or stage IE or IIE disease plus either B symptoms or tumor 10 cm in maximum diameter.
All patients were assigned a score using the IPI for diffuse large-cell lymphoma (DLCL) [12]. We grouped patients by IPI score into group 1 including those with scores 0–1; group 2, scores 2–3 and group 3, scores 4–5. The Eastern Cooperative Oncology Group scale was used to assess performance status.
Patients were treated according to diagnosis-, era- and age-specific treatment guidelines that included chemotherapy, irradiation or a combination of both. For meaningful comparison, we divided chemotherapy regimens into combination chemotherapy with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP)-like chemotherapy and other chemotherapy. CHOP-like chemotherapy included all regimens with at least standard doses of doxorubicin, cyclophosphamide, vincristine and prednisone administered in 2- or 3-week cycles [13, 14]. After March 2001, all patients with advanced-stage DLBCL received rituximab in addition to standard CHOP chemotherapy (CHOPR) [15, 16]. CHOPR was not used for limited-stage DLBCL until January 2005. Patients received planned irradiation based on their clinical stage. Limited-stage patients received brief chemotherapy and planned irradiation. In addition, radiotherapy was also administered if spinal cord compression was present at diagnosis. Finally, if residual disease was still present, radiation was electively recommended at the discretion of the treating oncologist.
Response to treatment was recorded as complete (CR) if there was elimination of all initial lymphoma-related abnormalities. Persistent radiological abnormalities in the bone were common after treatment completion due to the slow remodeling process. Such persistent abnormalities were disregarded for purposes of assigning response status. Partial response (PR) required >50% reduction of all measurable components of the tumor. No response was assigned if there was no change in measurable non-osseous lesions and progressive disease if there was at least a 25% increase in size of any lymphomatous mass or the appearance of lymphoma at a new site.
statistical methods
OS was calculated from the date of diagnosis to the date of last follow-up or death from any cause. Progression-free survival (PFS) was calculated from the date of diagnosis to the date of documented disease progression or death from treatment toxicity. Disease-specific survival (DSS) was calculated from the date of diagnosis to the date of death due to lymphoma or its treatment. Patients alive without evidence of disease were censored on the date of their last follow-up and patients who died due to other causes unrelated to lymphoma were censored at the time of death. Survival curves were constructed according to the method of Kaplan and Meier [17] and compared using the log-rank test. Differences were considered significant if the P value was 
0.05 (two tailed). Multivariate analysis using Cox regression was used to determine the prognostic significance of the following factors: stage, stage group, B symptoms, spinal cord compression, LDH level, IPI score, IPI group, addition of rituximab or irradiation to standard chemotherapy. All survival analyses were performed using the SPSS software for Windows, version 11.0.1.
|
results |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
patient characteristics
A total of 131 patients with PBL were diagnosed between January 1983 and March 2005 with a median follow-up duration of 59 months for living patients (range 1–253 months). Seven patients, who remained free of disease at their last clinic visit, were lost to follow-up from 1 to 13 years after diagnosis. The clinicopathological and therapeutic data of the whole cohort and the DLCL subgroup (includes DLBCL, n = 95, and DLCL with no immunophenotyping, n = 8) are summarized in Table 1. The most common diagnosis was DLCL (n = 103). Other pathologic diagnoses were infrequent as shown in Figure 1. Immunophenotypic analysis was performed in 122 patients, of which 115 patients (94%) were of B type.
|
|
From the total group of 131, approximately one third (n = 42) presented with disease in the spine and one third (n = 41) in the long bones. Those with spinal disease had a variety of different histologies; DLCL (n = 31); follicular lymphoma (FL) (n = 5); T-cell lymphoma (n = 4); low-grade lymphoma, unclassifiable (n = 1) and mantle cell lymphoma (n = 1). Figure 2 shows the distribution of sites of presentation. Nine patients presented with a pathological fracture in long bones, of which six had disease in the humerus. Half of the 42 patients who presented with spine disease presented with spinal cord compression.
|
survival analyses of patients with primary DLCL of bone
For the 103 patients in the DLCL group (102 were assessable), the overall response rate was 84% (65% CR and 19% PR). The OS at 5 and 10 years were 62% and 41%, respectively (Figure 3). The 5- and 10-year PFS were 57% and 40%, respectively (Figure 4).
|
|
analysis of prognostic factors in primary DLCL of bone
Patients with Ann Arbor stages IE (n = 27), IIE (n = 19) and IV (n = 57) had 10-year OS of 60%, 27% and 38%, respectively (P = 0.043). The 10-year DSS was 76%, 27% and 38% for stages IE, IIE and IV, respectively (P = 0.008). Patients with stage IIE disease (n = 19) had mostly advanced-stage disease (n = 12) and were mostly (n = 14) treated with a full course of six to eight cycles of chemotherapy. Twenty-three (22%) patients were considered to have limited-stage disease and were treated with brief chemotherapy followed by irradiation. The 10-year OS for patients with limited- and advanced-stage disease were 56% and 37%, respectively (P = 0.048). Patients with B symptoms at presentation had a 10-year OS of 25% compared with 50% for those without (P = 0.015). Another important prognostic factor by univariate analysis was the LDH level at presentation. Patients with a normal and elevated LDH level at presentation had 10-year OS of 52% and 35%, respectively (P = 0.008). Patients in IPI groups 1, 2 and 3 had a 10-year OS of 66%, 50% and 18%, respectively (P < 0.00001) (Figure 5).
|
In a multivariate analysis of prognostic factors for patients with DLCL of bone including the variables of age above or below 60 years, the following distinctions retained independent prognostic impact: stage IE versus IIE or IV, limited versus advanced stage, normal versus elevated LDH and absence or presence of B symptoms. Only four patients in the age range below 60 had four or five IPI factors present, making prediction of outcome in this group unreliable. As shown in Figure 6, the other three groups, age <60 years with IPI 0–3 (n = 43), age
60 years with IPI 0–3 (n = 23) and age 60 with IPI 4–5 (n = 33) have markedly different 5-year OS of 90%, 61% and 25%, respectively (P < 0.00001).
|
Of the 31 patients who had DLCL in the spine, 18 presented with spinal cord compression. Neither presence of spinal cord compression nor pathological fracture at presentation had a significant impact on PFS or OS. Tumor site was not a significant prognostic factor, except when the skull was affected, which put patients at risk to succumb to central nervous system (CNS) lymphoma due to direct local extension into the brain. Two patients with brain involvement died and one out of the four patients with dura mater involvement died due to lymphoma.
analysis of impact of treatment in primary DLCL of bone
In total, 92 of the 103 patients (89%) with DLCL of bone received CHOP or CHOP-like chemotherapy with or without irradiation. In order to assess the possible contribution of irradiation, we compared the outcomes after combined modality therapy to that of chemotherapy alone. The 10-year OS in patients with advanced-stage disease who received both irradiation and chemotherapy (n = 41) was 25% compared with 56% for those who received chemotherapy alone (n = 39) (P = 0.025). The discretionary nature of the irradiation precludes firm conclusions concerning its usefulness; however, these results are not consistent with a beneficial effect of adding irradiation.
Starting in March 2001, all patients in BC with advanced-stage DLBCL received CHOPR. In an intention to treat analysis comparing the outcome of patients treated with curative intent for DLCL of the bone in the era before (n = 66) with after (n = 26) March 2001, there was a significant difference. Although the follow-up period is short, adding rituximab has dramatically improved the PFS. The 3-year PFS for patients who received CHOPR chemotherapy was 88% compared with 52% for those who received standard CHOP chemotherapy without rituximab (P = 0.025) (Figure 7).
|
failure of primary therapy in primary DLCL of bone
Figure 8 is a flowchart showing the outcome of patients including those whose lymphoma was not cured by the primary treatment. Six patients were <60 years of age when disease relapsed. Only one patient was eligible for high-dose chemotherapy with autologous stem cell transplant (ASCT) and eventually died due to disease progression. The other five patients were ineligible for high-dose therapy (two, CNS relapse; two, refusal and one relapsed as acute lymphoblastic leukemia).
|
outcome of patients with other histological subtypes of PBL
Fourteen patients had indolent B-cell lymphoma [FL = 7, marginal zone lymphoma (MZL = 4), small lymphocytic lymphoma (SLL = 2) one unclassifiable]. None of these patients had spinal cord compression at presentation and only one, a patient with SLL, presented with pathological fracture (humerus). All seven patients with FL received multiagent chemotherapy and two died due to lymphoma. Patients with MZL of the bone presented with disease in the orbital bone (n = 3) or the skull (n = 1). One patient died due to lymphoma.
Six patients with T-cell lymphoma of bone (four anaplastic large cell and two peripheral T-cell-unspecified) were treated with CHOP or CHOP-like chemotherapy. Three died due to disease progression and three are alive and well. There were two patients with lymphoblastic lymphoma. One was treated with allogeneic bone marrow transplant but died due to disease progression and the other received high-dose chemotherapy and ASCT and is alive 5 years after transplant. Two patients with Burkitt's and Burkitt-like lymphoma were treated with high-dose chemotherapy and ASCT, but both died due to lymphoma. One elderly patient with mantle cell lymphoma was treated palliatively and died due to disease progression.
|
discussion |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
PBL was first described as a distinct clinicopathological entity in 1939 by Parker and Jackson [18]. Because of the considerable difficulty distinguishing primary from secondary bone lymphoma, many previous studies were restricted to stage IE and IIE disease. We describe a relatively large group of patients (N = 131) with PBL diagnosed and staged in homogeneous manner and treated with era-specific treatment protocols with a long median follow-up period (59 months). Our study included all pathological types and disease stages and describes the effect of different treatment modalities, including rituximab. In contrast, previous studies described relatively small cohorts (n = 19–63), included mainly stage IE and IIE disease and did not report experience with rituximab. Table 2 summarizes relatively recent representative studies reported in the literature.
|
Two previous studies emphasized a predominance of long bone involvement [6, 8]; however, in our study only one third of cases had such involvement. We found that the areas of bone involvement closely parallel the areas of active bone marrow in most cases, typically flat bones or the ends of long bones (results not shown). A relatively larger proportion of our patients presented with stage IV disease. This may be due to routine inclusion of CT, MRI and radionucleotide scanning. As expected from previous studies, DLCL was the most common histologic type (79%).
Prognostic factors evaluated for the DLCL group in this study included age, stage, LDH level, IPI score, spinal cord compression and treatment. The presence of pathological fracture or spinal cord compression did not have an impact on PFS or OS. Unlike nodal DLBC, stage IIE disease in primary DLCL of bone appears to have a similar outcome to stage IV. Other studies have failed to show a survival difference between stages IE and IIE, but their results were confounded by small numbers [5]. In our study, patients with IPI scores 0–1 and 2–3 had similar outcomes which was markedly better compared with IPI scores 4–5. Multivariate analysis of prognostic factors identified three distinct groups, those with age <60 years with IPI 1–3 (n = 43), age 60 with IPI 0–3 (n = 23) and age 60 with IPI 4–5 (n = 33) with markedly different 5-year OS of 90%, 61% and 25%, respectively (P < 0.00001).
Before chemotherapy became available, the initial treatment of PBL was irradiation or surgery. Several studies have established that combination chemotherapy plus irradiation is superior to irradiation alone [5, 10]. Now that highly effective chemotherapy is available for DLBCL, it is reasonable to re-examine the role of irradiation. In our study, patients with advanced-stage disease who received chemotherapy plus irradiation actually had a poorer outcome compared with those who received chemotherapy alone (10-year OS were 25% and 56%, respectively). However, this difference must be very cautiously interpreted because the decision to use irradiation was individualized. It is possible that patients with more biologically aggressive disease were more likely to receive irradiation, obscuring its impact. This individualization of treatment precludes definitive interpretation of the role of radiation treatment of lymphoma of the bone in our series.
The introduction of rituximab since March 2001 has been accompanied by a remarkable improvement in the PFS for patients with primary DLCL of bone compared with those treated earlier without rituximab (3-year PFS 88% versus 52%). Although the median follow-up is only 28 months, this is similar to reports for DLBCL in general and suggests a real improvement in the outcome of this condition [16, 20].
In summary, this population-based, retrospective study demonstrates that the majority of PBL are of DLCL subtype. Long bones and flat bones are equally affected. The most important adverse prognostic factors at diagnosis include age above 60 years and IPI score 4 or 5. Neither presenting bony site of lymphoma nor presence of pathological fracture has any prognostic impact. Disease progression or relapse, especially in patients 60, confers a very poor prognosis. Our analysis does not support routine use of irradiation in advanced-stage DLCL of the bone. Finally, results from adding rituximab to chemotherapy are encouraging.
|
Acknowledgements |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
The authors thank Jane Donaldson and Stephanie Noon for their help with data management and the physicians of British Columbia for referral of patients and provision of follow-up information. This study was supported in part by the British Columbia Cancer Agency Mary Toye Memorial Fund and the Turner Family Lymphoma Outcomes Unit Fund.
Received for publication May 31, 2006. Revision received August 10, 2006. Accepted for publication August 11, 2006.
|
References |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionAcknowledgementsReferences |
|---|
1. Dahlin DC and Bone UK. (1986) Tumors 4th edition (Thomas, Springfield, IL) pp. 208–226.
2. Limb D, Dreghorn C, Murphy JK, Mannion R. (1994) Primary lymphoma of bone. Int Orthop 18:180.[Web of Science][Medline]
3. Freeman C, Berg JW, Cutler SJ. (1972) Occurrence and prognosis of extranodal lymphomas. Cancer 29:252.[CrossRef][Web of Science][Medline]
4. Rosenberg S, Diamond HD, Jaslowitz B, Craver LF. (1961) Lymphosarcoma: a review of 1269 cases. Medicine (Baltimore) 40:31.[Medline]
5. Dubey P, Ha CS, Besa PC, et al. (1997) Localized primary malignant lymphoma of bone. Int J Radiat Oncol Biol Phys 37:1087–1093.[CrossRef][Web of Science][Medline]
6. Heyning FH, Hogendoom PC, Kramer MHH, et al. (1999) Primary non-Hodgkin's lymphoma of bone: a clinicopathological investigation of 60 cases. Leukemia 13:2094–2098.[CrossRef][Web of Science][Medline]
7. Stein ME, Kuten A, Gez E, et al. (2003) Primary lymphoma of bone-A retrospective study. Oncology 64:322–327.[CrossRef][Web of Science][Medline]
8. Mulligan ME, McRae G, Murphey MD. (1999) Imaging features of primary lymphoma of bone. AJR Am J Roentgenol 173:1691–1697.[Abstract]
9. Pettit CK, Zuckerberg L, Gray MH, et al. (1990) Primary lymphoma of bone: A B-cell neoplasm with a high frequency of multilobated cells. Am J Surg Pathol 14:329–334.[Web of Science][Medline]
10. Baar J, Burkes RL, Bell R, et al. (1994) Primary non-Hodgkin's lymphoma of bone: a clinicopathological study. Cancer 73:1194–1199.[CrossRef][Web of Science][Medline]
11. Jaffe ES, Harris NL, Stein H, Vardiman JW. (2001) World Health Organization Classification of Tumors: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues(IARC, Lyon, France).
12. The International Non-Hodgkin's Lymphoma Prognostic Factors Project. (1993) A predictive model for aggressive non-Hodgkin's lymphoma. New Engl J Med 329:987–994.
13. Armitage JO, Corder MP, Leimert JT, et al. (1980) Advanced diffuse histiocytic lymphoma treated with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) without maintenance therapy. Cancer Treatment Rep 64:649–654.[Web of Science][Medline]
14. Hoskins PJ, Le N, Gascoyne RD, et al. (1997) Advanced diffuse large-cell lymphoma treated with 12 week combination chemotherapy: natural history of relapse after initial complete response and prognostic variables defining outcome after relapse. Ann Oncol 8:1125–1132.
15. Coiffier B, Lepage E, Herbrecht R, et al. (2000) Mabthera (rituximab) plus CHOP is superior to CHOP alone in elderly patients with diffuse large B-cell lymphoma (DLCL): interim results of a randomized GELA trial. Blood 96:223a (Abstr 950).
16. Coiffier B, Lepage E, Briere J, et al. (2002) CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. New Engl J Med 346:235–242.
17. Kaplan EL and Meier P. (1958) Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457–481.[CrossRef][Web of Science]
18. Parker F and Jackson H. (1939) Primary reticulum cell sarcoma of bone. Surg Gynecol Obstet 68:45.
19. Sehn LH, Donaldson J, Chhanabhai M, et al. (2005) Introduction of combined CHOP plus rituximab therapy dramatically improved outcome of diffuse large B-cell lymphoma in British Columbia. J Clin Oncol 23:5027–5033.
20. Fairbanks RK, Bonner JA, Inwards CY, et al. (1994) Treatment of stage IE primary lymphoma of bone. Int J Radiat Oncol Biol Phys 28:2363–372.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F H Heyning, P C W Hogendoorn, M H H Kramer, C T Q Holland, E Dreef, and P M Jansen Primary lymphoma of bone: extranodal lymphoma with favourable survival independent of germinal centre, post-germinal centre or indeterminate phenotype J. Clin. Pathol., September 1, 2009; 62(9): 820 - 824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-S. Chim, C. P. Bik Yiu, and W. H. Shek Pathological Bone Fracture in Non-Hodgkin's Lymphoma J. Clin. Oncol., July 20, 2007; 25(21): 3175 - 3176. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||