ORIGINAL ARTICLE. Abstract. Introduction

Transcription

1 ORIGINAL ARTICLE Prognostic Factors and Treatment Efficacy in Patients with Primary Diffuse Large B-cell Lymphoma of the Bone: Single Institute Experience Over Years Yao-Chung Liu,2, Jyh-Pyng Gau,, Yuan-Bin Yu,, Ying-Chung Hong,, Chueh-Chuan Yen,, Chun-Yu Liu,, Ta-Chung Chao,, Liang-Tsai Hsiao,, Jin-Hwang Liu,, Tzeon-Jye Chiou, and Cheng-Hwai Tzeng, Abstract Objective Primary bone lymphoma is a rare disorder, accounting for less than % of all cases of malignant lymphoma. Primary bony diffuse large B-cell lymphoma (PBDLBCL) is the most common histological type. In our study, a favorable response and lower risk of emergent surgery were observed following the administration of systemic chemotherapy with or without rituximab. Methods Patients diagnosed with malignant lymphoma at our hospital between 2000 and 20 were evaluated for PBDLBCL. Pertinent data, including the clinical presentation, histological type, treatment modalities, long-term outcome, survival curve and prognostic factors, were analyzed. Results Twenty-four patients with a median age of 6 years were diagnosed with PBDLBCL. A complete response (CR) was achieved in 8.% (n=) of the patients. With treatment of the disease, nine of patients with initially impending pathological bone fractures ultimately did not undergo surgery. The median follow-up duration was 8 months. Two patients experienced disease relapse. In a Kaplan-Meier analysis, the -year overall survival (OS) and disease-free survival (DFS) rates were 66.7% and 77.8%, respectively. In the univariate analyses, the significantly favorable prognostic factors for OS were an International Prognostic Index (IPI) score of <, an age of 60 years, a performance status with an Eastern Cooperative Oncology (ECOG) score of <2, a CR and chemotherapy 6 cycles. For DFS, a CR was the only favorable factor. In the multivariate analysis, a CR was the only independent factor for both OS and DFS. Conclusion Our study confirms the good prognosis of this rare disorder. Once a CR is achieved, even elderly patients may exhibit long-term survival, possibly obviating the need for surgery for less severe bone lesions. Key words: primary bone lymphoma, diffuse large B-cell lymphoma, chemotherapy, rituximab (Intern Med : 9-, 20) (DOI:.269/internalmedicine..0967) Introduction Primary bone lymphoma (PBL) is a rare manifestation of non-hodgkin lymphoma (NHL), accounting for less than % of all cases of malignant lymphoma and % of malignant bone tumors (-). Diffuse large B-cell lymphoma (DLBCL) is the most common pathological type of this rare disorder, representing over 80% of cases of primary bony NHL (). In 99, Parker and Jackson reported the first case series of PBL (). Since then, there has been some controversy regarding the definition of this disorder. Ostrowski et Division of Hematology and Oncology, Department of Medicine, Taipei Veterans General Hospital, Taiwan, Department of Medicine of Yang- Ming Branch, Taipei City Hospital, Taiwan, Faculty of Medicine, National Yang-Ming University, Taiwan and School of Medicine, National Yang-Ming University, Taiwan Received for publication May, 20; Accepted for publication August 8, 20 Correspondence to Dr. Jyh-Pyng Gau, jpgau@vghtpe.gov.tw 9

2 Intern Med : 9-, 20 DOI:.269/internalmedicine al. suggested that PBL be classified into four groups: solitary bone lymphoma, bone lymphoma with multifocal bony lesions, bone lymphoma with distant nodal disease and bone lymphoma with visceral disease (). The common definition of PBL refers to lymphoma confined to the bone and adjacent bone marrow without other concurrent systemic involvement (). However, a diagnosis of PBL is not excluded when regional lymph nodes are also involved in association with major bone lesion(s) (6). Although PBL is not categorized as a disease entity in the current WHO classification of lymphoid malignancy, the unique clinical features of this rare disorder deserve further clinical investigation. Our study focused on a patient population with a homogeneous diagnosis of DLBCL. In our study, the patients were older (median age >60 years) than that reported in previous studies, which demonstrated a median age of diagnosis of approximately 0-60 years (, 6-9). The aim of this study was to understand the disease course and identify prognostic factors affecting the treatment outcomes. Patients Materials and Methods All patients diagnosed with malignant lymphoma at Taipei Veterans General Hospital between January, 2000 and December, 20 were reviewed to identify cases of PBL. To focus on a homogeneous patient population, only patients with the pathological classification of DLBCL were recruited for the analysis. The diagnostic criteria for PBL include lymphomas involving a single bone lesion with or without involvement of the adjacent lymph nodes or lymphomas presenting with multiple bone lesions but exhibiting no evidence of distant nodal disease (, ). Solitary bone lesions with lymphomatous involvement of the distant bone marrow or other organs were defined as stage IV disease. The stage was designated based on the Ann Arbor staging system (). Accordingly, we classified primary DLBCL of the bone into four stages: stage IE includes solitary localized bone lesions; stage IIE includes localized lymph node involvement on the same side of the diaphragm; stage IIIE includes distant nodal disease and stage IV includes visceral disease or distant bone marrow involvement noted during the staging procedures. To prevent confusion in diagnosis between PBL and disseminated lymphoma with bone involvement, we deliberately excluded patients with stage IIIE PBL from our study. Pertinent clinical data, including age, gender, initial presenting symptoms, laboratory examination findings, pathologic diagnosis, involved organs, International Prognostic Index (IPI) score (, 2), detailed treatment modalities, time to death and the date of the last follow-up, were collected for the analysis. Additional workup examinations included bone marrow biopsies, plain X-ray films of the bone, computed tomography (CT), magnetic resonance imaging (MRI) of the bone lesion(s) and positron-emission tomography (PET). The retrospective review of the patients medical records was approved by the local Institutional Review Board of our hospital. Statistical methods and response evaluation: The treatment responses and end points were defined according to the International Working Group response criteria. A complete response (CR) was defined as the disappearance of the initial lymphoma-related abnormalities or lymphoma masses after treatment. MRI and CT were the most common tools used to evaluate the response after the completion of treatment, according to our institution s policy. When a residual tumor was detected on MRI or CT after the completion of treatment, a PET scan was obtained to identify whether a viable tumor remained. A partial response (PR) was defined as a reduction of over 0% in all measurable tumors. A non-response (NR) was defined as no change or a reduction of less than 2% in measurable non-osseous lesions. Disease progression was defined as an increase in size of at least 0% in the sum of the product of the diameters of the lymphoma masses or the development of any new lymphoma lesions (, ). Overall survival (OS) was evaluated from the date of diagnosis to the date of the last follow-up or death. Diseasefree survival (DFS) was calculated from the date of response assessment to the date of documented disease relapse or death. Survival curves were plotted according to the Kaplan- Meier method. In a univariate analysis of prognostic factors, a survival analysis was performed using the log-rank test, with a p value of <0.0 being defined as statistically significant. A multivariate analysis with a Cox regression proportion hazards model was used to determine the independent prognostic factors. All survival analyses were performed using the SPSS software program for Windows, version 7.0. Results After reviewing 2,8 patients with a diagnosis of malignant lymphoma, we identified 870 patients diagnosed with DLBCL between January, 2000 and December, 20. Among the patients with DLBCL, 2 patients with PBDLBCL were collected for the analysis. The estimated incidence of PBDLBCL in the patients with malignant lymphomas and DLBCL was.% and 2.76%, respectively. The patient s characteristics and treatments are presented in Table. There were men (8.%) and women (.6%), with a median age at diagnosis of 6 years (range: years). The most common initial presentation was bone pain (/2). Other complaints included masses or soft tissue swelling (n=6) and pathological fractures requiring immediate surgery (n=). patients presented with constitutional B symptoms. The sites of bone involvement included the extremities in 6% (/2) of the patients, including five patients with femur lesions, and the axial skeleton in % (/2) of the patients, including patients 96

3 Intern Med : 9-, 20 DOI:.269/internalmedicine Table. Patients Characteristics (n=2) Parameter Patient (n) % Sex (M/F) / 8./.6 Age, year Median Range 60 >60 Primary sites Extremities Tibia/fibula Femur Humerus Axial skeleton Spine Scapula Clavicle Sternum Initial symptoms or findings Pain Mass/swelling Operation of pathologic fracture B symptoms Ann Arbor Stage IE.6 IIE IV 8. Elevated ALK-P level.6 Elevated calcium level.2 Elevated LDH level IPI score Treatment regimen R-CHOP/R-CHOP-like.8 CHOP/CHOP-like.2 Additional radiotherapy 9 7. ALK-P alkaline phosphatase, LDH: lactate dehydrogenase, IPI: International Prognostic Index, R: rituximab, CHOP: cyclophosphamide, hydroxydaunorubicin, vincristin and prednisone with spine lesions. Fourteen patients had a low IPI (IPI 2). Most patients presented with elevated levels of alkaline phosphatase (ALK-P) and lactate dehydrogenase (LDH). However, an initial presentation of hypercalcemia was observed in only one patient who had hypercalcemia (corrected level:. mg/dl) and hypoalbuminemia (serum albumin: 2. gm/l) with initially normal levels of renal function parameters, serum sodium, potassium and intact parathyroid hormone (ipth). Despite the administration of intensive supportive care and the prompt initiation of chemotherapy, the clinical course of the disease was relentlessly aggressive, and the patient died.2 months later, representing the shortest OS among the six patients with bone marrow involvement. Sixteen patients had early disease of Ann Arbor stage I or II. All of these patients received chemotherapy consisting of CHOP (cyclophosphamide, hydroxydaunorubicin, vincristine and prednisone) or a CHOP-like regimen combined with local radiotherapy in 9 cases and rituximab in cases. Radiotherapy was performed to treat the residual tumors following chemotherapy. Among the nine patients treated with chemotherapy and local radiotherapy, six out the seven patients with stage I/II disease and two patients with stage IV disease received a full course of CHOP therapy followed by local radiotherapy for the primary tumor bed. One patient with stage II disease received local radiotherapy after only one cycle of CHOP therapy for intractable bone pain. The treatment response was a complete remission in the six patients treated with full-course CHOP therapy followed by radiotherapy. Fifteen of the 2 patients received six or more cycles of chemotherapy, which was a statistically significant factor favorably influencing OS in the univariate analysis (p=0.00). Among the 2 patients, five underwent emergent surgery for the bone lesions, including internal fixation for pathological fractures in long bones or laminectomy for compression fractures in the spine with neurological symptoms at diagnosis. In addition, of the remaining 9 patients (2.6%) were found to have impending pathological fractures or compression fractures on advanced imaging studies. However, after treatment with combined chemotherapy and radiotherapy, only one of the patients (%) ultimately received surgery for the pathological fracture. The response to treatment was evaluated in all patients, with a median follow-up of 8 months. After the initial treatment, 8.% (/2) patients achieved a CR (/2), 6.7% (/2) achieved a PR and 8.% (2/2) achieved an NR. Four patients died of uncontrolled infections during the initial treatment without adequate follow-up to evaluate the treatment response. Among the four patients who died of intractable infections, an older age (median: 78. years) was noted. In addition, three of the four patients had stage IV disease, including two patients with distant bone marrow involvement at the initial diagnosis. When comparing the patients who achieved a CR (n=) with those who did not (n=), a younger age (median age: 9. vs. 68. years), good performance status (Eastern Cooperative Oncology (ECOG) <2: 78.6% vs. 0%) and greater frequency of treatment with chemotherapy 6 cycles (92.9% vs. 20%) were found in the CR group. However, the frequency of rituximab use was almost equal between the two groups (0% in the CR group vs. 0% in the non-cr group). Only three of the patients with a CR received PET to evaluate the residual masses, which showed no evidence of fluorodeoxyglucose (FDG)- avid viable tumors in any case. The -year OS and DFS were 66.7% and 77.8%, respectively (Figure a, b). Distant progression or local recurrence was observed in two patients after a disease-free interval of.0 months in one patient and 9.6 months in the other patient. In addition to six or more cycles of chemotherapy, other favorable factors for OS in the univariate analyses included an IPI score of < (log-rank p=0.008), patient age ( 60 years) (log-rank p=0.0), a performance status with an ECOG score of <2 (log-rank p=0.08) and a CR at the end of treatment (log-rank p=0.00). For DFS, the achievement of a CR at the end of treatment was the only signifi- 97

4 Intern Med : 9-, 20 DOI:.269/internalmedicine a cantly favorable factor (log-rank p=0.009). The administration of chemotherapy for more than six cycles was not statistically significant as a favorable factor for DFS. The use of rituximab did not influence the outcomes in our study according to the log-rank test. The results of the univariate analyses are summarized in Table 2. According to the multivariate analyses including the parameters of age, achievement of a CR, the performance status, the IPI score and the administration of chemotherapy for more than six cycles, the achievement of a CR remained the only independent prognostic factor for both OS (Hazard ratio (HR):0.07; 9% Confidence interval (CI): ; p=0.0) and DFS (HR:0.09; 9%CI: ; p=0.08). b Discussion In our study, 92% (2/26) of all cases of PBL belonged to the DLBCL histological type, while primary DLBCL of the bone accounted for.% (2/2,8) of all malignant lymphomas and 2.76% (2/870) of all DLBCL cases diag- Figure. a: Overall survival of the 2 patients with primary DLBCL of the bone. b: Disease-free survival of the 8 patients with primary DLBCL of the bone. Table 2. Univariate Analyses for Prognostic Factors of DFS and OS (log-rank Test) Overall survival (OS) Disease-free survival (DFS) Factors n MS 9% CI p n MS 9% CI p Age (years) 60 > C/T cycles 6 <6 IPI score < CR Rituximab use Stage I/II IV Elevated LDH PS (ECOG) <2 2 Extranodal site <2 2 BM involvement MS: median survival (months), CI: confidence interval, HR: hazard ratio, C/T: chemotherapy, IPI: International Prognostic Index, CR: complete response, LDH: Lactic dehydrogenase, PS: performance status, ECOG: Eastern Cooperative Oncology, BM: bone marrow 98

5 Intern Med : 9-, 20 DOI:.269/internalmedicine nosed at our hospital within a period of years. This frequency is consistent with the figures noted in previous reports (, 6, 7). Other histological types of bone lymphoma reported in the literature include follicular, T-cell, mantle cell, anaplastic large cell and Burkitt s lymphoma (, 8, 9, ). The current WHO classification of lymphoid malignancy recognizes many subtypes of DLBCL with a predilection to occur in particular body sites, such as the brain, mediastinum and even legs. However, DLBCL of the bone is not yet regarded as a unique disease entity (6). Despite this, there were some characteristic features observed in our patients with PBDLBCL that deserve further discussion. Being a veterans general hospital, the average age of all patients at our hospital is older than that observed at other local medical centers. Accordingly, the median age of the patients in this study was 6 years, which is older than that noted in previous studies, reporting an average age at diagnosis of between 0 and 60 years, or even less than 0 years (9,, 7-20). In those reports, the -year OS in patients younger than 60 years of age was better than that observed in older patients. Only one article described patients with PBL with an average age at diagnosis of more than 60 years, with OS and progression-free survival rates at five years of 62% and 7%, respectively (8). Although the age of our patient population is older, the survival rates were promising, with -year OS and DFS rates of 66.7% and 77.8%, respectively. These results indicate that patients with this rare disorder have a high likelihood of achieving a cure. Once a CR is achieved, even elderly patients may exhibit long-term survival. This prognosis is seemingly better than that of other aggressive lymphomas with extranodal involvement. In addition to the five patients who initially required emergent surgery, the real incidence of surgery for those with impending pathological fractures was low (/) following successful treatment. Therefore, it is noteworthy that successful treatment with combined chemotherapy and radiotherapy obviated the need for surgical intervention in cases of less severe bone lesions. Several of the clinical features discussed below suggest that PBL originates from the bone and is a localized disease of the bone at the outset, which likely contributes to its relatively good prognosis after active treatment. A large cohort of adults with PBL indicated that the only identifiable prognostic indicators were localized disease and a younger age (2). In contrast to other lymphomas with dissemination, B symptoms are experienced much less frequently in PBL patients, indicating the inherent localized nature of the disease (6, 7). Hypercalcemia is commonly seen in cancer patients with bone metastasis. In contrast, hypercalcemia is rarely reported in PBL patients in the literature, another indicator of primary local disease rather than disseminated systemic disease. In our series, only one patient presented with hypercalcemic crisis resulting from advanced disease with extensive involvement of the spine, sternum, clavicle and bone marrow. That patient exhibited the worst OS among our patients with bone marrow involvement. Another intriguing observation is that despite its bony location and frequent invasion of the adjacent bone marrow, bone lymphoma does not necessarily disseminate to involve the bone marrow of the pelvic bones, as examined using routine staging procedures. Bone marrow involvement is observed in a subgroup of patients with disseminated disease and is characterized by worse survival (22, 2). In our study, only 2% (6/2) of the patients with PBDLBCL had distant bone marrow involvement, proven on bone marrow biopsies of the iliac crest. Among the six patients with bone marrow involvement, only two achieved a CR at the end of treatment, while the other four patients died of infection. The median OS among the patients with bone marrow involvement was 2.6 months (range:.2-9.). In this study, PBL was generally treated with multimodality treatment, and relatively poor results were achieved with local radiation therapy alone. However, a recent report of a prospective study from Australia showed high rates of survival and considerable local failure when the patients were treated with limited chemotherapy and shrinking field radiotherapy, lending support to the disease nature involving initially local bone disease with later systemic dissemination (2). Regarding other clinical manifestations, bone pain is the most common presentation, followed in order of frequency by palpable masses, pathologic fractures and neurologic symptoms. Previous studies have reported that the pelvis and/or femur are the most commonly presenting sites of PBL (2, 8, 9,, 7-20). In contrast, the spine (.6%) was the most common primary site in our study. Regardless, the site of involvement had no impact on either OS (p=0.296) or DFS (p=0.78). A slight trend of male predominance has been reported in some studies of bone lymphoma (, 6, 8, 7, 9). This study also included a majority of men (8.%). Previous studies have identified several prognostic factors, including solitary bone lesions or multifocal invasion, age, stage, IPI, the achievement of a CR, the treatment course, treatment with combined modalities or elevated LDH (8,, 7-20, 2). In the univariate analyses in our series, the significantly favorable prognostic factors for OS were an IPI score of <, an age 60 years, a performance status with an Eastern Cooperative Oncology (ECOG) score of <2, a CR and the administration of chemotherapy 6 cycles. For DFS, a CR was the only favorable factor. In the multivariate analysis, a CR was the only independent factor for both OS and DFS. When compared to the patients without remission (n=), the patients who achieved a CR (n=) were younger, had a better performance status and included a higher proportion (>90%, n=) of patients who completed the full course of chemotherapy ( 6 cycles).however,the rate of rituximab use was almost equal (7/ vs. /) between the two subgroups of patients. These results indicate that the achievement of a CR is likely related to age, the performance status and tolerance of a full course of systemic chemotherapy. We believe that the addition of rituximab to 99

6 Intern Med : 9-, 20 DOI:.269/internalmedicine the treatment regimen will improve the outcome (8, 8, 9). However, this speculation was not verified in our study. It is most likely that the small number of patients receiving rituximab resulted in inadequately low power to detect the statistical difference with the addition of this agent. In our study, neither stage I/II disease (p=0.) nor solitary bone lymphoma (p=0.0) were associated with survival benefits in the univariate analysis. However, this is likely due to the small number of patients, hence resulting in the lack of statistical power. Controversial results have also been reported in the literature. Ostrowski et al. reported that patients with solitary bone lesions had a much better OS than those with soft tissue invasion or associated nodal disease (). Other studies have shown no differences in OS between patients with early- and advanced-stage disease (8, 9, 9). Owing to the small patient number, the optimal treatment regimen for bone lymphoma is unknown. In view of the characteristics of local disease, radiotherapy may be of value, as demonstrated in some previous reports (, 7, 20). However, the addition of radiotherapy to chemotherapy did not influence OS in our study (p=0.9). The effects of radiotherapy were likely masked by those of active chemotherapy, as supported by the observation that a treatment courses of more than six cycles associated with the achievement of a CR determined the survival in our study. In both the univariate and multivariate analyses, the achievement of a CR was the most significant factor determining the OS and DFS. These results are consistent with those of previous reports (9, 20). Four patients died of intractable infections during the initial therapy. Among them, distant bone marrow involvement was detected in two patients (0%). We believe that the treatment-related mortality observed in these four patients was attributed to the patients older age (median age: 78. years), initial bone marrow involvement and advanced disease status. After chemotherapy, a poor marrow recovery due to old age and the initial bone marrow status resulted in a high infection rate. Conclusion In conclusion, our study shows that primary DLBCL of the bone has characteristic clinical features and demonstrates promising survival in elderly patients with this diagnosis. The administration of adequate chemotherapy with the achievement of a CR is the most important factor determining survival. Successful treatment with combined chemotherapy and radiotherapy can obviate the need for surgical intervention in cases of less severe bone lesions with pathological fractures. Early recognition of this possible diagnosis and the prompt initiation of systemic chemotherapy after confirming the diagnosis can cure the majority of patients with this rare disorder. The authors state that they have no Conflict of Interest (COI). Acknowledgement This work was partially supported by grants from the Taipei Veterans General Hospital (V2C-202) and Taiwan Clinical Oncology Research Foundation. References. Unni KK, Hogendoorn PCW. World Health Organization classification of tumours. In: Pathology and Genetics of Tumours of Soft Tissue and Bone. IARC Press, Lyon, 2002: Limb D, Dreghorn C, Murphy JK, Mannion R. Primary lymphoma of bone. Int Orthop 8: 80-8, 99.. Rosenberg S, Diamond HD, Jaslowitz B, Craver LF. Lymphosarcoma: a review of 269 cases. Medicine (Baltimore) 0:, Parker F, Jackson H Jr. Primary reticulum cell sarcoma of bone. Surg Gynecol Obstet 68:, 99.. Ostrowski ML, Unni KK, Banks PM, et al. Malignant lymphoma of bone. Cancer 8: , Mikhaeel NG. Primary bone lymphoma. Clin Oncol (R Coll Radiol) 2: 66-70, Bhagavathi S, Fu K. Primary bone lymphoma. Arch Pathol Lab Med : , Ramadan KM, Shenekier T, Sehn LH, Gascoyne RD, Connors JM. A clinicopathological retrospective study of patients with primary bone lymphoma: a population-based study of successively treated cohorts from the British Columbia Cancer Agency. Ann Oncol 8: 29-, Horsman JM, Thomas J, Hough R, Hancock BW. Primary bone lymphoma: a retrospective analysis. Int J Oncol 28: 7-7, Carbone PP, Kaplan HS, Musshoff K, Smithers DW, Tubiana M. Report of the committee on Hodgkin s disease staging classification. Cancer Res : , 97.. Advani RH, Chen H, Habermann TM, et al. Comparison of conventional prognostic indices in patients older than 60 years with diffuse large B-cell lymphoma treated with R-CHOP in the US Intergroup Study (ECOG 9, CALGB 979): consideration of age greater than 70 years in an elderly prognostic index (E-IPI). Br J Haematol : -, Abdelhamid T, Samra M, Ramadan H, Mehessin M, Mokhtar N. Clinical prognostic factors of diffuse large B cell non-hodgkin lymphoma: A retrospective study. J Egypt Natl Canc Inst 2: 7-2, 20.. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92: 20-26, Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol 2: 79-86, Beal K, Allen L, Yahalom J. Primary bone lymphoma: treatment results and prognostic factors with long-term follow-up of 82 patients. Cancer 6: , Harris NL. Mature B-cell neoplasms. In: World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Eds. IARC Press, Lyon, 2008: Zinzani PL, Carrillo G, Ascani S, et al. Primary bone lymphoma: experience with 2 patients. Haematologica 88: , Alencar A, Pitcher D, Byrne G, Lossos IS. Primary bone lymphoma--the University of Miami experience. Leuk Lymphoma : 9-9, Pellegrini C, Gandolfi L, Quirini F, et al. Primary bone lymphoma: evaluation of chemoimmunotherapy as front-line treatment in 2 patients. Clin Lymphoma Myeloma Leuk : 2-2, Cai L, Stauder MC, Zhang YJ, et al. Early-stage primary bone 0

7 Intern Med : 9-, 20 DOI:.269/internalmedicine lymphoma: a retrospective, multicenter Rare Cancer Network (RCN) Study. Int J Radiat Oncol Biol Phys 8: 28-29, Jawad MU, Schneiderbauer MM, Min ES, Cheung MC, Koniaris LG, Scully SP. Primary lymphoma of bone in adult patients. Cancer 6: , Kajiura D, Yamashita Y, Mori N. Diffuse large B-cell lymphoma initially manifesting in the bone marrow. Am J Clin Pathol 27: , Chang H, Hung YS, Lin TL, et al. Primary bone marrow diffuse large B cell lymphoma: a case series and review. Ann Hematol 90: , Christie D, Dear K, Le T, et al. Limited chemotherapy and shrinking field radiotherapy for Osteolymphoma (primary bone lymphoma): results from the trans-tasman Radiation Oncology Group 99.0 and Australasian Leukaemia and Lymphoma Group LY02 prospective trial. Int J Radiat Oncol Biol Phys 80: 6-70, Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-hodgkin s lymphoma. N Engl J Med 28: 02-06, The Japanese Society of Internal Medicine