Determining Response and Recurrence in Pediatric B-Cell Lymphomas of the Bone

Transcription

1 Pediatr Blood Cancer 2013;60: Determining Response and Recurrence in Pediatric B-Cell Lymphomas of the Bone Alexandra J. Borst, MD, 1 Lisa J. States, MD, 2 Anne F. Reilly, MD, 1 and Susan R. Rheingold, MD 1 * Background. Primary lymphomas of the bone are rare in children, but have an excellent response to therapy. Evaluating patients for remission and recurrence can be challenging given the difficulties of distinguishing healing bone from residual tumor on imaging. A review of imaging in patients treated for primary bone lymphoma (PBL) in one center was performed in an effort to determine best practice. Methods. Twelve cases of PBL diagnosed and treated from 2000 to 2011 at the Children s Hospital of Philadelphia were identified. Information about presentation, histology, imaging, treatment, and outcomes was collected. Results. There were no recurrences of the primary bone tumor after therapy. One patient developed therapy-related AML. Although PET-avid lesions usually fell below a SUV max of 3 within 3 months, low-level SUV max lesions often remained up to 12 months post-therapy. At no point during therapy did radiographs, MRI, bone scans or CT of bones normalize, most remaining abnormal for several months to years. Patients were exposed to significant ionizing radiation, with estimated levels ranging from 9.5 to msv per patient. Over 10% of scans had, which led to 17 extra imaging studies and 4 biopsies, but no clinically significant outcomes. Conclusions. In our cohort, frequent imaging did not affect or improve outcome. Due to the low risk of relapse and high rate of repeated imaging for, minimization of post-therapy imaging should be considered. This modification in practice will significantly reduce radiation exposure, as well as potentially decrease parent and patient anxiety. Pediatr Blood Cancer 2013;60: # 2013 Wiley Periodicals, Inc. Key words: bone; imaging; lymphoma; pediatric; radiation; recurrence INTRODUCTION Primary bone lymphoma (PBL) is a rare but highly treatable tumor in children [1], accounting for 1% of all non-hodgkin lymphomas (NHL) and 5 7% of primary bone tumors [1 3]. Most PBLs are B-cell NHL [2,4,5], with T-cell PBL being extremely rare, except in Japan [1]. In adults, most cases of PBL are diffuse large B- cell lymphoma [5]. In children, PBL is more heterogeneous and also includes lymphoblastic lymphoma (LL) and Burkitt lymphoma [5]. The most common site for PBL is the femur, with other frequent locations including the pelvis, humerus, head/neck, tibia, and vertebrae [1,3,6]. Half of cases involve a single bone [3]. The tumors are often diagnosed after months or even years of symptoms, suggesting an indolent course [3]. Children with PBL appear to have significantly better outcomes than adults [5]. This may have to do with differences in tumor biology, suggesting that pediatric and adult PBLs may be distinct entities [4,5]. Most studies report a 5-year overall survival of over 90% for patients treated with standard chemotherapy [4,5,7]. Some studies suggest a poorer prognosis for females, children under 9 years of age, or those having non-large cell histology or marrow involvement [3,7]. Prognosis is not influenced by the bone involved, presence of soft tissue mass, or pathologic fracture [3,7]. Monitoring of treatment response is challenging in patients with PBL, as healing bone is difficult to distinguish from diseased bone on imaging studies. Difficulties include distinguishing remaining tumor from bone remodeling, fibrosis, or even new non-malignant lesions. Imaging characteristics of PBL at diagnosis are well described [1,3,6,8,9] but there is no agreement on the best way to follow the tumors throughout treatment and remission. Traditional imaging studies such as radiographs and CT scan often show persistent bony abnormalities for years despite clinical remission [2,10]. Gallium scans have become outdated in bone tumor imaging and have been replaced by the much more sensitive PET/ CT scan [10]. MRI is more sensitive compared to CT and gallium scan [2,11] when imaging bone tumors and allows avoidance of ionizing radiation exposure. However, there is a high false-positive rate and some studies have shown that MRI signal abnormalities can C 2013 Wiley Periodicals, Inc. DOI /pbc Published online 20 March 2013 in Wiley Online Library (wileyonlinelibrary.com). persist for up to 2 years after resolution of disease [2,11]. MRI does not correlate well with extent of tumor fibrosis or intralesional vascularity, and has been shown to both over- and under-estimate presence of disease in PBL [8,9]. PET/CT scan shows a rapid decline in FDG activity with complete remission [10], but has a high false-positive rate, possibly due to bone remodeling, necrotic tumor, or use of growth factors [4]. In our institution, there is significant variability in the imaging practices of pediatric oncologists following children with PBL. Many are concerned about cumulative radiation exposure, as well as repeated use of sedation for imaging young children. The goal of this review was to examine imaging modalities and follow-up intervals used to evaluate patients with PBL in our institution and determine best practice. MATERIALS AND METHODS The Children s Hospital of Philadelphia (CHOP) electronic patient record and tumor registry databases were queried for all patients with a diagnosis of lymphoma involving bone from 2000 to Records were reviewed to determine presenting symptoms, initial and follow-up imaging, histologic subtype, sites of involvement, length of therapy, treatment protocol, complications, and outcome. The timing, indication, and results of each radiologic image were reviewed with attention to the clinical decision-making that occurred with each image and office visit. Actions initiated as a result of an imaging review were recorded. All in-house imaging 1 Division of Pediatric Oncology, The Children s Hospital of Philadelphia, Philadelphia, Pennsylvania; 2 Division of Pediatric Radiology, The Children s Hospital of Philadelphia, Philadelphia, Pennsylvania Conflict of interest: Nothing to declare. Correspondence to: Susan R. Rheingold, Colket Translational Research Building, RM 10052, 3501 Civic Center Boulevard, Philadelphia, PA rheingold@ .chop.edu Received 17 September 2012; Accepted 12 February 2013

2 1282 Borst et al. was read using American College of Radiology Practice Guidelines [12 14]. Five patients had initial imaging studies at outside institutions, which were re-reviewed at CHOP by an in-house radiologist whenever hard-copy images were available. Five outside MRIs, one outside CT, four outside X-rays and one outside bone scan were read solely by outside radiologists. With the exception of one patient, outside PET/CT scan images were reviewed by CHOP radiologists at the time of disease evaluation and by one author (L.S.). Images were defined by five time points. Time point 1 (TP1) included all initial imaging. Time point 2 (TP2) was the first imaging after start of therapy. Time point 3 (TP3) included all other imaging done on-therapy. Time point 4 (TP4) was the first offtherapy imaging. Time point 5 (TP5) included all remaining offtherapy imaging. Images obtained for purposes other than disease evaluation were excluded from analysis. Incidental were defined as potentially clinically significant imaging abnormalities that indicated a potential for relapse and were not simply changes in the initial lesion. Combined CT scans were listed as one scan when counting total number of scans performed, but radiation totals accounted for all radiation exposure to the multiple body sites. Calculations of estimated effective dose were based on a normalized dose per dose length product (DLP) for adult and pediatric patients of various ages with conversion factors (k) for pediatric patients assuming the use of a 16 or 32 cm phantom, depending on patient age. CT doses were calculated using guidelines from the Diagnostic Imaging Council CT Committee [15]. For nuclear medicine studies, the whole-body effective dose was generated using guidelines from the International Commission on Radiologic Protection [16]. Due to the minimal contribution to total radiation exposure and frequent use for nondisease evaluation, plain radiographs were not included in radiation totals. RESULTS Twelve patients, ages 4 18 years, with PBL were diagnosed and treated between 2000 and 2011 (Table I). Ten patients had diffuse large B-cell lymphoma (DLBCL) and two had B-cell LL. Six of the 10 DLBCL patients were staged as group B disease and treated per the Children s Oncology Group trial CCG-5961 Arm B4 [17,18] or ANHL01P1 group B therapy [19], standard DLBCL therapy for the time period. Four patients were considered to have Group C disease, two with bone marrow (BM) involvement, one with liver metastases, and one with both BM and liver metastases. No patient had CNS involvement. The patients with Group C disease were treated like ANHL01P1, three with rituximab. The two patients with pre-b cell LL were treated like COG A5971 regimen B [20]. In all patients, X-ray was used as the initial imaging modality. The median time to diagnostic biopsy from initial abnormal imaging was 15.5 days (mean ¼ 103 days, range: 1 1,004 days). The median time from diagnostic biopsy to initiation of chemotherapy was 10.5 days (mean ¼ 15.3 days, range: 1 67 days). Four patients with DLBCL required repeat biopsy for confirmation of the diagnosis, leading to a delay in therapy initiation. Initial staging evaluation varied, with bone scans being replaced by PET/CT scans in more recent years. Follow-up imaging included MRI in 11 patients, CT scan in 10 patients, PET/CT scan in 11 patients, bone scan in 5 patients, and gallium scan in 2 patients. Total number of radiologic examinations (Table II) from TP1-TP5 TABLE I. Patient Demographics Patient Age at diagnosis (years) Gender Diagnosis Involvement Therapy # Months of follow-up Recurrence #1 10 F DLBCL 1 primary (tibia), BM negative CCG-5961, group B 82 No #2 17 M DLBCL 1 primary (scapula), possible ANHL-01P1-like, group B 65 No involvement at T10 #3 8 F DLBCL Diffuse skeletal (>5 lesions), BM ANHL-01P1-like, group C 39 No involved #4 14 M DLBCL Diffuse skeletal (>5 lesions), BM ANHL-01P1-like, group B 37 No negative #5 18 M DLBCL Diffuse skeletal (>5 lesions) plus ANHL-01P1-like, group C 35 No liver metastases, BM negative #6 14 F DLBCL More than 1 primary (vertebrae) ANHL-01P1-like, group C 27 No plus liver metastases, BM involved #7 13 F DLBCL Diffuse skeletal (>5 lesions), BM positive ANHL-01P1-like, group C 26 Developed secondary AML #8 12 M DLBCL More than 1 primary (tibia and CCG-5961, group B 26 No calcaneus), BM negative #9 15 M DLBCL 1 primary (femur), BM negative ANHL-01P1-like, group B 7 No #10 18 M DLBCL 1 primary (femur), BM negative ANHL-01P1-like, group B 6 No #11 6 F LL More than 1 primary (maxilla and femurs), BM negative COG A-5971, regimen B2 120 No #12 4 M LL Diffuse skeletal (>5 lesions), BM involved (chemo and cranial XRT) COG A-5971, regimen B1 5 (still on therapy) DLBCL, diffuse large B-cell lymphoma; LL, precursor B lymphoblastic lymphoma; M, male; F, female; BM, bone marrow. No

3 Pediatric B Cell Lymphomas of the Bone 1283 TABLE II. Imaging Studies # CT* # PET/CT # Bone scan # Gallium scan # MRI # X-ray Total # radiologic studies Total estimated effective dose (msv) Patient # Patient # Patient # Patient # Patient # Patient # Patient # Patient # Patient # Patient # Patient # Patient # Total Median Mean Range Number of CT scans combined scans (e.g., chest, abdomen, pelvis) count as 1 scan for totaling number of scans, but radiation totals are based on multiple body sites. ranged from 11 to 61; and 2 30 studies were done off-therapy (TP4 TP5) at the time of data collection. There was no consistent pattern of imaging across the 12 patients. All but one patient (with LL) had some form of follow-up imaging of their bony sites within 2 months of starting therapy (TP2). Four patients were imaged with MRI and six were imaged with PET/CT at TP2; one patient had both. Five patients had other types of imaging done at this time point as well. During therapy (TP3), the DLBCL patients were generally imaged every 1 4 months. At no point during therapy did radiographs, MRI, bone scans, or CT of bones normalize in appearance for any patient. For the 11 patients who completed therapy, the first off-therapy imaging (TP4) was done within 1 month. Two patients were imaged with MRI, six patients with PET/CT, and three patients with both MRI and PET/CT. Five patients had additional types of imaging. Off-therapy scans (TP5) were planned approximately every 3 months for the DLBCL patients for the first months, after which imaging became less frequent (every 4 6 months), if performed at all. Time to normalization by imaging type varied greatly. In general PET/CT scans reverted most quickly to normal, defined as SUV max < 3. There were a total of 49 PET/CT scans performed. SUV max for initial scans was on average 6.4 (range: ) for the primary lesions. TP2 and TP3 scans had an average SUV max of 3.0 (range: 2 3.8). TP4 and TP5 scans had an average SUV max of 2.5 (range: ). Three patients still had SUV max > 3.0 at completion of therapy. Radiographs, CT scans, and MRIs took the longest to normalize given the continued bone healing seen on these modalities. Eight of the 12 patients had at least two MRIs. In three patients, MRIs of the bone were finally read as normal at 24, 59, and 64 months from therapy initiation. Five patients still had abnormal bone MRI imaging at the time of last MRI (4, 7, 24, 26, and 30 months from start of therapy). Bone or gallium scans were followed in three patients, and normalized in two patients, one at 6 months into therapy and the other at 6 months off therapy. Bone scan was still abnormal in one patient when it was last performed at 4 months off therapy. CT scans were used more commonly as part of an initial staging work-up, with only five patients having CT scans after diagnosis and only one having repeated CT scans specifically to follow bony lesions. Other patients had repeat CT scans to monitor for metastases or lung nodules, although changes in bone lesions were also noted on these scans. In a single patient there was CT normalization of the bone 8 months off therapy. Five patients had a soft tissue mass or local edema surrounding their bone lesion at diagnostic imaging. In three, the soft tissue component resolved by TP2 post-induction imaging. Two patients had longer resolution periods, both 7 months from start of therapy. Patients with BM disease at diagnosis all had normal BM examinations by 5 weeks. Imaging performed during and after completion of therapy found 32 in seven patients (Table III). Incidental consisted of with concerns of recurrence such as lymphadenopathy, mediastinal widening from thymic rebound, or new bone abnormalities. Eighteen percent of PET/CT scans, 19% of MRIs, and 12% of CT scans had. There were no as a result of bone scan or radiograph. Overall, 11% of imaging was associated with. This was associated with an estimated additional msv of radiation. MRIs commonly showed new bone abnormalities at separate sites from the original bone lesion. Seven of 15 MRIs with led to three extra MRIs, three extra PET/CT scans, and two additional bone biopsies. PET/CT scans revealed new sites of increased FDG uptake and in many cases led to more imaging of that site. None of the on PET/CT were found to be clinically significant, but they led to an additional four CT scans, one ultrasound, one plain radiograph, and two additional biopsies to rule out recurrence. Three patients had imaging performed due to musculoskeletal complaints and one for headaches and visual changes; all with concerns of recurrence. In all patients imaging was reassuring or

4 1284 Borst et al. TABLE III. Incidental Findings Modality Total # of scans performed # of scans with Percent with Estimated additional scans due to Estimated additional radiation exposure Types of Clinical interventions other than further imaging CT msv Lymphadenopathy, None lung nodule, thymic rebound PET/CT msv FDG avid signal 2 biopsies in lungs liver, mediastinum, and presumed lymph nodes Bone or gallium scan 18 0 NA 0 NA NA None MRI NA Abnormal bone 2 biopsies signals at other bone sites, abnormal enhancement in vertebral body X-ray 94 0 NA 0 NA NA None Total msv Number of CT scans combined scans (e.g., chest, abdomen, pelvis) count as 1 scan for totaling number of scans, but radiation totals are based on multiple body sites. found alternative etiologies (i.e., sinusitis). In three patients, concerns on two PET/CTs and two MRIs during post-therapy imaging led to four repeat biopsies, all of which were negative for malignancy. Of note, none of these patients had clinical symptoms or bone pain at the abnormal imaging site. One patient showed a slight increase in FDG uptake on PET/CT and another had new hyperintense bony foci seen on MRI. The patient who had two repeat biopsies had multiple bone abnormalities in new locations on two separate MRIs, which were associated with low level PET/CT positivity (Fig. 1). Although the second biopsy was not recommended, the family insisted for peace of mind. All biopsies were negative for recurrence and revealed necrosis, ablated BM, and/or fibrosis. Total estimated radiation exposure from CTs, PET/ CTs, bone scans, and gallium scans is listed in Table II. Radiation exposure ranged from 9.5 to msv with a mean of 97.6 msv per patient. There have been no recurrences or relapses to date. One patient did develop therapy-related AML, diagnosed by peripheral blasts on a routine CBC. DISCUSSION PBL presents oncologists with a variety of challenges, particularly with regards to surveillance imaging following remission. Once therapy is initiated, it is unclear how best to monitor patients for response to therapy and screen for potential recurrence. In the context of growing concerns over radiation exposure, especially in a pediatric population, careful attention should be paid to the frequency and type of monitoring used. This is particularly critical given the ambiguity of imaging results during the period of highest risk follow-up, namely the first year off therapy. Our cohort of patients provides an example of these difficulties and calls into question follow-up that emphasizes regular imaging, regardless of radiologic modality used. Our cohort of patients was similar to previously studied groups in presentation and site of lesions, but different in age and survival. Also consistent with previous was the rapidity of soft tissue resolution and high rate of persistent abnormalities on MRI and CT scan despite clinical remission [2,11]. There are no definitive guidelines for frequency of imaging in pediatric PBL, but some protocols provide recommendations [17 19]. On-therapy imaging intervals in our PBL patients typically followed the timelines suggested by the clinical protocols. All patients received off-therapy imaging with follow-up approximately every 3 months, likely based on imaging practices for other NHL subtypes. There was no clear pattern to use or timing of particular imaging modality, although there was a trend over time towards less use of MRI, bone scan, and CT, and more reliance on PET/CT. This likely reflects the increased availability of PET/CT imaging as well as greater understanding of the drawbacks of the other modalities. Incidental were common, led to more frequent imaging, and almost assuredly increased anxiety for patients and families. One in five PET/CT scans and MRIs had concerning for recurrence, and led to earlier than planned imaging or biopsy in about half of those cases. All repeat biopsy results were negative for lymphoma and none of these patients has experienced a recurrence. None of the patients had complications as a result of the biopsy procedure, but all had significant periods of immobility postoperatively. The data on PBL recurrence primarily come from the adult oncology literature, which typically uses radiation and chemotherapy for treatment [4,7]. Two studies looking at pediatric cohorts found significant differences in recurrence [3,5], but neither of these

5 Pediatric B Cell Lymphomas of the Bone 1285 Fig. 1. A: 18F-FDG PET image in the coronal plane demonstrates a heterogeneous pattern of mild increased activity (SUV ¼ 2.2) in the right proximal tibia (oval). B: Coronal contrast-enhanced MRI of tibias reveals heterogeneous enhancement at the same site (oval). Biopsy revealed fibrosis and no evidence for tumor. studies commented on the role of imaging in detecting relapse. We did not find any studies describing the utility of surveillance imaging in detecting relapse in PBL, but a recent study by Voss et al. [21] described surveillance imaging in Hodgkin s lymphoma. They found that detection of relapse after the 1st year off-therapy, either through routine imaging or symptom surveillance, did not change overall survival and was likely overused [21]. None of the 10 DLBCL patients described here have had a recurrence to date. The limited data available on frequency and site of recurrence makes it difficult to recommend routine non-primary site screening. Given the lack of useful information, the frequent occurrence of, and the low risk of relapse, frequent imaging performed for patients with PBL may be unnecessary. In addition to the time, resources, and anxiety involved, there is the concern of excess radiation exposure. Based on the average radiation dose with each type of imaging modality, we estimated that our patients were exposed to an average of 97.6 msv during their treatment course ( msv). Estimates for high exposure range from 20 to 50 msv/year, which is the upper annual occupational exposure limit for workers [22,23]. For comparison, the US Nuclear Regulatory Commission recommends that the annual occupational dose limit for nuclear workers be limited to 50 msv and that the exposure for minors be 5 msv. This is in comparison to an average US annual dose of 6.2 msv and annual natural background dose of 3.1 msv [24]. While the long-term risks of this kind of radiation exposure are uncertain, several studies have looked at radiation doses from diagnostic imaging and the subsequent risk of cancer [25 30]. Some have even claimed that the risk is enhanced in children with primary malignancies because of the frequency of imaging and the risks of therapy itself [27,28]. However, more recent evidence calls into question the idea that excess ionizing radiation exposure can predict increased risk of malignancy [31,32]. Although information about radiation exposure cannot be used to predict increased risk for any patient or cohort of patients, the information can be used to create protocols and standards that will have less radiation exposure and less risk [32]. Based on the experience with this cohort of patients, we suggest there is little role for frequent imaging of pediatric patients with PBL, and recommend off-therapy imaging for symptoms only. PET/CT is likely to be the most useful modality in this setting given its relatively fast time to resolution. MRI may also be useful in patients with musculoskeletal symptoms as other etiologies may be discovered. We would recommend continued on-therapy imaging at least until resolution of all soft tissue and metastatic non-bony components, and remind clinicians that bony abnormalities at tumor sites may remain abnormal for years. Although we will be adopting this practice in our institution, larger cooperative group studies may better confirm patterns of relapse and best imaging procedures.

6 1286 Borst et al. REFERENCES 1. Krishnan A, Shirkhoda A, Tehranzadeh J, et al. Primary bone lymphoma: Radiographic-MR imaging correlation. Radiographics 2003;23: MengiardiB, HoneggerH, HodlerJ, etal. Primary lymphomaofbone: MRIandCTcharacteristics during and after successful treatment. Am J Roentgenol 2005;184: Glotzbecker MP, Kersun LS, Choi JK, et al. Primary non-hodgkin s lymphoma of bone in children. J Bone Joint Surg Am 2006;88: Reddy N, Greer JP. Primary bone lymphoma: A set of unique problems in management. Leuk Lymphoma 2010;51: Zhao XF, Young KH, Frank D, et al. Pediatric primary bone lymphoma Diffuse large B-cell lymphoma: Morphologic and immunohistochemical characteristics of 10 cases. Am J Clin Pathol 2007;127: Mulligan ME, McRae GA, Murphey MD. Imaging features of primary lymphoma of bone. Am J Roentgenol 1999;173: BealK, AllenL, YahalomJ. Primary bonelymphoma: Treatment resultsandprognosticfactorswithlongterm follow-up of 82 patients. Cancer 2006;106: HeyningFH, KroonHMJA, HogendoornPCW, etal. MRimagingcharacteristics inprimarylymphomaof bone with emphasis on non-aggressive appearance. Skeletal Radiol 2007;36: WhiteLM, Schweitzer ME, KhaliliK, etal. MRimagingofprimarylymphomaofbone: VariabilityofT2- weighted signal intensity. Am J Roentgenol 1998;170: Okada M, Sato N, Ishii K, et al. FDG PET/CT versus CT, MR imaging, and 67Ga scintigraphy in the posttherapy evaluation of malignant lymphoma. Radiographics 2010;30: Stroszczynski C, Oellinger J, Hosten N, et al. Staging and monitoring of malignant lymphoma of the bone: Comparison of 67Ga scintigraphy and MRI. J Nucl Med 1999;40: American College of Radiology (ACR). ACR Practice Guideline for Performing and Interpreting Magnetic Resonance Imaging (MRI) [online publication]. Reston, VA: American College of Radiology (ACR); p. 13. American College of Radiology (ACR), Society for Pediatric Radiology (SPR). ACR-SPR Practice Guideline for Performing FDG-PET/CT in Oncology [online publication]. Reston, VA: American College of Radiology (ACR); p. 14. American College of Radiology (ACR). ACR Practice Guideline for Performing and Interpreting Diagnostic Computed Tomography [online publication]. Reston, VA: American College of Radiology (ACR); p. 15. American Association of Physicists in Medicine. The measurement, reporting and management of radiation dose in CT. Report 96. AAPM Task Group 23 of the Diagnostic Imaging Council CT Committee. College Park, MD: American Association of Physicists in Medicine, ICRP. Environmental Protection The concept and use of reference animals and plants. ICRP Publication 108. Ann ICRP 2008;38: Patte C, Auperin A, Gerrard M, et al. Results of the randomized international FAB/LMB96 trial for intermediate risk B-cell non-hodgkin lymphoma in children and adolescents: It is possible to reduce treatment for the early responding patients. Blood 2007;109: Cairo MS, Gerrard M, Sposto R, et al. Results of a randomized international study of high-risk central nervous system B non-hodgkin lymphoma and B acute lymphoblastic leukemia in children and adolescents. Blood 2007;109: Shiramizu B, Goldman S, Kusao I, et al. Minimal disease assessment in the treatment of children and adolescents with intermediate-risk (stage III/IV) B-cell non-hodgkin lymphoma: AChildren s Oncology Group report. Br J Haematol 2011;53: Termuhlen AM, Smith LM, Perkins SL, et al. Outcome of newly diagnosed children and adolescents with localized lymphoblastic lymphoma treated on Children s Oncology Group trial A5971: A report from the Children s Oncology Group. Pediatr Blood Cancer 2012;59: ; doi: /pbc Voss SD, Chen L, Constine LS, et al. Surveillance computed tomography imaging and detection of relapse in intermediate- and advanced-stage pediatric Hodgkin s lymphoma: A report from the Children s Oncology Group. J Clin Oncol 2012;30: ICRP. The 2007 recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Ann ICRP 2007;37: Fazel R, Krumholz HM, Wang Y, et al. Exposure to low-dose ionizing radiation from medical imaging procedures. New Engl J Med 2009;361: United States Nuclear Regulatory Committee. NRC Regulations: Subpart C Occupational dose limits. NRC Library Documents Collection. part html Published May 21, Accessed November 15, Amis ES, Jr., Butler PF, Applegate KE, et al. American College of Radiology white paper on radiation dose in medicine. J Am Coll Radiol 2007;4: Brody AS, Frush DP, Huda W, et al. Radiation risk to children from computed tomography. Pediatrics 2007;120: Chawla SC, Federman N, Zhang D, et al. Estimated cumulative radiation dose from PET/CT in children with malignancies: A 5-year retrospective review. Pediatr Radiol 2009;40: Hall EJ. Lessons we have learned from our children: Cancer risks from diagnostic radiology. Pediatr Radiol 2002;32: KleinermanRA. Cancer risksfollowingdiagnostic andtherapeutic radiation exposurein children. Pediatr Radiol 2006;36: Pearce MS, Salotti JA, Little MP, et al. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: A retrospective cohort study. Lancet 2012;380: ; doi: /S (12) Hall DJ, Goodwin M, Clarke T. Lifetime exposure to radiation from imaging investigations. Can Fam Physician 2006;52: HendeeWR, O ConnorMK. Radiationrisks ofmedicalimaging: Separatingfact fromfantasy. Radiology 2012;264: