Purpose/Objective(s) This prospective phase II trial was designed to assess disease control and late effects following proton radiotherapy in children and young adults with non-metastatic Ewing Family of Tumors (EFT) or other bone sarcomas and non-rhabdomyosarcoma soft tissue sarcomas (NRSTS). Materials/Methods Sixty-nine pediatric and young adult patients (age<30) with NRSTS or bone sarcomas were enrolled between September 2006 and July 2016. Chemotherapy and surgery were administered to patients in accordance with current standard practices or according to co-enrollment on cooperative group protocols. Results Median follow-up was 7.7 years (range 0.4-15.7). 39 patients had EWS sarcoma, 5 patients had osteosarcoma, and 25 patients had other histologies. Eight-year overall survival (OS), event-free survival (EFS), local regional control (LC), and distant control (DC) rates for the whole cohort were 77.5%, 67.2%, 82.1%, and 87.9%, respectively and 81.0%, 76.8%, 92.3%, and 89.7% for EWS and 73.2%, 56.5%, 69.7%, and 86.3% for other histologies. The median age at the time of radiation therapy was 12.3 years (range 1-22). The most common primary tumor locations were head and neck (42%); abdomen/pelvis (26%) and paraspinal (23%). Seventy percent of patients had gross residual disease at the time of radiotherapy. The median radiation dose was 55.8GyE (range 41.4-72). Patients with a GTR had no difference in EFS (8-year: 69.9% vs 65.8%, p = 0.9) or LC (8-year: 74.7% vs 85.3%, p = 0.4). The non-significant relationship between surgical extent and EFS/LC did not change across histology (among EWS, 8-year EFS was GTR: 85.7% vs STR 75.0%, p = 0.6 and 8-year LC was GTR: 85.7% vs STR 93.8%, p = 0.5; among all other histologies, 8-year EFS was GTR: 64.3% vs STR 49.2%, p = 0.6 and 8-year LC was GTR: 71.4% vs STR 68.8%, p = 0.8). Eight-year follow-up toxicity assessments showed 22 patients (31.9%) with at least one grade 2 toxicity and 13 patients (18.8%) with at least one grade 3 toxicity, both at least partially attributable to radiation therapy. There were no grade 4 toxicities reported. Conclusion Eight-year OS, EFS, LC, and DC rates were similar to those observed in comparable studies of photon radiation. Proton radiation represents a safe and effective radiation modality for pediatric non-rhabdomyosarcoma bone and soft tissue sarcomas.

Purpose/Objective(s) The objective of this study was to update a previously reported retrospective series with long-term outcomes on the efficacy and safety of re-irradiation with proton therapy for recurrent ependymoma. Materials/Methods With IRB approval, we searched our pediatric database to identify children with a diagnosis of ependymoma who received at least two courses of radiation therapy. Only patients who received proton therapy for their re-irradiation were included in our analysis. Progression free survival (PFS) and overall survival (OS) were calculated using the Kaplan-Meier method from the start date of the second radiation. Results The median age at re-irradiation was 8.5 years (n = 35). The majority of patients were female (57%) with anaplastic histology (71%), infratentorial tumors (69%), and received a median of 54.0 Gy (52.2-59.4) with proton radiation therapy (86%) for their initial treatment. The median time to first failure was 2.39 years from the initial treatment. The most common site of first failure was local (77%) followed by distant (14%) and both (9%). Salvage therapy typically included re-resection (83%), chemotherapy (57%), and proton radiation therapy (100%) to a median dose of 52.2 GyE (14.0 – 58.0). The median time to second failure was 1.16 years from the second treatment. The most common site of second failure was local (64%) followed by distant (32%) and both (4%). The median follow up was 5.22 years (0.9-14.0). Six-year OS and PFS are 55.5% (39.9-76.2) and 25.5% (13.5-45.1). An improved OS was associated with patients who did not receive chemotherapy as part of the treatment for their first recurrence (HR 3.17, p = 0.014). Four of 30 patients (13.3%) locally re-treated experienced grade 2 radiation-associated treatment change and 1 patient (3.3%) experienced a grade 3 radiation-associated treatment change. Conclusion Among a cohort of patients with long-term follow up, proton therapy remains a safe and effective option for re-irradiation in the treatment of recurrent ependymoma and can be an effective salvage strategy for patients with recurrent disease.

Denosumab has been advocated as a potential treatment for the rare skeletal disorder fibrous dysplasia (FD); however, there is limited data to support safety and efficacy, particularly after drug discontinuation. We report a case of successful treatment of aggressive craniofacial FD with denosumab, highlighting novel insights into the duration of efficacy, surrogate treatment markers, and discontinuation effects. A 13-year-old girl presented with persistent pain and expansion of a maxillary FD lesion, which was not responsive to repeated surgical procedures or bisphosphonates. Pre-treatment biopsy showed high RANKL expression and localization with proliferation markers. Denosumab therapy was associated with improved pain, decreased bone turnover markers, and increased lesion density on computed tomography scan. During 3.5 years of treatment, the patient developed increased non-lesional bone density, and after denosumab discontinuation, she developed hypercalcemia managed with bisphosphonates. Pain relief and lesion stability continued for 2 years following treatment, and symptom recurrence coincided with increased bone turnover markers and decreased lesion density back to pre-treatment levels. This case highlights the importance of considering the duration of efficacy when treating patients with FD and other nonresectable skeletal neoplasms that require long-term management.

Proton radiotherapy (PRT) is being increasingly used in the United States for pediatric medulloblastoma due to its ability to decrease the incidence and severity of late effects. The purpose of this study is to report mature clinical outcomes, including disease control and late toxicity, for pediatric patients treated with PRT for medulloblastoma. 180 patients with standard, intermediate (large cell/anaplastic histology but otherwise standard risk) or high risk medulloblastoma were treated with surgery, adjuvant PRT, and chemotherapy at an institution between February 2002 and November 2016. Patients underwent craniospinal irradiation of 18-36 Gy radiobiological equivalents (GyRBE) at 1.8 GyRBE per fraction followed by a boost dose to the whole posterior fossa or tumor bed. 131 (72.8%) patients were enrolled on a prospective clinical trial. Event Free Survival (EFS) and Overall Survival (OS) was assessed using the Kaplan Meier Method. The cumulative incidence of brainstem injury and secondary malignancies was also evaluated. The median follow up was 8.1 years (1-16.3). The median age at start of PRT was 8.1 years (2.5-24.1). The percentage of patients undergoing a gross total resection (GTR) or near-GTR/NTR, subtotal resection (STR), and biopsy only was 88.3%, 11.1%, and 0.6%, respectively. 58.3% of patients were categorized as standard risk and 41.7% of patients as intermediate/high risk. The 7-year EFS and OS for the entire cohort was 76.9% and 80.5%, respectively. The 7-year EFS and OS was 85.1% and 87.5% for the standard risk cohort and 65.4% and 70.5% for intermediate/high risk patients. On univariate analysis, intermediate/high risk disease (HR: 2.0; p = 0.02) was associated with inferior EFS, while anaplastic subtype (HR: 1.9; p = 0.07) trended towards worse EFS. The 7-year cumulative incidence of both secondary tumors and brainstem injury was 2.0%. PRT for medulloblastoma is effective, with disease control in our cohort comparable to other published data. Furthermore, it is associated with low rates of grade II or higher brainstem injury and secondary malignancies. Strategies to improve survival outcomes for high risk patients are needed.

Children treated with radiotherapy for pediatric brain tumors are at risk for late effects, including neurocognitive effects, endocrine alterations, hearing deficiencies, and impaired HRQOL. Proton radiotherapy (PRT) should partially mitigate these adverse effects of treatment. Here we report health and PedsQL HRQOL outcomes for pediatric brain tumor patients treated on a prospective phase II PRT protocol. Eligibility criteria are as follows: age 1-25 years and potentially curable brain tumors requiring local field radiotherapy. PRT doses used were 50.4-59.4 GyRBE. Participants had baseline radiologic imaging, endocrine, audiologic, and neurocognitive evaluations during PRT and at specified follow-up intervals. Toxicities were scored in accordance with CTCAE v4 except for hearing loss (HL), which was scored present if hearing decibel thresholds were 25+ at any frequency. Hearing evaluations were only required if cochlea dose exceeded 20 GyRBE. One hundred patients were enrolled from 2011-2017. Diagnoses include glial tumors (23), ependymoma (41), craniopharyngioma (20), and other (16). Median follow-up and age was 4.0 and 8.0 years, respectively. Five-year EFS and OS was 84% and 94% for the whole cohort. Five-year EFS for gliomas, ependymoma, craniopharyngioma, and other patients was 94%, 68%, 100%, and 89%. Nine patients had HL at baseline and the same 9 patients had HL at follow up. Twenty-nine patients had an endocrine deficiency at the start of treatment and 13 developed a deficiency after treatment. Development of an endocrine deficiency at follow-up was associated with increased dose to the combined pituitary and hypothalamus (p <0.0001). The change in mean FSIQ was +0.84 points at latest follow-up (p = 0.33). Baseline average total core QOL score was 75 and 73 (p = 0.24) at latest follow-up. Doses received by normal brain structures are shown in the table below. There was no incidence of brainstem or other brain necrosis. Correlation of dose with outcome measures of hearing, endocrine, and neurocognitive effects will be presented. Proton radiotherapy for pediatric brain tumors is safe, feasible and likely to contribute to improved long-term outcomes. Disease control is comparable to other published studies. Neurocognition, hearing, and HRQOL were stable from baseline to follow-up, but new endocrine deficits were noted after treatment and correlated with increased dose to the HPA.Abstract 1151; Table 1StructureMedian Mean Dose (GyRBE) [range]Supratentorial brain6.9 [0 - 24.9]Infratentorial brain10.5 [0 - 51.2]Pituitary12.1 [0 - 60.0]Cochlea1.9 [0 - 56.6]Hypothalamus18.9 [0 - 56.1]Optic nerves and chiasm13.5 [0 - 53.4]Hippocampi17.9 [0 - 53.2]Temporal lobes8.5 [0 - 43.1]Brainstem46.7 [28.2 - 52.7] Open table in a new tab

Disease-Free Survival in pediatric medulloblastoma is high, though treatment-related late sequelae are common and adversely impact quality of life. 20-40% of patients still fail despite best treatment, and currently there are few biomarkers to predict outcome. Lymphocyte counts have been shown to predict for outcome after RT in other disease sites, though often correlate with other clinical factors. Here we investigate if WBC counts early during and throughout RT can independently predict recurrence. We assessed a total of 197 medulloblastoma patients treated at our institution, the median age was 7.8 years [2.3-24]. We focused on recurrence and therefore excluded disease-free patients with less than 5 years of follow-up. In the remaining 149 patients, of which 42% were classified as high risk, 36 recurrences were observed (24%) and the median follow-up in disease-free patients was 7.5 years. We analyzed WBC counts and its subcomponents, focusing on 2 metrics. Grade 3+ Lymphopenia during week 2 was chosen as a clinically relevant, early marker that was previously published on in other indications. Average counts were chosen as an exploratory data-driven metric to provide a measure of immune surveillance throughout RT. On univariate analysis, several clinical factors (high risk, p=0.05; M stage≥1, p=0.02; anaplastic histology, p=0.04) predict for the risk of recurrence, as well as the overall leukocyte counts, driven by the lymphocytes: grade 3+ lymphopenia at week 2 (LW2, p=0.02) and to a lesser degree the average lymphocyte counts (LA, p=0.09) predict for progression. Patients experiencing LW2 (44% of total) had 5-year disease-free survival (DFS) of 65% vs. 85% for patients without. Pre-RT WBC were not significant. LW2 stays significant when the analysis excludes concurrent carboplatin patients and those who received pre-RT chemotherapy, which can alter circulating WBCs. In a multi-variate model including confounding factors influencing outcome and possibly influencing circulating lymphocytes (carboplatin, chemotherapy pre-RT, CSI dose), LW2 is the only factor remaining significant (p=0.009). A small subgroup (n=7) that received modified CSI fields to spare the vertebral body demonstrated significantly higher LW2 and LA (Spearman p=0.002). Circulating Lymphocytes during the second week of RT predict recurrence and are an independent prognostic factor in medulloblastoma. Reduced CSI fields may impact depletion of circulating lymphocytes. This early finding may indicate which patients could benefit from intensified treatment or additive therapy.Abstract 1135; Table 1Multivariate Cox regressionHR [95% CI]pGrade 3+ Lymphopenia in week 2 of RT2.6 [1.3-5.4]0.009High Risk – omitted as comprised of M1+ and hist, inclusion yields identical resultsM1+0.36 [0.13-1.01]0.053Anaplastic histology1.8 [0.8-3.8]0.14CSI dose0.97 [0.91-1.03]0.30Concurrent carboplatin0.8 [0.2-2.5]0.68Chemotherapy pre-RT0.5 [0.2-1.5]0.22 Open table in a new tab

We have designed a comprehensive pediatric ototoxicity grading scale that accounts for and grades sensorineural hearing loss (SNHL) in children treated with ototoxic therapy (radiation and/or chemotherapy). This comprehensive (COMP) scale captures all patients with SNHL > 20 dB between 250 to 12000 Hertz (Hz). The SIOP consensus scale was designed for high-frequency pattern SNHL (2000+ Hz) in cisplatin-treated. Because the SIOP scale omits deficits at 250-1000 Hz, assigning a ototoxicity grade in patients treated with cranial radiation can be challenging. It also deficiently distinguishes between grade 3 (moderately severe) and 4 (severe) SNHL. We compare the SIOP and the proposed COMP ototoxicity grades applied to the audiograms of medulloblastoma patients. We designed an audiogram-based ototoxicity grading scale that captures and accurately describes the severity of SNHL defined as hearing thresholds > 20 dB at 250-12000 Hz. Grades should correlate with the following qualitative functional impact: [1] mild SNHL - no intervention; [2] moderate SNHL - hearing intervention recommended (i.e., preferential seating, FM system); [3] moderately-severe SNHL - hearing aid indicated; [4] severe SNHL - hearing aid/cochlear implant required. Patients with audiograms from two prospective medulloblastoma studies (NCT01063114, NCT00105560) comprise the study cohort. SIOP and COMP grades were applied to each ear audiogram. Grade was marked as missing if SNHL could not be captured by the grading scale. One hundred thirty-two patients had 914 ear-audiograms eligible for analysis. The SIOP scale could not grade 23 ear-audiograms (2.5%) with SNHL, whereas the COMP scale had no ears with SNHL with missing grades (0%, p<0.001). Using the COMP grading scale, 13 missing SIOP grades were classified as grade 1, 1 was classified as grade 2, 6 were classified as grade 3, and 3 were classified as grade 4. The table shows the distribution of grades by scale. There were 53 discordant grades between the SIOP and COMP scales: 27 SIOP grade 4’s were downgraded to COMP grade 3’s; and 3 SIOP grades were upgraded 1-2 grades by the COMP scale. The COMP scale improves upon the current SIOP consensus ototoxicity scale, better capturing the spectrum of SNHL possible in children exposed to radiotherapy. We have proposed and applied a more comprehensive grading scale that can accurately assign a grade to any child with SNHL assessed in a standard audiogram. The inefficiency of the SIOP scale (designed to replace previous ototoxicity scales) to capture and grade the full spectrum of SNHL possible in cancer survivors may underrepresent adverse hearing outcomes. We will validate the COMP scale with existing scales, correlate it with functional outcomes, and evaluate inter-rater reliability.Abstract TU_21_3102; Table 1Ototoxicity GradeSIOP-Boston Scale, n (%)Comprehensive Scale, n (%)Missing23 (2.5)0 (0)0385 (42.1)385 (42.1)1230 (25.2)240 (26.3)296 (10.5)99 (10.8)3144 (15.8)178 (19.5)436 (3.9)12 (1.3) Open table in a new tab

Standard treatment for childhood medulloblastoma currently sequences radiation therapy (RT) administration prior to maintenance chemotherapy (CT), based on a disease-free survival (DFS) benefit found in earlier sequencing trials (HIT '91 Kortmann, IJROBP 2000; SIOP II, Bailey 1993). However, these studies incorporated RT interruptions to minimize acute hematologic toxicity, with RT prolonged by 11.5 days following neoadjuvant chemotherapy and 35% suffering RT interruptions. Since then, evidence has shown that prolongation of RT treatment time in medulloblastoma compromises DFS (DelCharco 1998; Paulino 2003). It remains unclear whether delivering CT before compressed RT with minimal interruptions (the current standard) would also compromise DFS. Standard treatment guidelines dictate starting adjuvant radiation within 31 days of surgery. This goal can preclude referral of children with surgical complications or those living geographically distant to a proton center, despite the potential advantage for decreased late effects. We report OS and EFS in children with medulloblastoma, stratified by treatment sequencing. All children ages 3-21 with medulloblastoma treated between 2000 and 2016 were included and stratified by adjuvant treatment sequencing after surgery, either compressed proton radiation first (RT1) followed by maintenance chemotherapy or chemotherapy first (CT1) followed by consolidative RT. Patients in the CT1 arm received either (1) typical standard-dose adjuvant CT (vincristine, CCNU, cisplatin, etc.) of short duration (i.e., 1-3 cycles) who started RT >31 days post-surgery, or (2) induction and high-dose CT (HDCT) resulting in multi-month delays prior to consolidative RT. Patients undergoing salvage RT for recurrence following HDCT alone were excluded. 161 consecutively treated patients were eligible for inclusion (88% co-enrolled on protocol) with 136 (84%) in RT1 arm and 24 (15%) in the CT1 arm with a median follow-up of 4.28 years. There was no difference in 4 year EFS 84% vs. 81% (p=0.84) or OS 94% vs. 87% (p=0.91) between cohorts. Notably, fewer patients (14% vs. 31%) in the RT1 vs. CT1 arm had anaplastic (vs. non-anaplastic) histology (p=0.12); and M+ (vs. M0) disease (18% vs. 64%, p=0.0007). RT interruptions were rare, with median treatment time being 41.6 days and 41.2 days, respectively (p=NS). In childhood medulloblastoma, the sequencing of chemotherapy or radiation therapy after surgery did not appear to alter disease outcome in the cohort where overall radiation treatment time was not prolonged despite worse prognostic factors in the CT1 arm. Interval chemotherapy may afford additional time for referring institutions to navigate post-operative complications and the logistical barriers of referral to radiation centers.

Long-Term survival rates for pediatric medulloblastoma patients are encouraging, though treatment-related late sequelae are common and adversely impact quality of life. However, due to non-negligible rates of recurrence, ∼20-30% at 5 years, a general de-escalation of treatment is not warranted. Currently there are few markers to predict outcome in this patient group, impeding efforts to personalize therapy further. We hypothesize that hematologic toxicities during therapy can significantly improve predictions and help identify patients at very high/low risk of recurrence. We analyzed data for 59 medulloblastoma patients aged 3-21 years treated with combined modality therapy including proton therapy in a clinical trial at our institution. Median CSI dose and total dose were 23.4Gy and 54Gy respectively. The following standard variables were extracted for each patient and included in the model building process: age, risk stratification (standard or high/intermediate), histology, radiation dose and cumulative cisplatin dose. In addition, the following acute toxicities, scored according to CTCv3.0, were added: Lymphopenia, Anaemia, Neutrophil and Leukocyte counts. Based on these features we employed 4 commonly used classification techniques (k nearest neighbours, support vector machines (SVM), decision trees and random forests) to predict recurrence within 5 years after treatment (binary classification). 5-fold cross validation was used to protect against overfitting and area under the receiving operator characteristic curve (AUC) and specificity were used to score the models. 11 out of 59 patients had relapsed within 5 years post treatment. None of the clinical predictors was able to significantly separate the PFS curves (all log-rank p>0.2). Taking all 5 variables together did not result in any useful predictions, with AUC for all models between 0.48-0.6. However, analyzing the acute hematological toxicities yielded numerous significant correlations, especially high grade Neutrophils (log rank p<0.001) and Leukocytes (log rank p=0.003) clearly separated the PFS curves. Including these variables into the predictive models boosted their performance to an AUC of ∼0.73. The best models were SVM and tree-based approaches, with the latter showing the highest specificity: they can classify a significant part of the patients recurring within 5 years with a 95% confidence, avoiding inclusion of patients with good prognosis into the high-risk group. This study suggests hematological toxicities can predict outcomes in pediatric medulloblastoma. Several acute toxicities correlate with PFS, and their inclusion may allow for timely modification of treatment regimen before the end of therapy. Similar models could be used to guide clinical trials that aim at personalizing treatment regimen, improving outcomes for high-risk patients while maintaining a high quality of life for long term survivors.

Fazit Nach diesen Daten scheint das Nebenwirkungsspektrum der Protonenbestrahlung bei Medulloblastom mittelfristig vertretbar, so die Autoren. Die damit erreichbare Kontrolle der Erkrankung ist vergleichbar mit der in der Literatur bei herkömmlicher Bestrahlung beschriebenen.

A relationship between RT dose and risk of endocrine dysfunction has been described for children treated with photon RT, however, this has not been examined thoroughly in children treated with proton RT, with some expressing concerns about a higher relative biological effect of proton RT on normal brain tissue. We examined the effect of RT dose to the hypothalamus on development of endocrine dysfunction in children with brain tumors treated with proton RT, and evaluated other factors that affect the risk of endocrine dysfunction. Dosimetric and clinical data was collected from pediatric patients with brain tumors treated with proton RT on 3 prospective single institution clinical trials (2010-2014) and a retrospective registry from 2002-2014 with a minimum of 2 years of follow up. Patients had a pre-treatment endocrine evaluation and were followed yearly by a pediatric endocrinologist, and monitored for clinical and serological evidence of thyroid, growth hormone (GH), cortisol, and sex hormone dysfunction. Hypothalamus and pituitary were contoured on all planning CT scans and dose statistics, including D90, D10, median, and mean dose were collected at the time of treatment planning. Incidence of endocrine deficiency over time was estimated using the Kaplan-Meier method, and multivariate analysis was conducted using a binary logistic regression model. One hundred eighteen patients were evaluable (60% male; 40% female) with a median follow up of 3.4 yrs (range, 2-11 yrs). Medulloblastoma (73%, n = 86) was the most common diagnosis, with all receiving CSI and boost. All other patients received partial brain RT (n = 32), with the most common diagnoses of ependymoma (9.3%, n = 11), glioma (7.6%, n = 9), and craniopharyngioma (3.4%, n = 4). GH deficiency was the most common endocrinopathy (48%), followed by cortisol (21%), thyroid (20%), and sex hormone deficiency (16%). For patients receiving partial brain RT, median hypothalamic dose ≥50 GyE was associated with a rate of 64% for GH deficiency and 45% for thyroid deficiency. Patients with a median hypothalamic dose <50 GyE did not develop GH or thyroid deficiency. Patients with a diagnosis of medulloblastoma who received CSI and concurrent chemotherapy developed higher rates of endocrine deficiency with a hypothalamic dose of 35-40 GyE associated with a higher risk. D90, D10, mean hypothalamic dose, and median pituitary dose correlated with median hypothalamic dose and were not independent predictors of outcome. Hypothalamic tolerance to proton RT is comparable to previously reported values for photons. In addition to RT dose, a diagnosis of medulloblastoma with CSI and concurrent chemotherapy were associated with an increase in risk for endocrine dysfunction in children treated with proton RT for brain tumors.

Purpose/Objective(s)Mesenchymal chondrosarcomas are a rare subtype of chondrosarcomas, distinctive for affecting patients of relatively younger age and for its aggressive behavior. The role of radiotherapy has yet to be described in meaningful detail. A single institutional retrospective analysis was undertaken with a focus on radiotherapy technique.Materials/MethodsThis retrospective study includes a search of patients with a histologically-confirmed mesenchymal chondrosarcoma treated at a single institution from 1970-2010.ResultsTwenty-three patients had confirmed pathological diagnoses of mesenchymal chondrosarcomas. Median follow-up was 47.6 months (range: 9.3-263.7). There were 12 males and 11 females. The median age at diagnosis was 25.4 years old (range, 4-71). Primary sites included ten within the head and neck, five of the trunk, and eight extremities. Systemic therapy was used in 15 patients. Definitive surgery was used in three patients. Radiotherapy was used in 21 patients (4 preoperatively only, 13 postoperatively only, 3 pre-and postoperatively and 1 pre-, intra- and postoperatively). Proton therapy was used in 12 patients. Median preoperative, postoperative and total radiotherapy doses were 45 Gy RBE, 54.9 Gy RBE, and 60 Gy RBE (range, 44-78 Gy RBE), respectively. There were 4 local, 1 regional, and 8 distant failures. Three local failures occurred after surgery alone in 4 patients at a median of 2.37 years (range, 1.4-3.5). In these 3 cases, one was salvaged successfully after amputation, but one failed at 8.5 years after salvage surgery and adjuvant 46.8 Gy and another failed at 5.5 years after definitive 78 Gy RBE for unresectable disease. Of the remaining 19 patients who received primary treatment with a combination of surgery and radiotherapy, only 1 local failure occurred at 17.3 months after 50.4 Gy RBE for microscopic residual disease in the left skull base. Of 12 patients who received systemic therapy as part of management of primary disease, 2 patients had local failures followed by distant metastases. Distant failures occurred at a median of 3 years in 8 patients with a median of 5.1 years of follow-up. Five of these 8 patients had previous or synchronous locoregional recurrences, while 3 failed distantly without locoregional failures. Two year disease-free and overall survival rates were 75.0% and 100%, respectively. Estimated 5 year disease-free and overall survival rates were 60.0% and 95.7%, respectively.ConclusionsMesenchymal chondrosarcomas can be effectively locally controlled with a combination of surgery and radiotherapy for primary disease, with only 1 local failure seen using this approach. Local recurrences treated with surgery and radiotherapy were not durably controlled. Purpose/Objective(s)Mesenchymal chondrosarcomas are a rare subtype of chondrosarcomas, distinctive for affecting patients of relatively younger age and for its aggressive behavior. The role of radiotherapy has yet to be described in meaningful detail. A single institutional retrospective analysis was undertaken with a focus on radiotherapy technique. Mesenchymal chondrosarcomas are a rare subtype of chondrosarcomas, distinctive for affecting patients of relatively younger age and for its aggressive behavior. The role of radiotherapy has yet to be described in meaningful detail. A single institutional retrospective analysis was undertaken with a focus on radiotherapy technique. Materials/MethodsThis retrospective study includes a search of patients with a histologically-confirmed mesenchymal chondrosarcoma treated at a single institution from 1970-2010. This retrospective study includes a search of patients with a histologically-confirmed mesenchymal chondrosarcoma treated at a single institution from 1970-2010. ResultsTwenty-three patients had confirmed pathological diagnoses of mesenchymal chondrosarcomas. Median follow-up was 47.6 months (range: 9.3-263.7). There were 12 males and 11 females. The median age at diagnosis was 25.4 years old (range, 4-71). Primary sites included ten within the head and neck, five of the trunk, and eight extremities. Systemic therapy was used in 15 patients. Definitive surgery was used in three patients. Radiotherapy was used in 21 patients (4 preoperatively only, 13 postoperatively only, 3 pre-and postoperatively and 1 pre-, intra- and postoperatively). Proton therapy was used in 12 patients. Median preoperative, postoperative and total radiotherapy doses were 45 Gy RBE, 54.9 Gy RBE, and 60 Gy RBE (range, 44-78 Gy RBE), respectively. There were 4 local, 1 regional, and 8 distant failures. Three local failures occurred after surgery alone in 4 patients at a median of 2.37 years (range, 1.4-3.5). In these 3 cases, one was salvaged successfully after amputation, but one failed at 8.5 years after salvage surgery and adjuvant 46.8 Gy and another failed at 5.5 years after definitive 78 Gy RBE for unresectable disease. Of the remaining 19 patients who received primary treatment with a combination of surgery and radiotherapy, only 1 local failure occurred at 17.3 months after 50.4 Gy RBE for microscopic residual disease in the left skull base. Of 12 patients who received systemic therapy as part of management of primary disease, 2 patients had local failures followed by distant metastases. Distant failures occurred at a median of 3 years in 8 patients with a median of 5.1 years of follow-up. Five of these 8 patients had previous or synchronous locoregional recurrences, while 3 failed distantly without locoregional failures. Two year disease-free and overall survival rates were 75.0% and 100%, respectively. Estimated 5 year disease-free and overall survival rates were 60.0% and 95.7%, respectively. Twenty-three patients had confirmed pathological diagnoses of mesenchymal chondrosarcomas. Median follow-up was 47.6 months (range: 9.3-263.7). There were 12 males and 11 females. The median age at diagnosis was 25.4 years old (range, 4-71). Primary sites included ten within the head and neck, five of the trunk, and eight extremities. Systemic therapy was used in 15 patients. Definitive surgery was used in three patients. Radiotherapy was used in 21 patients (4 preoperatively only, 13 postoperatively only, 3 pre-and postoperatively and 1 pre-, intra- and postoperatively). Proton therapy was used in 12 patients. Median preoperative, postoperative and total radiotherapy doses were 45 Gy RBE, 54.9 Gy RBE, and 60 Gy RBE (range, 44-78 Gy RBE), respectively. There were 4 local, 1 regional, and 8 distant failures. Three local failures occurred after surgery alone in 4 patients at a median of 2.37 years (range, 1.4-3.5). In these 3 cases, one was salvaged successfully after amputation, but one failed at 8.5 years after salvage surgery and adjuvant 46.8 Gy and another failed at 5.5 years after definitive 78 Gy RBE for unresectable disease. Of the remaining 19 patients who received primary treatment with a combination of surgery and radiotherapy, only 1 local failure occurred at 17.3 months after 50.4 Gy RBE for microscopic residual disease in the left skull base. Of 12 patients who received systemic therapy as part of management of primary disease, 2 patients had local failures followed by distant metastases. Distant failures occurred at a median of 3 years in 8 patients with a median of 5.1 years of follow-up. Five of these 8 patients had previous or synchronous locoregional recurrences, while 3 failed distantly without locoregional failures. Two year disease-free and overall survival rates were 75.0% and 100%, respectively. Estimated 5 year disease-free and overall survival rates were 60.0% and 95.7%, respectively. ConclusionsMesenchymal chondrosarcomas can be effectively locally controlled with a combination of surgery and radiotherapy for primary disease, with only 1 local failure seen using this approach. Local recurrences treated with surgery and radiotherapy were not durably controlled. Mesenchymal chondrosarcomas can be effectively locally controlled with a combination of surgery and radiotherapy for primary disease, with only 1 local failure seen using this approach. Local recurrences treated with surgery and radiotherapy were not durably controlled.

A 14 year-old girl was referred to our institution with stage IVA Hodgkin's disease that involved the sacrum. The patient initially presented with sacral pain. She had no associated B-symptoms such as fever, weight loss, or night sweats. A CT scan of the pelvis showed a destructive lytic lesion in the right sacrum, and small lytic lesions in the left sacrum (figure 1A). A CT-guided needle biopsy of the right sacral lesion showed classic Hodgkin's disease of mixed cellularity. Although CT of the chest and abdomen did not show enlarged lymph nodes, an 18-fluorodeoxyglucose ([18F]FDG) PET scan showed increased uptake of radiolabel in the sacrum, and in the superior mediastinum and left axilla.

Schultz, Karl MD; Rosenberg, Andrew E. MD; Ebb, David H. MD; Mankin, Henry J. MD Author Information