Background
The Ewing’s family of tumours (EFT) are the second most common malignant bone tumour seen in children and young people[
1,
2]. Histologically, they are characterized by small round blue cells with immunohistochemical staining for CD99 and neural markers. A reciprocal translocation between chromosomes 11 and 22 is evident in more than 85% of these tumours[
3,
4]. The family of small round blue cell sarcomas also includes desmoplastic small round cell tumour (DSRCT), a rare soft tissue sarcoma characteristically presenting in young males with extensive multifocal intraabdominal disease. Similar chemotherapy approaches to those utilized for EFT are used, albeit with less satisfactory results as progression and ultimately death due to disease is almost universal[
5‐
7].
Since the introduction of multimodality treatment in EFT, survival has improved from 10% to 75 % in patients with localized disease[
8‐
11]. Since the 1980’s, chemotherapy regimens have evolved both in Europe and the United States to include anthracyclines and alkylating agents with only modest variations in dose and schedule[
9,
10,
12‐
15]. To contend with a lack of recent survival improvement or new agents with major activity, investigators have concentrated on investigating the benefits of scheduling and dose intensity.
The current European Ewing tumour Working Initiative of National Groups 1999 (EURO-EWING 99) study has enrolled over 3,200 patients in a study evaluating an intensive induction regimen (VIDE, vincristine, ifosfamide, doxorubicin and etoposide) and, in selected cohorts of randomized patients, high dose chemotherapy with stem cell rescue. Toxicity associated with VIDE chemotherapy is substantial. For example, neutropenia and related fever is reported in 60.8% and 65.8% of courses respectively[
16].
In North America, alternative approaches to dose optimization have been explored. The Children’s Oncology Group (COG) study, INT-0154, using a regimen of vincristine, doxorubicin, cyclophosphamide alternating with ifosfamide and etoposide (VDC/IE), randomized patients to an increased treatment intensity by higher doses of cyclophosphamide and ifosfamide and a decreased length of treatment to 11 cycles over 30 weeks in the test arm compared with a standard 17 cycles over 48 weeks but with equivalent total drug doses in each arm. There was no survival improvement but more toxicity in the dose intense arm[
17]. In contrast a survival advantage has been reported in the preliminary results from a further COG study of patients with localized EFT, AEWSOO31, randomizing between a standard three weekly schedule and an ‘interval compressed’ two weekly schedule of VDC/IE, the latter made possible by growth factor support[
18]. The mean cycle durations were 18.5 and 23.3 days for the two and three weekly cycles respectively. Event free survival at 3 years was significantly extended in the two weekly arm, 76% vs. 65%, p = 0.028. Toxicity was similar in the two treatment arms but with the reported frequency of febrile neutropenia and other major toxicities apparently lower than those seen with VIDE. Interval compressed VDC/IE has consequently been adopted as the standard of care for future studies of EFT by COG.
There are significant advantages to defining a standard chemotherapy regimen for EFT, not least as a platform for testing new agents in an international setting, which is essential in studying rare cancers. Additional goals for all investigators are reducing both short and late toxicity in a young population of whom approximately two thirds will achieve long term survival, and of course limiting the treatment burden in those with poor prognostic factors. Whether interval compressed VDC/IE is less toxic than VIDE is speculative in the absence of comparative data. Furthermore, the AEWS0031 study was limited to patients with localised disease and had only a small proportion of patients aged over 20 years, so it may not be appropriate to extrapolate the data to the entire EFT population.
In preparation for a planned randomized comparison in Europe of first line treatment for EFT between VIDE and VDC/IE, we retrospectively examined the feasibility of interval compressed VDC/IE in patients with metastatic EFT. In recognition that less toxic treatment would be appropriate for older patients[
19] and those with a poor prognosis, we did not restrict the age of our cohort, and also included those with DSRCT. This latter group responds temporarily to chemotherapy regimens used in EFT but with a 5-year survival of only 15%[
20]. Our institutional standard for DSRCT and UHGRCS includes treatment with VIDE. Identifying a more tolerable but effective regimen is a priority.
Materials and methods
Patient selection
A consecutive series of patients were treated with interval compressed VDC/IE if either ineligible for the EURO-EWING 99 study (including those with extrapulmonary metastases), or were newly diagnosed with DSRCT or UHGRCS. A histological diagnosis was required in all patients. Staging included plain X-rays and magnetic resonance imaging (MRI) or computed tomography (CT) scan of the primary site, chest CT scan and whole body technetium (99m Tc) bone scan were performed to confirm the presence of metastases. Routine assessments of renal and cardiac function were conducted at baseline, during and after therapy. Institutional ethics guidance was followed.
Chemotherapy
The VDC/IE regimen was alternate 14 day cycles of VDC (vincristine 1.4mg/m2, maximum 2mg, doxorubicin 75mg/m2 and cyclophosphamide 1200mg/m2 administered over 2 days) and IE (ifosfamide 9gm/m2, etoposide 500mg/m2 fractionated over 5 days). Patients had a full blood count measured on the day prior to the cycle due date and treatment was given if the neutrophil count was greater than 1 × 109/l, platelet count greater than 80 × 109/l and biochemical parameters were within normal range. Nadir blood counts were not routinely measured. All patients received growth factor support with pegylated granulocyte colony stimulating factor given within 72 h of last chemotherapy. Toxicity was assessed using the Common Terminology Criteria for Adverse Events (CTCAE) Version 3.0. In the event of lack of recovery from toxicity, chemotherapy was delayed or modified at the discretion of the treating physician. Radiological response assessment was carried out with restaging CT and MRI scans after 4 and 8 cycles of treatment. Fourteen cycles of chemotherapy were planned subject to tolerability and response, with treatment interrupted after 6 cycles or 12 weeks for surgery if applicable. When radiotherapy was the primary modality for local control, this was given concurrently with chemotherapy beginning after cycle 6 and with the omission of anthracyclines from relevant cycles.
Discussion
In response to emerging data from the US indicating that interval compressed VDC/IE was a well-tolerated, effective regimen for EFT, we selected patients who were ineligible for randomisation in Euro EWING 99 to receive this treatment in preference to the institutional standard, VIDE[
16,
21]. We wished to gain preliminary experience of the feasibility of this regimen and extend its use to a population with metastatic disease and older patients, groups not represented in the randomised study from COG. The results indicate that this regimen can be safely delivered and interval compression achieved in this patient group. Although this is a small study, these data also lend support to the view that this is a tolerable regimen, even in an older patient group with considerable disease burden. Finally, we observed clinical benefit of dose compressed VDC/IE in DSRCT.
Toxicities associated with VIDE are well recorded both from the initial single institution study[
21] and from a large analysis of the first 851 patients included in EURO-E.W.I.N.G. 99[
16]. A total of 4,746 courses of VIDE in 851 patients were analysed with respect to toxicity. The rate of febrile neutropenia was 60.8% with, and 65.8% without, GCSF support.
Toxicity data from the AEWSOO31 study is currently available only in abstract but reports a rate of grade 3/4 febrile neutropenia of 6.9% in the two weekly arm[
18]. While emphasising the very small number of patients reported here, this figure is lower than that reported here (13.5%). However, the several differences in patient demographics may account for this. Firstly the AEWS0031 study was in a paediatric population with only 75 of the 587 (12.8%) patients over the age of 18, whereas in this study 12 patients were over the age of 18 (75%). Secondly, the AEWS0031 study was carried out in patients who did not have metastatic disease. Again, in our cohort, the majority of patients had metastatic disease and a large overall tumour burden, 11 (69%) patients having bone, bone marrow and lung metastases, and 7 (44%) with primary pelvic disease.
The determinants of chemotherapy tolerance in EFT require further clarification. Comparisons of the influence of factors such as age are limited by the absence of planned prospective analyses and the reporting of cohorts containing varying proportions of children, adolescents and adults. Hence studies that report the adverse influence of younger age may contain few adults[
22]; others focus exclusively on much older adults[
23]; while still others fall between these extremes[
19,
24]. In the largest study, from EURO-E.W.I.N.G. 99, chemotherapy toxicity was not clearly greater in older patients but dose modifications were more frequent[
16]. Most clinicians will remain cautious when treating older patients or those with heavy tumour burden with the current intensive regimens discussed here.
The treatment interval for the two weekly regimen was successfully maintained in this patient cohort with a mean treatment interval of 16.7 days. This resulted in a shorter overall treatment time when compared to a 3-weekly regimen. For the 8 patients who completed all 14 cycles of treatment the average overall treatment time was 253 days, which is markedly shorter than the 294 days it takes to complete standard treatment with VIDE/VAI. In the EURO-E.W.I.N.G 99 clinical trial, dose modifications were recorded in 1020 of the 4746 courses (21%) of all cycles. In this study, only 26 cycles were given at a reduced dose (14%), therefore we can postulate that dose intensity was not compromised in order to maintain a shorter dose interval. Three patients terminated treatment early due to chemotherapy-related toxicity; two at cycle 11 and one at cycle 12.
The main limitation of this study is that it is a small non-randomised cohort of patients with heterogeneity of presenting clinical features. The poor prognostic factors present in several patients and the need to gain familiarity with the regimen will have had an effect on the application of clinical thresholds for decisions on dose reductions and cessation of treatment. The information on clinical outcome is provoking but should be considered only with caution. Despite this, the data do have value in supporting the premise that this may be a less toxic regimen than VIDE, which is deliverable in patients outside of the limited eligibility criteria for AEWS0031.
In conclusion, this study demonstrates that interval compression of chemotherapy is feasible in an older and higher risk cohort of patients with EFT and other small round cell sarcomas. This schedule appears well tolerated compared with the standard European treatment of VIDE followed by VAC or VAI. We also included patients who would traditionally have a poor long term survival even with intensive chemotherapy. A shorter, less toxic regimen is an attractive option for these patients in whom life expectancy is likely to be limited by their disease. Further data regarding efficacy and long term toxicity will be available in the future as this regimen will now be compared with the VIDE/VAI schedule in a multi-centre randomised controlled trial.
Competing interest
The authors declare that they have no competing interest.
Authors’ contributions
JW conceived the study; analysed and interpreted data; drafted the manuscript. AK, AS and CR collected, analysed and interpreted the data and contributed to drafting the manuscript. BS, MM, SS contributed to the study design, collection and interpretation of data and contributed to the manuscript. All authors read and approved the final manuscript.