DISCUSSION
Across an 11-year period at our institution, fourteen patients (5 with single kidney) were treated with an intensive ifosfamide-containing VDC-ICE treatment plan. This regimen was well tolerated, with no patient experiencing toxicities severe enough to prevent completion of planned therapy, and no significant renal toxicity occurred in any patient. Although only 9 out of 14 patients completed all 8 intended cycles of chemotherapy, early termination of VDC-ICE chemotherapy was only due to disease progression, not lack of tolerability. There were no toxic deaths or severe organ toxicities. While examining outcomes was not our primary objective, our results add to existing evidence of an association of improved outcomes with the use of intensified treatment.
The COG clinical trial, AREN0321, found that chemotherapy regimens (UH-1 and UH-2) containing nearly all of the same agents as VDC-ICE, with the notable omission of ifosfamide, improved relapse-free survival for stages II-IV DAWT. The UH-1 and UH-2 regimens were accompanied by significant treatment-related CTCAE grade 4 toxicities and mortality – with four patients with DAWT (6%) dying of toxicities – prompting a reduction in regimen intensity and eventual early termination of the trial. As VDC-ICE is similar or even more intense than UH-1 (Supplemental Table S3), we examined the toxicities experienced by our patients receiving VDC-ICE. Importantly, we did not observe any non-hematologic grade 4 or 5 toxicities. As the most significant complications of AREN0321 were related to cardiac and pulmonary dysfunction, we hypothesize that the universal use of the cardioprotectant dexrazoxane in our study, which was allowed but not recommended on AREN0321, may have contributed to the lack of grade 4 and 5 toxicities. However, future studies are needed to confirm this hypothesis. Additional supportive care, including administration of prophylactic infectious measures may have helped decrease infectious complications, but these were not administered in a standardized way. Kidney function may have been protected by extended home hydration implementation.
The potential nephrotoxicity of ifosfamide-containing regimens like VDC-ICE is a major concern in patients with HRR tumors, especially those with single kidneys. In our cohort, kidney injury was rare, with just a single episode of temporary grade 1 acute kidney injury due to an episode of dehydration associated with carboplatin-induced delayed nausea and vomiting, which resolved in 24 hours. Six of our patients had primary renal tumors, all of whom underwent bilateral partial or unilateral nephrectomies prior to the majority of their chemotherapy. All six completed the intended 8 cycles of VDC-ICE, making the lack of severe toxicity or renal insufficiency noteworthy in light of the inclusion of the potentially nephrotoxic ifosfamide in VDC-ICE as compared to the AREN0321 regimens. Possible explanations include improvements in antiemetic management with more widespread use of aprepitant and the ability to provide home hydration via nasogastric tube or intravenous fluids. We acknowledge that our institution’s catchment area has excellent home nursing resources, enabling us to offer these options to our patients.
Despite meticulous supportive care, the VDC-ICE regimen is not without toxicities. Not surprisingly, given the myelosuppression expected with the agents and doses administered, all but one patient (who only received 3 cycles of VDC-ICE due to disease progression) experienced at least one CTCAE grade 3 toxicity, each having at least one hospitalization for febrile neutropenia. Half of our patients had grade 3 nausea/vomiting/dehydration, and more than half had anorexia requiring nutritional support through a nasogastric tube. Other significant toxicities were rare and limited to one episode of uncomplicated bacteremia, hepatotoxicity in patients with tumors of their livers, and admissions to the ICU. While 14% (2/14) of patients requiring ICU admission seems substantial, none were thought to be related to VDC-ICE chemotherapy.
Although all patients treated with this regimen required significant supportive care (e.g., unplanned ED/clinic visits, transfusion, and nutritional support), it appears that overall, the complications and toxicities observed in our cohort were manageable. Even though treatment delays of greater than a week were observed in 50% of our patients, these were mostly due to delayed count recovery. Furthermore, this is not entirely unexpected given the intensity of this regimen and does not justify considering this regimen as intolerable. Other regimens, such as standard of care for Ewing sarcoma (interval compressed alternating cycles of VDC and ifosfamide/etoposide), are also plagued by the need for significant symptomatic and supportive care, treatment delays, and admissions for febrile neutropenia. Yet, this regimen has been used for years as the standard of care for this disease, lasts longer than VDC-ICE (28 versus 24 weeks) and has higher cumulative doses of doxorubicin (375 mg/m2), cyclophosphamide (8400 mg/m2), ifosfamide (63 g/m2), and etoposide (3500 mg/m2).
Our findings are limited by the small number of patients and their management at our single institution. It is possible that the lack of toxic deaths we have seen is due to chance alone. For example, given toxic deaths on AREN0321 were rare, if the toxic death rate of VDC-ICE were comparable to UH-1 and UH-2 regimens, we would have expected to see just 1 toxic death in our 14 patients. Even discounting toxic death rates, it is notable that our patients experienced no grade 4 toxicities. Because these tumors are rare, the challenge of low patient numbers is difficult to overcome even through multicenter trials. Nevertheless, we represent a high-volume center, and this is the largest cohort of patients with HRR/INI- tumors described with this chemotherapy regimen. The other similarly-sized cohort of pediatric oncology patients treated with VDC-ICE comprised patients with Ewing sarcoma who were much older, lacked renal tumors, and thus would have been expected to tolerate the regimen better than our cohort.9
Additionally, because this represents a retrospective analysis, controlling for differences in patient characteristics, treatment regimen administration, or supportive care is difficult. Each patient’s treatment plan was developed by our center’s disease-specific expert in renal tumors and MRT, with some patients receiving customized dosing or the addition of other chemotherapy agents based on individual characteristics. Therefore, whether our experience is generalizable remains an important question.
Our findings suggest that VDC-ICE chemotherapy can be safely used for children with HRR and INI-deficient tumors, with comparable toxicity to other widely accepted pediatric oncology regimens. Most notably, use of ifosfamide was tolerated without nephrotoxicity in a population generally thought of as being at unacceptably high-risk of renal toxicity. Despite the known activity of ifosfamide in HRR/INI- tumors, the use of post-nephrectomy ifosfamide has been reserved for the relapsed setting. Yet for patients with these aggressive malignancies, for whom death is overwhelmingly due to progressive and relapsed disease rather than treatment toxicity, withholding effective agents is difficult to justify. With our findings of excellent tolerance of VDC-ICE combined with no meaningful chance of post-relapse survival for patients with these tumors, inclusion of ifosfamide in upfront treatment regimens is imperative. The use of this regimen has been adopted as our standard of care for these rare malignancies and should be considered as a reasonable option until future studies identify more promising treatments. Furthermore, we believe this regimen warrants further evaluation at a greater scale through a prospective, multicenter clinical trial.