Discussion
KHE and TA are aggressive vascular tumors that are associated with significant morbidity and are rarely cured. Until the discovery of efficacy of mTOR inhibition with sirolimus in KHE in mid-2000s, vincristine and steroids were the mainstay of medical therapy. The rare nature and widely heterogenous clinical presentation of KHE/TA have made comparison of treatment modalities difficult. This study compares treatment response and durability in the largest cohort of children with KHE/TA to date.
In this cohort, we found similar rates of KMP (40.3%) but a much lower mortality rate (1.3%) compared to historical cohorts. This may be attributed to the introduction of sirolimus for long-term tumor control.4,12,32,33 Sirolimus was the most common treatment and overall treatment response rates were high (>70% overall by 6 months), and there were no differences in radiologic, hematologic, or clinical response between the VCR and Siro treatment groups at 6 months. Due to chronicity of KHE/TA we also felt it was important to try to evaluate durability of treatment response. We found no significant differences in the rates of progressive or recurrent disease at 3 and 6 months between the VCR and Siro treatment groups. While increased rates of reported persistent disease were noted in the Siro group compared to VCR, we think the lack of a standardized definition of persistent disease across investigators and the frequent addition of sirolimus to patients on VCR regimen precludes confirming this as a valid finding. KHE/TA can be considered a chronic tumor, requiring months to years of therapy with infrequent complete cures, and persistent disease needs to be defined more effectively in future studies.
A recent prospective trial in 73 patients with KHE with KMP showed that upfront sirolimus therapy in combination with steroids improves time to resolution of KMP, durability of platelet response, and overall tumor response at 12 months.31 In this retrospective cohort, low patient numbers limited comparison of regiments with and without steroids. More patients reported persistent disease in the Siro + steroid group compared to the VCR + steroid treatment group, however again noting that nearly 50% of the patients in the VCR + steroid group added sirolimus to their regimen later.
Different side effect profiles often influence the choice of medical agent. Sirolimus is associated with concerns for delayed wound healing, mucositis, neutropenia, and lymphopenia. Vincristine is associated with constipation, jaw pain, and peripheral neuropathy. Sirolimus may be an agent of choice or an agent of transition for some due to its increased ease of administration, lack of need for central venous access, and demonstrated efficacy. Although there are concerns about immunosuppression secondary to sirolimus, we found a lower rate of infectious complications in the Siro group compared to patients receiving vincristine. However, high adverse event rates in the groups receiving vincristine (VCR and VCR+Siro) could be confounded if these patients had more severe disease leading their physicians to elect treatment with vincristine.
With equivalent treatment response and durability at 6 months, the decision to treat with vincristine or sirolimus, with or without steroids, should account for additional clinical factors, including route of administration, side effects, perceived severity of KMP or urgency to treat, and patient/family preference. While vincristine may bring about more rapid radiologic response and perhaps should be considered for high urgency situations, this is not associated with improved longer-term outcomes in these indolent tumors. Although full surgical resection of KHE is the only curative option, we found that many patients who underwent surgery ultimately ended up needing systemic medical therapy. Fortunately, the high overall treatment response rates and low mortality highlighted in this cohort demonstrate that individual patient factors and family and physician preferences can be encouraged in the management of this rare vascular tumor.
This study represents the largest cohort to date investigating treatment practices and short-term response rates in patients with KHE/TA. Results are limited by the retrospective nature of this study, the clinical heterogeneity of KHE/TA, and variable investigator experience and clinical perspective. Although KHE and TA have overlapping clinical and histopathologic features, their presentation and the urgency for medical therapy can be quite distinct, which may affect assessment of treatment choices and duration in this cohort. Many patients with mild TA are observed only or treated with monotherapy initially, whereas patients who present more acutely with large KHE causing clinical compromise or KMP, multi-modal therapy may be the initial approach. Due to this significant disease heterogeneity, determining baseline response parameters for this population will allow for historical comparison in future treatment protocols. These results augment the ability to make educated decisions for individualized medical management for patients with KHE/TA with and without KMP, particularly when complete surgical excision is not possible or successful. Given the rarity of these tumors, multi-institutional studies are needed to further knowledge of the natural history of KHE/TA and optimal treatment modalities.