Challenges and treatment outcome of patients with Kaposiform hemangioendothelioma: a retrospective single center experience

Previously, KHE was shown to have an equal sex predilection. However, a slight male predominance has been indicated by two large retrospective studies, both of which collected data from more than 100 patients with KHE [1], in our cohort there was obvious male predominance compared to other studies.

Interestingly, all our patients had thrombocytopenia, nine of them were classified as very severe, as mentioned KMP may occur in 42 to 71% of patients with KHE [2], but the higher percentage in our cohort may be related to the fact of being a tertiary referral center.

Previously, the decision for treatment of patients with KHE was based on small cases series and retrospective data using multiagent chemotherapy or interferon alpha [10] then after a first report of successful treatment of one case report with vincristine and corticosteroids in 2009 [11], a consensus-derived practice treatment for complicated KHE has been published in 2013 recommending first-line therapy with intravenous vincristine and prednisolone [12]. After successful results for sirolimus in patients with KHE [13, 14], a randomized controlled trial of sirolimus monotherapy compared to combined treatment of sirolimus and prednisolone showed a durable platelet response in 94.5% in the combination group compared to 66.7% in the monotherapy shifting the paradigm of initial treatment for patients with KHE [6].

Surprisingly, one of the patients who did not show response to therapy had a spontaneous hematological recovery of platelet after 12 months of stoppage of treatment, likely in a study by Boccara and colleagues 2015, suggested that KMP may resolve spontaneously with suitable supportive care alone as no adequate response in his group of studied patients [13].

There were no clear definitions in the previous recommendations for hematologic response or treatment duration; furthermore, the risk factors for refractoriness to treatment and when to declare failure of treatment are not clearly stated [15].We have put definition for hematological response as change from one grade of severity to another especially in platelet count till reaching normal platelet count more than 100 × 10 * 3/µL. We tried to identify the time period required to reach each stage and to identify the risk factor for refractoriness. Hematological response in our cases to first line (steroid and vincristine) appears after 3–7 days of therapy by changing from grade of severity to another then normalization of platelet count occurred after 2 to 4 weeks of therapy. The intended length of vincristine therapy is approximately 20–24 weeks tailored according to high-risk symptoms, tumor response and acceptable toxicity [16].

Our center is using sirolimus as second line treatment for non-responding patients to first line based on studies reporting the importance and response of sirolimus in patients with KHE [17, 18]. Second line treatment varies between centers, in a survey done in 27 centers in USA and Canada, vincristine was used in 38%, rapamycin in 21%, and propranolol in 21% of treating centers [19]. We had problem for using sirolimus as first line due to shortage of solution form in Egypt being a low to middle income country; this approach points to a need to modification of standard of care used in certain diseases based on drug availability. Combination therapy of sirolimus and steroids was not used in our cohort because the results of randomized controlled trial of combined sirolimus and steroids had not been published until 2022 and we are reporting patients before this era [20], that is why it was not used in patients who failed sirolimus as monotherapy.

We found that patients with delay in starting treatment, those who sought medical advice after 1 year of age, large size of cutaneous tumor (> 5% of body surface area), mediastinal or visceral KHE, life threatening presentation especially of systemic nature, very severe thrombocytopenia at presentation, no hematological response after maximum 2 months of therapy with vincristine and steroids are those who were considered poor responders or refractory to treat.

Likley, Wang and his colleagues reported that the deeper lesions in patients with KHE are associated with significant morbidity and mortality rates and poor response [21], similarly, Keane and colleagues had a patient with,mediastinal KHE/KMP and responded only after using steroids with sirolimus [22]. Also, KMP occurs in 87% of the masses > 5 cm and 100% of those > 10 cm, suggesting that dimensions are related to the risk of developing KMP and so carry risk of poor response and high relapse rate after stopping treatment [23].

Long term outcome reported in our patients was good with limited functional disability such as distal lymphedema at the site of the lesion in responders, hemiplegia secondary to intracranial bleed improved by physiotherapy. The residual lesions previously reported in patients with KMP were either a cutaneous red stain, with or without associated red papules overlapping a fibrotic infiltration, or telangiectatic streaks and swelling a minor, firm, irregular, subcutaneous mass, or sequelae in muscles and/or joints [24]. Lymphatic involvement is frequently reported in patients with KHE, where D2 - 40 immune stain was positive in 96% of KHE. It stained the neoplastic spindled cells and lymphatic channels adjacent to vascular lobules of KHE suggesting, lymphothelial differentiation of its neoplastic component [25]. Similarly, Ji and his colleagues conducted a multicenter retrospective analysis of patients who had a minimum of 3 years of follow-up after the onset of KHE and/or Kasabach–Merritt phenomenon (KMP). They concluded that chronic lymphedema is a common sequela of KHE and can occur independently of KMP and sirolimus treatment [26].

Study limitations

The findings of this study have to be seen in light of some limitations. It included experience of one center with a small sample size as an example of low-middle income countries. Further collaborative prospective study needs to be conducted with other centers with similar settings in order to enroll more patients and therefore elucidate the patterns of response and the toxicities of the various treatment modalities.

Comments (0)

No login
gif