Abstract
Background:
Hepatic vascular malignancies (HVMs) are rare malignancies, with no standardized treatment regimens. The most common HVMs, angiosarcoma and malignant epithelioid hemangioendothelioma (EHE), are often grouped together in the literature complicating our ability to achieve reliable survival data and treatment strategies.
Objective:
To compare the disease characteristics of HVMs, with a subanalysis on pediatric patients.
Methods:
The 2016 National Cancer Database was queried for patients with HVMs using international classification of diseases-oncology-3 (ICD-O-3) codes yielding 699 patients. Descriptive statistics, chi-square, Kaplan-Meier, and log-rank analyses were performed.
Results:
We found 478 patients (68%) with angiosarcoma and 221 (32%) with EHE. The median (Q1, Q3) age for angiosarcoma patients was 65 years (56, 75) versus 54 years (37, 65) in EHE patients (P < .001). The rate of resection was lower in patients with angiosarcoma than EHE (13% vs 32%, P < .001). The mean 1-, 3-, and 5-year overall survival for angiosarcoma patients was 17%, 8%, and 6%, respectively, versus 80%, 65%, and 62% in EHE patients (P < .0001). A subgroup analysis was performed on pediatric patients demonstrating six with angiosarcoma and 10 with EHE. The mean 1-, 3-, and 5-year overall survival for pediatric angiosarcoma patients was 67%, 50%, and 50%, respectively, and 90%, 90%, and 90% for pediatric EHE patients.
Conclusion:
In the largest study of HVMs to date, we found angiosarcoma has significantly worse overall survival than EHE. Pediatric patients appear to have improved survival and higher rates of resection. Larger studies of HVMs are needed to clearly differentiate tumor types, standardize care, and improve survivorship.
Keywords: hepatic angiosarcoma, hepatic vascular malignancies, malignant epithelioid hemangioendothelioma, National Cancer Database, pediatric surgical oncology, rare liver tumor
1 |. INTRODUCTION
Hepatic vascular malignancies (HVMs) are extremely rare malignancies in both adult and pediatric populations; as such, treatment paradigms are developed based upon case reports or case series. HVMs are mesenchymal neoplasms that arise from vascular or lymphatic vessels with high proliferative capacity.1 The most common pathologies are hepatic angiosarcoma and malignant epithelioid hemangioendotheliomas (EHE), but these two tumors are difficult to distinguish, often misclassified histologically, and often discussed as one malignancy in the literature.2–4 HVMs can be asymptomatic or present with right upper quadrant or vague abdominal pain, postprandial fullness, fever, weight loss, hepatomegaly, elevated transaminases, anemia, congestive heart failure, liver failure, or rupture leading to hemoperitoneum.5–8 Several chemical exposures have been postulated to increase the incidence of HVMs, including: vinyl chloride, Thorotrast, androgenic steroids, phenylhydrazine, arsenic, diethylstilbestrol, urethane, and cyclophosphamide.9–14
The differential diagnoses for HVMs are often broad, and include many benign and malignant lesions of the liver.1 The radiologic appearances are often described as neoplasms with irregular margins, heterogenous architecture, internal hemorrhage, and progressive contrast uptake on dynamic contrast-enhanced evaluation.15–17 HVMs are highly vascular tumors that are often multifocal, nodular, and contain central necrosis.18 Pretreatment percutaneous tissue diagnosis is complicated by the vascular nature of the tumors and the potential for hemorrhage.6,19 Under histological examination, angiosarcoma has large, atypical spindle and polyhedral-shaped endothelial cells that can grow alongside or invade sinusoidal structures,20 while EHE has cords and nests of neoplastic cells that invade vessel walls, obliterating the lumen and spreading outward into surrounding tissue.2 There are no single immunohistochemical markers for HVMs; however, epidermal growth factor is most often positive followed by cluster of differentiation (CD) 31, CD34, and factor VIII-related antigen.21 Due to the rarity of HVMs, differentiating angiosarcoma from EHE is very difficult, but misclassification can lead to incorrect prognosis and possibly treatment.4
There are no standardized treatment practices for HVMs.1 Surgical resection is the mainstay of therapy due to the aggressive nature of these malignancies, though few patients are eligible for resection at time of diagnosis.18 Due to advanced disease and lack of standardized treatment regimens, previous studies estimated angiosarcoma to have a median survival of 5 months following diagnosis.22 In comparison to angiosarcoma, EHE occurs in younger patients and has a more indolent course with overall survival as high as 73%.23 Additionally, HVMs are often cited as a contraindication for transplantation in the adult literature. However, some reports in the pediatric population advocate for transplantation or other aggressive surgical techniques.24 It is unclear whether the tumor biology and outcomes are different between adult and pediatric patients, as no comparator studies have previously been performed.
Given the rare nature of these tumors—especially in children—scant literature, and lack of treatment guidelines, we sought to compare tumor characteristics, treatment modalities, and outcomes of angiosarcoma versus EHE using the National Cancer Database (NCDB) with a subgroup analysis on pediatric patients. We hypothesized that all patients with angiosarcoma would experience worse overall survival as compared to EHE, adults would have worse survival than pediatric patients in both malignancies, and most patients with HVMs would undergo nonstandardized treatment regimens.
2 |. METHODS
The 2016 NCDB hepatic file was reviewed for international classification of diseases-oncology-3 (ICD-O-3) codes of primary liver malignancies (C22.0), with histology codes 9120 and 9133 to identify all persons with a histologically confirmed diagnosis of angiosarcoma and EHE, respectively. Duke University Medical Center Institutional Review Board approval was sought, and the study was determined eligible for exemption as a retrospective review of a national de-identified dataset.
Patients under 21 years of age were classified as pediatric patients and those 21 years and older were classified as adults. While some literature on HVMs uses the intra-abdominal and intrathoracic soft tissue sarcoma tumor, node, metastasis staging system developed by the American Joint Committee on Cancer in 2018, consistent tumor staging is not available in NCDB. Therefore, we instead report tumor size, lymph node involvement, and presence of metastasis. We compared tumor characteristics at presentation, management, and survival in angiosarcoma versus EHE. A subgroup analysis was performed on pediatric patients; however, given the small number of patients with heterogenous outcomes, we elected to provide descriptive statistics only. Percentages were calculated from known variables; unknowns were excluded for percentage count. Univariable analyses including Pearson’s chi-square test or Student’s t-test were performed, as appropriate. Survival data were visualized with Kaplan-Meier curves, differences in restricted mean survival times are reported, and log-rank tests were performed. Significance was defined as P < .05.
Restricted mean survival was calculated due to small sample sizes. In survival analyses in oncology, restricted mean survival time estimation confers several advantages over hazard ratio and/or median survival. This is because it better captures all information presented in Kaplan-Meier curves and it does not hinge on assumptions of proportional hazards. In brief, differences in restricted mean survival times are calculated as the difference in the area under the Kaplan-Meier curve up to a certain time point.
3 |. RESULTS
A total of 699 patients were identified with histologically confirmed HVMs. Sixteen (2%) were pediatric and 683 (98%) were adult patients. There were 478 patients (68%) with angiosarcoma and 221 (32%) with EHE. Median age at diagnosis was 65 years (quartile [Q] 1, Q3: 56, 75 years) for patients with angiosarcoma and 54 years (Q1, Q3: 37, 65 years) for patients with EHE (P < .001). There was no significant difference in the distribution of race across tumor categories, but there was a male predominance in angiosarcoma versus EHE (64% vs 40%, P < .001) (Table 1). There was a significant difference between the angiosarcoma and EHE groups in median (Q1, Q3) tumor size (7.5 cm [4.2, 11.7] vs 4.0 cm [2.6, 6.7], P = .0001), but not in the presence of metastasis at diagnosis (27% vs 32%).
TABLE 1.
Demographic and tumor characteristics at diagnosis for angiosarcoma and EHE Abbreviation: EHE, malignant epithelioid hemangioendothelioma.
| Angiosarcoma N (%)a | EHE N (%)a | P-value | |
|---|---|---|---|
| Patients | 478 (68) | 221 (32) | |
| Age: median (Q1, Q3) | 65 (56, 75) | 54 (37,65) | <.001 |
| Male sex | 305 (64) | 88 (40) | <.001 |
| Race | .963 | ||
| White | 403 (84) | 185 (84) | |
| Black | 34 (7) | 17 (8) | |
| Other | 41 (9) | 19 (8) | |
| Tumor size (cm) | |||
| <5 | 91 (33) | 88 (61) | <.001 |
| 5–10 | 97 (35) | 39 (27) | |
| 11–15 | 60 (22) | 11 (8) | |
| >15 | 30 (11) | 7 (5) | |
| Metastasis at diagnosis | 130 (27) | 70 (32) | .223 |
| Histologically confirmed | 445 (93) | 211 (96) | .471 |
| Overall survival, % (SE) | |||
| 1year | 17% (2%) | 80% (3%) | <.0001 |
| 2 years | 10% (1%) | 71% (3%) | |
| 3 years | 8% (1%) | 65% (3%) | |
| 5 years | 6% (1%) | 62% (4%) |
Percentages are calculated per category, as not all patients had data for each category.
The rate of surgical resection was significantly lower in patients with angiosarcoma than in patients with EHE (13% vs 32%, P < .001). However, in patients who underwent resection, there was no significant difference in the extent of liver resection performed, resected tumor size, or achieving a residual tumor (R) 0 resection margin in patients with angiosarcoma versus EHE (Table 2). There was a significantly lower number of lymph nodes examined during surgery in patients with angiosarcoma versus EHE (8% vs 49%, P < .001), and number of positive lymph nodes identified (18% vs 59%, P = .003). There was no significant difference in the 30-day re-admission rate, which was 15% in patients with angiosarcoma versus 5% in patients with EHE. The 30-day mortality (10% vs 3%) was not statistically different; however, 90-day mortality was higher in patients with angiosarcoma versus EHE (21% vs 4%, P = .003). Six (1%) patients with angiosarcoma and four (2%) patients with EHE underwent local tumor destruction. A small minority of patients received radiation therapy, and there was no significant difference in radiation usage between patients with angiosarcoma versus EHE (1% vs 5%). Fewer patients with angiosarcoma underwent chemotherapy (5% vs 10%, P = .037), but more patients with angiosarcoma pursued palliative care than those with EHE (11% vs 4%, P = .016) (Table 2).
TABLE 2.
Multimodal therapy approaches for angiosarcoma and EHE
| Angiosarcoma N (%)a | EHE N (%)a | P-value | |
|---|---|---|---|
| Surgery | 63 (13) | 71 (32) | <.001 |
| Extent of surgery | .277 | ||
| Segment/wedge | 21 (33) | 28 (39) | |
| Lobectomy | 18 (29) | 10 (14) | |
| Extended lobectomy | 6 (10) | 5 (7) | |
| Transplant | 18 (29) | 28 (39) | |
| Tumor size among those who underwent surgery (cm) | .085 | ||
| <5 | 21 (38) | 40 (61) | |
| 5–10 | 16 (29) | 13 (20) | |
| 10–15 | 11 (20) | 7 (11) | |
| >15 | 8 (14) | 6 (9) | |
| Resection margins | .544 | ||
| R0 | 45 (65) | 53 (71) | |
| R1 | 5 (7) | 6 (8) | |
| R2 | 7 (10) | 3 (17) | |
| Regional lymph nodes examined | .206 | ||
| 0 | 22 (92) | 19 (51) | |
| 1–5 | 2 (8) | 13 (30) | |
| 6–10 | 0 (0) | 3 (8) | |
| >10 | 0 (0) | 1 (3) | |
| Node examined but not counted | 0 (0) | 1 (3) | |
| Positive regional lymph nodes | .238 | ||
| 0 | 9 (82) | 13 (41) | |
| 1–5 | 1 (9) | 16 (50) | |
| 6–10 | 0 (0) | 2 (6) | |
| >10 | 0 (0) | 0 (0) | |
| Positive node but not counted | 1 (9) | 1 (3) | |
| 30-day readmissions | 11 (15) | 4 (5) | .053 |
| 30-day mortality | 6 (10) | 2 (3) | .102 |
| 90-day mortality | 13 (21) | 3 (4) | .003 |
| Radiation therapy | 5 (1) | 7 (5) | .164 |
| Chemotherapy | 20 (5) | 20 (10) | .037 |
| Neoadjuvant | 1 (0.2) | 3 (2) | |
| Adjuvant | 10 (2) | 10 (5) | |
| Intraoperative | 2 (0.5) | 0 (0) | |
| Sequence unknown | 7 (2) | 7 (4) | |
| Pursuance of palliative care | 54 (11) | 7 (4) | .016 |
| Radiation therapy | 6 (1) | 2 (1) | |
| Chemotherapy | 16 (3) | 1 (0.5) | |
| Pain management | 15 (3) | 0 (0) | |
| Any/unknown combination | 17 (4) | 4 (2) |
Abbreviation: EHE, malignant epithelioid hemangioendothelioma.
Percentages are calculated per category, as not all patients had data for each category.
3.1 |. Survivorship
The mean 1-, 3-, and 5-year overall survival for patients with angiosarcoma was 17%, 8%, and 6% respectively, versus 80%, 65%, and 62% in patients with EHE (P < .0001) (Figure 1). The restricted mean overall survival (95% confidence interval [CI]) for all patients with angiosarcoma was 14 months (10–17). Mean survivorship for patients with angiosarcoma dramatically improved if surgical resection was performed (42 months [27–57]) and was worse if resection was not performed (9 months [6–11]) (Figure 2A, Figure S1a). There were similar mean overall survivals for patients who underwent resection alone (13 months [5–20]) versus resection with chemotherapy (11 months [4–18]) versus chemotherapy alone (14 months [9–19]) versus neither chemotherapy nor resection (13 months [8–17]) (Figure 3A). There was an improved mean overall survival (95% CI) for patients with angiosarcoma who underwent lobectomy or extended lobectomy (74 months [47–101]) versus orthotopic liver transplantation (13 months [6–20]). There was a significantly worse overall survival for patients with angiosarcoma with lymph nodes examined (24 months [3–44]) than not examined (44 months [29–60]). There was not a significant difference in mean overall survival (95% CI) between patients with angiosarcoma who had R0 versus non-R0 resection (47 months [29–66] vs 21 months [2–40], P = .09). Finally, there was a significantly longer mean overall survival (95% CI) for angiosarcoma tumors <10 cm versus ≥10 cm (23 months [16–30] vs 12 months [5–19], P = .001).
FIGURE 1.
Overall survival for pediatric and adult patients with hepatic vascular malignancies (P < .001)
FIGURE 2.
Overall survival for all patients with (A) angiosarcoma and (B) malignant epithelioid hemangioendothelioma who received surgical versus nonsurgical treatment (P < .001)
FIGURE 3.
Overall survival of all patients with (A) angiosarcoma and (B) malignant epithelioid hemangioendothelioma who received surgical treatment alone versus surgical and chemotherapeutic treatment versus chemotherapeutic treatment alone (P < .0001)
The mean overall survival (95% CI) for all patients with EHE was 98 months (87–109). Similar to angiosarcoma, survivorship for patients with EHE improved if surgical resection was performed (122 months [107–136]), and was worse if resection was not performed (82 months [68–96]) (Figure 2B, Figure S1b). Mean overall survivals for patients who underwent resection alone (121 months [99–142]) and those who underwent resection with chemotherapy (128 months [95–161]) were better than those with chemotherapy alone (69 months [50–88]) or neither chemotherapy nor resection (99 months [84–113]) (Figure 3B). The mean overall survival (95% CI) for patients with EHE was similar for those who underwent lobectomy or extended lobectomy (96 months [66–126]) versus orthotopic liver transplantation (118 months [98–138]). There was no significant difference in mean overall survival (95% CI) between patients with EHE who had all negative lymph nodes versus at least one positive lymph node identified (115 months [78–153] vs 106 months [81–132], P = .99), R0 vs non-R0 resection (117 months [102–132] vs 100 months [68–132], P = .78), or tumors <10 cm versus ≥10 cm (100 months [88–112] vs 78 months [50–106], P = .19).
3.2 |. Pediatric patients
Subgroup analysis was performed on pediatric patients and found six (1%) with angiosarcoma and 10 (5%) with EHE. The median (Q1, Q3) age of presentation of pediatric patients with angiosarcoma was 7 years (0, 20) and 15.5 years (13, 19) in pediatric patients with EHE. Pediatric patients with angiosarcoma and EHE had a similar demographic distribution, tumor size, lymph node involvement, metastasis, and rates of treatment with resection, chemotherapy, radiation, and palliation to the adult patients. There were two (33%) pediatric patients with angiosarcoma and six (60%) pediatric patients with EHE who underwent surgical resection (Figure S1c,d). The mean 1-, 3-, and 5-year overall survival for pediatric patients with angiosarcoma was 67%, 50%, and 50%, respectively, and 90%, 90%, and 90% in pediatric patients with EHE, both of which are higher than the adult populations (17%, 7%, 5% and 80%, 64%, 61%, respectively) (Figure S2). The restricted mean survival (95% CI) for patients with angiosarcoma was significantly longer for pediatric patients than adult patients (80 months [23–137] vs 13 months [9–16], P = .004), whereas for patients with EHE there was a trend of longer survival for pediatric patients than adult patients (123 months [98–149] vs 97 months [86–109], P = .15).
4 |. DISCUSSION
In this large national study of children and adults with HVMs, we found that the mean 1-, 3-, and 5-year overall survival for patients with angiosarcoma was 17%, 8%, and 6%, respectively, versus 80%, 65%, and 62% for patients with EHE (P < .0001). The mean 1-, 3-, and 5-year overall survival for pediatric patients with angiosarcoma was 67%, 50%, and 50%, respectively, and 90%, 90%, and 90% in pediatric patients with EHE, both of which are much higher than the respective adult populations. The adult survival is consistent with smaller prior studies of hepatic angiosarcoma, as the largest retrospective multicenter study of 44 adult patients reported similarly poor 1-, 3-, and 5-year overall survival of 30.0%, 8.1%, and 5.6%, respectively.25 The pediatric outcomes reported here, however, are much better than recent case reports in the literature, which all report dismal outcomes with few survivors.24 The overall survival for EHE in this study was also higher than previously reported, with a mean 1-, 3-, and 5-year overall survival of 83.4%, 55.8%, and 41.1%, respectively.26 The dramatically different overall survival between angiosarcoma and EHE stresses the importance of distinguishing these types of HVMs histologically. Unfortunately, these tumor types are often collapsed into a single group of malignancies in many larger studies, especially those using large national databases.
Surgical resection is often the preferred therapy for HVMs; however, less than 15–25% of patients with angiosarcoma and 10–40% of patients with EHE present with localized liver involvement and resectable disease.4,25 In our study, 13% of adult patients with angiosarcoma and 32% of adult patients with EHE underwent surgical resection in line with prior reports, while 33% of children with angiosarcoma and 60% of children with EHE were able to undergo surgical resection. Although the number of patients in the pediatric cohort is smaller, and likely underpowered to detect a statistical difference, these discrepancies bring into question whether HVMs have a more favorable pathogenesis and/or are treated more aggressively in children. It is unlikely that children present at an earlier stage than in the adult population as the tumor size and rate of metastasis was not found to be statistically significant. Therefore, this may suggest that if surgical resection were performed more frequently in adults, a greater percentage of patients could benefit.
A review on angiosarcoma by Zheng et al reported that 76.5% of those who undergo complete surgical resection alone or in combination with chemotherapy survived longer than 6 months.22 In our study, survivorship for patients with angiosarcoma dramatically increased if resection was performed (42 months [27–57]) than if resection was not performed (9 months [6–11]). Survivorship for patients with EHE also increased if resection was performed (122 months [107–136]) than if resection was not performed (82 months [68–96]). Similarly, improved rates of survival were reported in a review on EHE by Mehrabi et al, suggesting that the 5-year survival for patients who underwent surgical resection was 55–75% versus 5–30% for those who did not.26 The survival difference in both types of HVMs supports the conclusion that patients able to undergo surgical resection for either type of HVM have a dramatically improved overall survival, and also that patients with EHE have a more favorable tumor biology than angiosarcoma. Efforts should therefore focus on prioritizing surgical resection for all patients with HVMs whenever possible.
In this study, there was no significant difference in the extent of liver resection between tumor types or age groups; however, the pediatric cohort is likely underpowered to detect a meaningful difference. There did not appear to be a standard surgical approach, with similar proportions of wedge resection (33%), lobectomy (29%), and transplantation (29%) in patients with angiosarcoma and wedge resection (39%) or transplantation (39%) in patients with EHE. The extent of resection performed varies in the literature, but a report from Tripke et al advocates for anatomical hemihepatectomy or radical hepatectomy for curative intent in angiosarcoma.27 Their report also stressed the importance of R0 resection, as the nine patients that achieved this result had a median overall survival of 59 months. While there was a trend supporting R0 resection in our study, we did not find a statistically significant difference in overall survival based upon achieving R0 resection in either type of HVM. However, due to missing data in NCDB, this result is likely underpowered to detect a difference.
HVMs have previously been considered a contraindication to orthotopic liver transplantation in the adult population due to high recurrence rate in the allograft and poor median posttransplant survival of <6 months.28 Despite this recommendation, there are a number of case reports of transplantation for HVMs, especially in the pediatric population and patients with EHE.29 Furthermore, some patients with HVMs have been reported to undergo transplantation due to the presence of multifocal tumors that are not amenable to surgical resection.26 This study included 18 (29%) patients with angiosarcoma and 28 (39%) patients with EHE who underwent transplantation. The transplant cohort’s mean overall survival for patients with angiosarcoma was 13 months (95% CI: 6–20), which is modestly improved compared with previously published estimates.25,28 Further, patients with angiosarcoma who underwent lobectomy or extended lobectomy had a much longer mean overall survival (74 months [47–101]) than the transplanted cohort. As patients who underwent transplantation for angiosarcoma had poor overall survival and worse overall survival compared to the resection cohort, it supports the current belief that angiosarcoma is most often not an appropriate malignancy for transplantation.28 In contrast, the EHE transplanted cohort’s mean overall survival was 118 months (98–138), which is much improved from previously published results.29 Additionally, for patients with EHE, mean overall survival was slightly worse for patients who underwent lobectomy or extended lobectomy (96 months [66–126]), suggesting EHE could be an appropriate malignancy for transplantation. Further studies to confirm these findings would be helpful in informing a discussion of whether HVMs should remain a contraindication for transplantation, especially in patients with EHE.
Alternative oncologic treatments such as locoregional transcatheter chemotherapy and ablative techniques have been shown to improve survival in other hepatic tumors for unresectable disease.30,31 In HVMs, transcatheter embolization has been utilized in the setting of acute hemorrhage, while transarterial chemoembolization has been utilized in unresectable disease.32 Six (1%) patients with angiosarcoma and four (2%) patients with EHE in this study underwent local tumor destruction. NCDB does not provide the granularity to determine what type of local tumor destruction techniques were used in the dataset, so larger studies with more granularity are needed to verify the clinical benefit of differing modalities of local tumor destruction.
From the few reports in the literature, there is no consensus on dose or selection of chemotherapeutic agents for HVMs. There is also no proven survivorship benefit.18 Typical primary chemotherapy regimens reported are Adriamycin- or Taxol-based regimens or include newer agents such as paxopanib.22,25 In this study, only 20 patients with angiosarcoma (5%) and 20 patients with EHE (10%) received chemotherapy, which is lower than other reports of up to 40% for angiosarcoma and 21% for EHE.18,26 In our study, there was also no apparent mean overall survival benefit with the addition of chemotherapy versus resection alone for patients with angiosarcoma (11 months [4–18] vs 13 months [5–20]) or for patients with EHE (128 months [95–161] vs 121 months [99–142]). Radiation treatment modalities have also been reported, but HVMs are generally considered radioresistant, and again there is no consensus on dose or frequency.1 In this study very few patients underwent radiotherapy (five [1%] angiosarcoma vs seven [5%] EHE).
Limitations of this study include the lack of granularity afforded by utilizing a large database that was not created for these specific malignancies. However, the number of patients able to be examined through a national database outweighs the granularity limitation.33 Although this study is the largest HVM study to date in both the adult and pediatric populations, the small number of patients with these rare malignancies—especially in the pediatric population—makes it difficult to draw firm conclusions about best practices for diagnosis and treatment. Further, the NDCB does not differentiate all-cause mortality from mortality related to oncologic disease, therefore the survival rate could be higher than reported. Also, recurrent disease is a common occurrence in HVMs, but the NCDB does not monitor for recurrence, and therefore this remains to be explored in future prospective studies. Finally, as mentioned throughout the manuscript, angiosarcoma and EHE are often misclassified histologically and grouped together under one malignancy. While there is no way to retroactively review all pathology specimens, future efforts should focus on clearly distinguishing these malignancies, as survival outcomes are dramatically different. If in doubt, the authors would encourage all clinicians to send pathology samples to specialized centers with more expertise in HVMs to ensure the correct diagnosis is achieved.
5 |. CONCLUSIONS
In the largest study of HVMs to date, we found that although often difficult to distinguish from each other, patients with angiosarcoma have a significantly worse overall survival than patients with EME. Furthermore, that pediatric patients with angiosarcoma and EHE are associated with higher rates of surgical resection and longer overall survival. Although the pediatric cohort is small, these discrepancies bring into question whether these tumors have a more favorable pathogenesis than in the adult population and if children are treated more aggressively than adults. Further studies, including studies of standardized ways to distinguish angiosarcoma from EHE, reports examining the biology of resected pediatric versus adult tumor specimens, along with treatment trials, are needed. Such studies would require coordination and commitments across institutions, and even internationally, given the rare nature of these tumors, but could yield invaluable insights and improve survivorship for pediatric and adult patients with HVMs.
Supplementary Material
FUNDING INFORMATION
National Institute of General Medical Sciences, NIH; Eunice Kennedy Shriver National Institute of Child Health & Human Development Pediatric Clinical Pharmacology Fellowship, Award Number: T32GM086330
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Sarah Jane Commander was supported by the National Institute of General Medical Sciences of the National Institutes of Health and the Eunice Kennedy Shriver National Institute of Child Health & Human Development Pediatric Clinical Pharmacology Fellowship under Award Number T32GM086330. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Abbreviations:
- CD
cluster of differentiation
- CI
confidence interval
- EHE
malignant epithelioid hemangioendothelioma
- HVM
hepatic vascular malignancies
- NCDB
National Cancer Database
- Q
quartile
- R
residual tumor
Footnotes
DISCLOSURE STATEMENT
The National Cancer Database (NCDB) is a joint project of the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The CoC’s NCDB and the hospitals participating in the CoC NCDB are the source of the de-identified data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.
SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of the article.
REFERENCES
- 1.Chaudhary P, Bhadana U, Singh RA, Ahuja A. Primary hepatic angiosarcoma. Eur J Surg Oncol. 2015;41(9):1137–1143. [DOI] [PubMed] [Google Scholar]
- 2.Antonescu C. Malignant vascular tumors-an update. Mod Pathol. 2014;27(Suppl 1):S30–S38. [DOI] [PubMed] [Google Scholar]
- 3.Costa MA, Sousa A, Vieira E. Hemangioendothelioma: a rare vascular tumor in childhood and adolescence. Pediatr Hematol Oncol. 1996;13(4):333–337. [DOI] [PubMed] [Google Scholar]
- 4.Studer LL, Selby DM. Hepatic epithelioid hemangioendothelioma. Arch Pathol Lab Med. 2018;142(2):263–267. [DOI] [PubMed] [Google Scholar]
- 5.Molina E, Hernandez A. Clinical manifestations of primary hepatic angiosarcoma. Dig Dis Sci. 2003;48(4):677–682. [DOI] [PubMed] [Google Scholar]
- 6.Locker GY, Doroshow JH, Zwelling LA, Chabner BA. The clinical features of hepatic angiosarcoma: a report of four cases and a review of the English literature. Medicine (Baltimore). 1979;58(1):48–64. [DOI] [PubMed] [Google Scholar]
- 7.Leowardi C, Hormann Y, Hinz U, et al. Ruptured angiosarcoma of the liver treated by emergency catheter-directed embolization. World J Gastroenterol. 2006;12(5):804–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Harrison JR, Faust TW, Blackstone MO. Hepatic angiosarcoma: an unusual cause of congestive heart failure. South Med J. 2001;94(3):336–338. [PubMed] [Google Scholar]
- 9.Popper H, Thomas LB, Telles NC, Falk H, Selikoff IJ. Development of hepatic angiosarcoma in man induced by vinyl chloride, thorotrast, and arsenic. Comparison with cases of unknown etiology. Am J Pathol. 1978;92(2):349–376. [PMC free article] [PubMed] [Google Scholar]
- 10.Elliott P, Kleinschmidt I. Angiosarcoma of the liver in Great Britain in proximity to vinyl chloride sites. Occup Environ Med. 1997;54(1):14–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ito Y, Kojiro M, Nakashima T, Mori T. Pathomorphologic characteristics of 102 cases of thorotrast-related hepatocellular carcinoma, cholangiocarcinoma, and hepatic angiosarcoma. Cancer. 1988;62(6):1153–1162. [DOI] [PubMed] [Google Scholar]
- 12.Falk H, Thomas LB, Popper H, Ishak KG. Hepatic angiosarcoma associated with androgenic-anabolic steroids. Lancet. 1979;2(8152):1120–1123. [DOI] [PubMed] [Google Scholar]
- 13.Rosenthal AK, Klausmeier M, Cronin ME, McLaughlin JK. Hepatic angiosarcoma occurring after cyclophosphamide therapy: case report and review of the literature. Am J Clin Oncol. 2000;23(6):581–583. [DOI] [PubMed] [Google Scholar]
- 14.Huang NC, Wann SR, Chang HT, Lin SL, Wang JS, Guo HR. Arsenic, vinyl chloride, viral hepatitis, and hepatic angiosarcoma: a hospital-based study and review of literature in Taiwan. BMC Gastroenterol. 2011;11:142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ohmoto K, Hirokawa M, Takesue M, Yamamoto S. Hepatic angiosarcoma with early central enhancement and arterioportal shunt on dynamic CT. Hepatogastroenterology. 2000;47(36):1717–1718. [PubMed] [Google Scholar]
- 16.Bruegel M, Muenzel D, Waldt S, Specht K, Rummeny EJ. Hepatic angiosarcoma: cross-sectional imaging findings in seven patients with emphasis on dynamic contrast-enhanced and diffusion-weighted MRI. Abdom Imaging. 2013;38(4):745–754. [DOI] [PubMed] [Google Scholar]
- 17.Wang L, Lv K, Chang XY, et al. Contrast-enhanced ultrasound study of primary hepatic angiosarcoma: a pitfall of non-enhancement. Eur J Radiol. 2012;81(9):2054–2059. [DOI] [PubMed] [Google Scholar]
- 18.Zhu YP, Chen YM, Matro E, et al. Primary hepatic angiosarcoma: a report of two cases and literature review. World J Gastroenterol. 2015;21(19):6088–6096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Smith EH, Complications of percutaneous abdominal fine-needle biopsy. Review. Radiology. 1991;178(1):253–258. [DOI] [PubMed] [Google Scholar]
- 20.Kojiro M, Nakashima T, Ito Y, Ikezaki H, Mori T, Kido C. Thorium dioxide-related angiosarcoma of the liver. Pathomorphologic study of 29 autopsy cases. Arch Pathol Lab Med. 1985;109(9):853–857. [PubMed] [Google Scholar]
- 21.Wang ZB, Yuan J, Chen W, Wei LX. Transcription factor ERG is a specific and sensitive diagnostic marker for hepatic angiosarcoma. World J Gastroenterol. 2014;20(13):3672–3679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zheng YW, Zhang XW, Zhang JL, et al. Primary hepatic angiosarcoma and potential treatment options. J Gastroenterol Hepatol. 2014;29(5):906–911. [DOI] [PubMed] [Google Scholar]
- 23.Lau K, Massad M, Pollak C, et al. Clinical patterns and outcome in epithelioid hemangioendothelioma with or without pulmonary involvement: insights from an internet registry in the study of a rare cancer. Chest. 2011;140(5):1312–1318. [DOI] [PubMed] [Google Scholar]
- 24.Grassia KL, Peterman CM, Iacobas I, et al. Clinical case series of pediatric hepatic angiosarcoma. Pediatr Blood Cancer. 2017;64(11):e26627. [DOI] [PubMed] [Google Scholar]
- 25.Wilson GC, Lluis N, Nalesnik MA, et al. Hepatic angiosarcoma: a multi-institutional, international experience with 44 cases. Ann Surg Oncol. 2019;26(2):576–582. [DOI] [PubMed] [Google Scholar]
- 26.Mehrabi A, Kashfi A, Fonouni H, et al. Primary malignant hepatic epithelioid hemangioendothelioma: a comprehensive review of the literature with emphasis on the surgical therapy. Cancer. 2006;107(9):2108–2121. [DOI] [PubMed] [Google Scholar]
- 27.Tripke V, Heinrich S, Huber T, et al. Surgical therapy of primary hepatic angiosarcoma. BMC Surg. 2019;19(1):5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bonaccorsi-Riani E, Lerut JP. Liver transplantation and vascular tumours. Transpl Int. 2010;23(7):686–691. [DOI] [PubMed] [Google Scholar]
- 29.Ercolani G, Grazi GL, Pinna AD. Liver transplantation for benign hepatic tumors: a systematic review. Dig Surg. 2010;27(1):68–75. [DOI] [PubMed] [Google Scholar]
- 30.Liapi E, Geschwind JF. Transcatheter and ablative therapeutic approaches for solid malignancies. J Clin Oncol. 2007;25(8):978–986. [DOI] [PubMed] [Google Scholar]
- 31.Brown DB, Nikolic B, Covey AM, et al. Quality improvement guidelines for transhepatic arterial chemoembolization, embolization, and chemotherapeutic infusion for hepatic malignancy. J Vasc Interv Radiol. 2012;23(3):287–294. [DOI] [PubMed] [Google Scholar]
- 32.Park YS, Kim JH, Kim KW, et al. Primary hepatic angiosarcoma: imaging findings and palliative treatment with transcatheter arterial chemoembolization or embolization. Clin Radiol. 2009;64(8):779–785. [DOI] [PubMed] [Google Scholar]
- 33.Rice HE, Englum BR, Gulack BC, et al. Use of patient registries and administrative datasets for the study of pediatric cancer. Pediatr Blood Cancer. 2015;62(9):1495–1500. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.