Pleuropulmonary blastoma (PPB) is a rare and highly aggressive pulmonary malignancy in children. In 1961 Spencer first coined the term and suggested that PPB arose from mesodermal blastema because of its similarities to nephroblastoma.1 In 1988 Manivel et al. described PPB in children as an entity that was distinct from the biphasic epithelial-stromal morphology of the classic adult type.2 In pediatric patients the lesion is a true dysembryogenic neoplasm of thoracopulmonary mesenchyma, without malignant epithelial cells. This tumor is characterized histologically by primitive blastema and a malignant mesenchymal stroma that often shows multidirectional differentiation (a rhabdomyosarcomatous, chondrosarcomatous, or liposarcomatous pattern). Immunohistochemical studies add information to the diagnostic assessment. Recently, Priest et al. reported that in 25% of cases, PPB patients or their siblings have other dysplastic or neoplastic conditions.3
PPB in children age < 5 years is characterized clinically by symptoms often mistaken for respiratory tract infection or pneumothorax. The tumor usually is located in lung periphery, but it may be extrapulmonary with involvement of the mediastinum, diaphragm, and/or pleura. Common metastatic sites include the brain, bone, lymph nodes, liver, pancreas, kidney, and adrenal glands.4–6
Despite the introduction of multimodal therapy (surgery, chemotherapy, and occasionally radiotherapy), the prognosis for PPB patients remains poor.
MATERIALS AND METHODS
- Top of page
- MATERIALS AND METHODS
The current study reports the clinical and pathologic findings of PPB cases observed in five Italian Associations for Pediatric Hematology and Oncology (AIEOP) centers between 1982 and 1998. This was part of a national retrospective search for rare pediatric tumors conducted by 18 AIEOP centers. Patients were enrolled after informed consent was obtained from their parents or guardians.
Eleven children (7 boys and 4 girls) were included in this analysis. The youngest patient was age 2 weeks at the time of diagnosis and the oldest was age 96 months (median age, 32 months). Clinical data, surgical notes, pathologic findings, and summaries of chemotherapy and radiotherapy were taken from the charts. Pathologic material from all patients was reviewed by one of the authors (V.N.). These cases were subclassified histologically as described previously by Dehner et al.4: type 1 PPB (exclusively cystic without a macroscopically detectable solid component and most likely a better prognosis), type 2 PPB (with solid and true cystic areas), and type 3 PPB (a true solid tumor). Tumor size was measured by computed tomography scans.
The extent of surgical resection was estimated after detailed review of surgical and histopathologic notes as: 1) biopsy only; 2) total resection, when tumors were free of surgical margins; 3) partial resection, when macroscopic tumor remained; and 4) macroscopic total resection, when the surgeon reported total resection but no histologic examination of the surgical margins was performed. Evaluation of tumor response was performed after 9–10 weeks of chemotherapy administration. Response to treatment, based on the degree of tumor volume reduction obtained from clinical and imaging evidence, was defined as follows: complete response (CR): the complete disappearance of disease; partial response (PR): a reduction in tumor volume of > 67%; and minor response (MR): a reduction > 33% but < 67%. Stable disease or a reduction in tumor volume of < 33% was recorded as no response, whereas an increase in tumor size or the detection of new lesions was considered as disease progression.
Standard statistical descriptions of parameters were used to characterize the data (mean, median, and range). Multiple logistic regression analysis was employed to calculate estimates of relative risk. Variables included in the regression model were gender, age at the time of diagnosis, side of tumor involvement (right vs. left), tumor size, and the presence of primary dysplastic disease. Overall survival and event free survival curves were estimated using the Kaplan–Meier method.
Survival was calculated for all patients from the date of diagnosis to the date of death or last follow-up if alive. Event free survival was calculated from the date of diagnosis to the date of induction failure, disease recurrence, death, or last follow-up if alive. The data were analyzed as of April 2000. The statistical significance level was set at P < 0.05.
- Top of page
- MATERIALS AND METHODS
The major clinicopathologic features of 11 cases are summarized in Table 1. In the patients in the current study the initial symptoms were fever (five patients), cough (five patients), respiratory distress (six patients), abdominal pain (one patient), and anorexia and weight loss (one patient). The initial symptoms were not known for one patient. At diagnosis, the primary tumor sites were limited to 1 hemithorax in 10 cases: 6 patients on the right side and 4 patients on the left side, with pleural involvement noted in 5 patients. Only one patient (Patient 6) presented with an exclusively pleural tumor. Involvement of the mediastinum, diaphragm, and pericardium, respectively, were recorded in three patients. Finally, one patient presented with bone metastasis at the time of diagnosis. In three cases the PPB developed from other primary dysplastic diseases: cystic adenomatoid malformation in one case and congenital lung cysts in two cases. Bilateral tumors were not observed. The lower right lobe was involved in four patients, the upper right lobe was involved in one patient, involvement of the middle lobe was in two patients. In one patient the disease was located in the upper left lobe and in three patients it was located respectively in the middle lobe and in the lower left lobe with involvement of the middle lobe reported in one patient (Patient 10). Tumor size was > 5 cm in 9 patients and unknown in 2 patients.
Table 1. Clinicopathologic Features of the 11 Pleuropulmonary Blastoma Patients Studied
|Patient no.||Age (mos)||Gender||Sites||Pathology typesa||Malformations|
Therapy and Clinical Outcome
The therapeutic management as well as the clinical outcome of the 11 PPB patients in the current study is summarized in Table 2. Total resection was performed in two patients with tumorectomy and lobectomy, respectively; partial resection was performed in two patients with tumorectomy; and macroscopic total resection was performed in four patients (two lobectomies, one tumorectomy, and one pneumonectomy). In this last patient pleural and pericardial involvement did not allow radical surgery. Finally, in three patients only surgical biopsy was performed.
Table 2. Therapy, Evolution, and Outcome of the 11 Patients with Pleuropulmonary Blastoma Studied
|Patient no.||Extent of surgery||Primary chemotherapy||Primary radiotherapy||Delayed surgery||Site(s) of recurrence or metastasis||Time to recurrence (mos)||Further therapy||Outcome||Length of follow-up (mos)|
|1||Macroscopic total||CEVAIE||Yes (44 Gy)||No||Local||22||CEVAIE||AWD||34|
|5||Macroscopic total||VAIA||Yes (44 Gy)||No||CNS||10||CH/XRT||DOD||18|
|8||Macroscopic total||VAIA||Yes (44 Gy)||Biopsy: no tumor||No||—||No||NED||119|
|9||Biopsy||(C + E) + IVA||No||Total: no tumor||No||—||No||NED||39|
|10||Biopsy||(Ep + E) + IVA||No||Macroscopic total||Local||3||XRT 20 Gy||DOD||6|
Delayed surgery was performed in three patients. A macroscopic total excision of the lesion was performed in 1 patient and total resection was performed in 2 patients who underwent biopsy at the time of diagnosis with tumor shrinkage of > 33% but < 67% after 9–10 weeks of initial chemotherapy. In this last patient, histology showed total tumor necrosis. After 39 months of follow-up 1 patient was healthy and the other died after disease recurrence. In the third patient (who had no clinical evidence of disease after chemotherapy and radiotherapy and underwent macroscopic total resection as initial surgery), no tumor was present at delayed biopsy. After 119 months of follow-up, this patient was without any evidence of disease. No adjuvant therapy was planned for one patient (Patient 7) who underwent primary radical surgery, the results of which were confirmed by second-look exploration.
The remaining patients received drug treatment. The majority of these drugs (vincristine, epirubicin, actinomycin D, ifosfamide, etoposide, doxorubicin, and carboplatin) also are included in therapy protocols for childhood soft tissue sarcoma. The groups of drugs used in our patients were carboplatin, epirubicin, vincristine, actinomycin D, ifosfamide, and etoposide (CEVAIE) (five patients); vincristine, actinomycin D, ifosfamide, and doxorubicin (VAIA) (two patients); (carboplatin plus etoposide) plus ifosfamide, vincristine, and actinomycin D (IVA) (one patient); and (epirubicin plus etoposide) plus IVA (one patient) (Table 2). In 6 patients tumor response was estimated after 9–10 weeks of initial chemotherapy; 2 patients with macroscopic residual disease at diagnosis were found to have achieved CR, (Patients 4 and 6); an MR was noted in 2 patients who underwent surgical biopsy only (Patients 9 and 10); and disease progression was noted in 2 other patients (Patients 3 and 5).
Five of 11 patients developed disease recurrence (local recurrence in 3 patients at 3, 9, and 22 months, respectively, after diagnosis; and central nervous system [CNS] and intraocular metastasis in 2 patients at 10 and 14 months, respectively, after diagnosis). Two of these children survived well into their second CR (1 patient after local recurrence and the other patient after intraocular metastasis) 14 and 63 months, respectively, from recurrence.7 Two cycles of carboplatin and etoposide after 45 grays (Gy) of local radiotherapy was planned in the last two patients. At last follow-up a third child was alive with disease after developing a local recurrence. Two other patients died from disease progression 3 and 8 months, respectively, after disease recurrence despite multimodal therapy. Finally, one patient died from disease progression after the initial chemotherapy.
At last follow-up 8 of the 11 patients in the current study were alive after a median of 36.5 months. Five of these patients were without disease recurrence and/or metastases after a median of 39 months of follow-up. Two were disease free after disease recurrence and one patient was alive with disease. Three patients died of disease. Overall survival at 2 years was 72% (95% confidence limits [95% CL], 46–99%) for all patients. Event free survival at 2 years from diagnosis was 45% (95% CL, 16–75%) for all patients (Fig. 1). Gender, age, side of involvement, tumor size, and the presence of primary dysplastic disease did not appear to influence survival significantly by univariate analysis. Extrapulmonary involvement at the time of diagnosis appeared to be related to prognosis. Five of six patients with pleural involvement developed disease recurrence or presented with disease progression, two of whom died. One of these five patients, with parenchymal involvement only, developed disease recurrence and died of disease.
Figure 1. Overall survival (SUR) and event free survival (EFS) of patients as calculated by the Kaplan–Meier method. 95% CL: 95% confidence limits.
Download figure to PowerPoint
Histologically, the three type 2 PPB patients all developed disease recurrence, two of whom died; of the seven type 3 PPB patients, three also developed a disease recurrence. Only the patient with type 1 PPB was alive and healthy at the time of last follow-up.
- Top of page
- MATERIALS AND METHODS
PPB is a very rare tumor of controversial histogenesis. It is a highly aggressive tumor in children with a projected overall survival rate of 45% at 5 years and an event free survival rate of 49% at 2 years in the population studied by Priest et al..8 The rarity of PPB has allowed only slow elucidation of its clinical features according to prognosis and its response to therapy. Therefore, the data presented regarding the 11 patients observed in Italy might provide further insight into the management of this rare disease.
The clinical presentation of the patients in the current study often was misinterpreted as an upper respiratory tract infection. For this reason the diagnosis of PPB was delayed for up to 45 days. At the time of diagnosis mediastinal or pleural involvement was found to be correlated significantly with a poorer survival. In keeping with the findings of Priest et al.8 the current study data appear to confirm their findings: five of six patients developed a disease recurrence.
PPB usually develops in younger children without congenital pulmonary malformations. Delahunt et al. and Joshi et al. described a familial association between PPB and cystic nephroma.9, 10 Others have reported lung cysts arising 4 months to 3 years before the time of diagnosis to be a predisposition to the development of malignancy.3, 11–15 For this reason surgical resection should be considered for any lesions with a solid mass.16, 17
Total resection of the tumor at the time of diagnosis appears to be the preferred treatment with, if necessary, a lobectomy or a pneumonectomy. Furthermore, in the majority of patients initial surgery is incomplete because of massive tumor involvement. Two of the patients in the current study who underwent total resection at the time of diagnosis were alive and healthy at the time of last follow-up, as was one other patient who achieved a CR after delayed surgery.
Microscopic residual disease after the initial surgery has been found to result in a 50% event free survival rate for all patients; 2 of the 4 patients in the current study were healthy, as were 9 of the 19 patients reported by Priest et al. after 14–199 months of follow-up (mean, 57.3 months; median, 49 months).8 Conversely, partial resection or biopsy at the time of diagnosis does not appear to prevent local recurrence or metastases. Therefore, radical surgery even with microscopic residual disease remains the cornerstone of PPB management. For this reason, patients with initially unresectable tumors should be treated with neoadjuvant chemotherapy to reduce the lesion to the point that it becomes resectable. It is interesting to note that in the current study cases, one patient with type 1 PPB without extrapulmonary involvement and who underwent total resection at the time of diagnosis was healthy after undergoing surgery only, similar to two other patients described in the literature.8 However, confirmation of these findings in a larger number of cases is required to obtain a subset of patients with a better prognosis whose disease was managed with surgery only.
To our knowledge the role of combination chemotherapy in these patients has not been defined on a statistically significant level, but common opinion dictates that chemotherapy should be considered for these children based on the aggressiveness of PPB. There are numerous reports regarding the need for combination chemotherapy for PPB,2, 5, 8, 18–21 although to our knowledge there is no worldwide agreement regarding the modality.22 Schmaltz et al. reported that the administration of high dose chemotherapy followed by autologous blood stem cell transplantation did not provide any significant benefit in terms of clinical outcome.23 Recently, Ozkaynak et al. argued that chemotherapy is important but should be tailored to treat sarcomatous and carcinomatous elements. Unfortunately, this argument does not take into account the histologic differences between adult pulmonary blastoma and pediatric PPB.24 The results of the current study favor chemotherapy, with two CRs reported after partial resection at initial surgery and no disease or microscopic residual disease found at second-look surgery in two other patients who underwent biopsy at the time of diagnosis. In the last two patients the tumor showed a minor response after the first course of chemotherapy. The same change in the homogeneity of the mass likely was secondary to tumor necrosis.24 Therefore, the findings of the current study confirm the effectiveness of chemotherapy, although chemotherapy alone is not able to offer a better prognosis; three of four patients with a better response to chemotherapy developed disease recurrence or metastasis. Chemotherapy appears to more efficacious after radical or microscopic residual disease is detected at initial surgery.
Although there is disagreement in the literature, local radiotherapy also has been applied to PPB. In a report of 16 children who received radiotherapy, 8 developed disease recurrence, in 5 of whom it developed in the irradiated field. To our knowledge survival rates do not differ significantly between patients who receive radiotherapy and those who do not (2-year survival rate of 64% vs. 65%).8 However, the irradiated patients presented with worse features; 13 of the 16 patients had extrapulmonary involvement, there were no patients with type 1 PPB, and one patient underwent total resection at the time of diagnosis. Other authors have emphasized the poor outcome after radiotherapy with 45 Gy to the thorax and mediastinum.24 However, our experience is better; in three patients radiotherapy after the initial radical surgery controlled the disease. Two patients developed a disease recurrence outside the irradiated field. Thus, the results of the current study suggest that local radiotherapy is an effective adjunct therapy with which to manage this disease.
Currently, local recurrence and distant metastasis frequently occur after or during therapy. Brain metastases are the most common distant lesion and usually are fatal.2, 8, 19, 20, 23 One of the patients in the current study (Patient 5) developed brain metastasis and died despite multimodal therapy. We know of one patient reported in the literature who was alive and free of disease after chemoradiotherapy after 24 months of follow-up.5 The aggressiveness of PPB is manifested by early local recurrence and/or metastases. In the cases in the current study these events occurred after a median of 10 months (range, 3–22 months) from the time of diagnosis, as in most of the patients studied by Priest et al.8 However, the relative frequency of CNS deposits, occurring even 60 months after diagnosis, suggests long term monitoring may be prudent.8 Bone is the second most common site of recurrence. Therefore, we believe the CNS and bone are the sites that should receive long term monitoring.
To our knowledge adequate therapy has yet to be defined for all children with this neoplasm. It is our opinion that patients with PPB type 1, without extrapulmonary involvement and who undergo total resection at the time of diagnosis, could be cured by radical surgery alone. Prospective studies in a larger number of patients are needed to confirm these findings. In other cases the main goal of therapy is radical surgery, even in patients with microscopic residual disease. Our experience suggests front-line chemotherapy (i.e., CEVAIE) is effective in the majority of patients when it is associated with local radiotherapy if there is a poor response to chemotherapy.
The issue remains how long should chemotherapy continue after surgery? Moreover, the poor prognosis of PPB patients requires a review of more cases with the administration of high doses of chemotherapy followed by autologous blood stem cell transplantation. It is the authors' opinion that the only way to answer these questions is through an international cooperative study of PPB.