The Role of Surgical Management in Pediatric ... - Semantic Scholar

1 downloads 175 Views 718KB Size Report
of first relapse or the date of last free follow- up. The OS was ... 1), sacrococcygeal and orbit in a single patient ea
Journal of the Egyptian Nat. Cancer Inst., Vol. 20, No. 1, March: 70-79, 2008

The Role of Surgical Management in Pediatric Germ Cell Tumors (GCTs), NCI Case Series IBRAHIM M.Y. FAKHR, M.D., M.R.C.S.*; EL-SAYED ASHRAF KHALIL, M.D., F.R.C.S.I.*; TAREK S. EL-BARADIE, M.D.*; MOHAMED A. SHAALAN, M.D.*; LOBNA M. SHALABY, M.D.**; SHIMAA L.A. NASSIF, M.Sc.** and IMAN G. FARAHAT, M.D.*** The Departments of Surgical Oncology*, Pediatric Oncology** and Pathology***, National Cancer Institute (NCI), Cairo University.

with irresectable disease had comparable overall survival while none could be rendered disease free with chemotherapy.

ABSTRACT Purpose: To review the experience of a tertiary referral center in pediatric germ cell tumors (GCTs) in the last 8 years and to investigate the impact of surgery and site of disease on prognosis.

Conclusion: The initial surgical approach to malignant GCTs at all sites should be complete resection when possible; the morbidity of extensive surgical resection should be weighed carefully against the good tumor control with chemotherapy. Surgical staging does not preclude preservation of fertility, which should always be considered in this young age. The site of primary disease plays a role in the prognosis of pediatric germ cell tumors with the extragonadal pelvic tumors being the worst regarding resectability. Good tumor response can be achieved with surgery and chemotherapy even for advanced stage and metastatic disease.

Patients and Methods: We retrospectively analyzed the cases of pediatric germ cell tumors at National Cancer Institute over an 8 years period. Data concerning diagnosis, surgery and medical decisions were reviewed and analyzed for all patients. A total of 34 children with (GCTS) were found, with a mean age, at presentation, of 6.7 years and a follow-up period ranging from 3-52 months. One patient with benign GCT was excluded during analysis of the results. Results: Among the 34 patients, there were 14 males and 20 females with mean age of 6.7 years (range: 9 months-15 years), with male to female ratio 1:1.4. All patients were symptomatic at presentation, most commonly with abdominal swelling (18 patients; 52.9%). Anatomic distribution of GCTs according to sex organ involvement was either gonadal in 21 patients (61.8%) or extragonadal in 13 patients (38.2%).

Key Words: Germ cell tumors – Childhood solid tumors – management – Surgery.

INTRODUCTION Gonadal and extragonadal germ cell tumors are infrequent in children representing approximately 3% of cancers diagnosed in persons younger than 15 years [1,2]. Germ cell tumors are presumed to share a common cell of origin, the primordial germ cell, yet they remain a heterogeneous group of tumors. Variations regarding age, sites of presentation, histopathology and malignant potential stem from the differences in the stage of germ cell development at tumorigenesis, the differences in the tumor environment secondary to the gender of the patient and to the location of the clone and the occurrence of specific genetic aberrations [3]. Sacrococcygeal teratomas are the most common germ cell tumors of childhood, accounting for 40% of all and up to 78% of extragonadal germ cell tumors. They also are the most frequently recognized neoplasm of fetuses, with approxi-

All patients had surgery either in the form of curative resection or biopsy after formal exploration and evidence of irresectability. No significant surgical morbidity or mortality were encountered in our patients. Yolk sac tumor and malignant teratoma were the commonest histologic subtypes in our series. Metastatic disease was encountered in nine out of 33 patients (27.2%). Adjuvant chemotherapy was administered in 28 out of 33 patients (84.8%), following surgery, including all patients with extragonadal disease. Our patients were followed-up to 52 months. Twentytwo patients (66.7%) had no recurrence while two patients (6.1%) died from disease. Pelvic extragonadal site was the worst site regarding resectability. Complete surgical resection showed better disease free survival, while those Correspondence: Ibrahim Fakhr, MD, MRCS [email protected]

70

The Role of Surgical Management in Pediatric Germ Cell Tumors (GCTs)

mately 75% of patients are female [4]. Gonadal tumors come next, followed by the rest of the extragonadal sites, namely: mediastinal, retroperitoneal, brain and multiple other rare sites such as neck, oral cavity and stomach [5]. Germ cell tumors show numerous histologic subtypes. The histologic features of each subtype are independent of presenting clinical characteristics; tumor biology and clinical behavior vary with site of origin, stage and age of the patient. For example, mature teratoma in the infants and in the ovary are almost invariably diploid and benign, whereas those in the adult testis that show the same histologic features are aneuploid and potentially malignant [6]. Clinically, ovarian tumors account for only 1% of childhood malignancies with peak of age around 19 years. They present with an abdominal pain in 80% of patients (most of the latter have associated ovarian torsion) followed by abdominal mass or vaginal bleeding. On the other hand pediatric testicular tumors account for 2% of solid malignant neoplasm in boys with almost all identified as irregular non tender scrotal masses [7]. The outcome for patients with malignant GCTs was poor before the advent of systemic chemotherapy, with 3-year survival rates of approximately 20% [8,9]. The introduction of cisplatin-based therapy improved the outcome for patients with localized tumors, but patients with advanced disease continued to have a very poor prognosis [10-13]. Surgical resection is the therapy of choice in benign tumors, such as teratomas. With malignant lesions surgical resection is indicated if possible. However, given the availability of effective chemotherapy, resection should not be undertaken to the point of sacrificing vital structures. In such situation only debulking or biopsy is appropriate. After initial chemotherapy second-look surgery with intent of complete resection serves to assist in achieving complete response in selected patients [14].

71

A total of thirty four patients were found with a follow-up period ranging between 10 and 52 months. Data was retrieved from NCI hospital-based registry. Demographic and tumorspecific data was reviewed and analyzed. It included clinical data examination (age, symptoms and signs at diagnosis) and tumor markers. Diagnostic imaging evaluation included chest radiograph, ultrasonography and trunk computed tomography, or magnetic resonance imaging to assess the anatomical site of the disease and as part of the metastatic work up. The type of chemotherapy delivered to patients differed with time and stage of the disease at diagnosis: In low risk disease (stage I testicular and ovarian tumors), no chemotherapy was given with postoperative observation only. Patients with moderate risk gonadal tumors or progression of disease in untreated tumors were managed adequately with three to four cycles of a platinum containing regimen. For higher risk patients (higher stage testicular or ovarian tumors and extragonadal tumors), four to six cycles of a platinum based chemotherapeutic regimen were indicated. The type of chemotherapy delivered to patients with favorable tumors varied with time and included one of these protocols: cisplatin (platinol), vinblastine and bleomycin (PVB); cisplatin, etoposide and bleomycin (PEB) and carboplatin, etoposide and bleomycin (JEB). Pediatric patients with unfavorable tumors received chemotherapy combination etoposide, ifosfamide, carboplatin or cisplatin. We classified chemotherapy as optimal if its delivery was scheduled each 3 weeks. Statistical method: All patients were evaluated for the following response criteria: Event-free survival (EFS) and overall survival (OS) at 3 years. The EFS was calculated from the date of diagnosis to the date of first relapse or the date of last free followup. The OS was calculated from the day of initial diagnosis until death or last date of news. Survival curves were estimated by the KaplanMeier method.

PATIENTS AND METHODS

RESULTS

This retrospective study included all pediatric patients with malignant germ cell tumors (GCTs) recorded at the National Cancer Institute, Cairo University from January 2000 to December 2007.

Demographic features: A total of 34 children diagnosed with germ cell tumors were enrolled in the present study. Their mean age at presentation was 6.7 years (range: 9 months-15 years). Among the 34 en-

72

Ibrahim M.Y. Fakhr, et al.

rolled patients; there was a female predominance with a male to female ratio of 1:1.4 with 14 (41.2%) males and 20 (58.8%) females. Males had a mean age at presentation of 5.9 years (range 9 months-13 years) versus a mean age of 7.8 years (range 1-15 years) in females. Difference was statically not significant (p value=0.30). Clinical manifestation at presentation: The most common presentation was abdominal swelling and/or abdominal pain which occurred in 21 patients. Less common clinical manifestations included: testicular swelling in 9 patients, respiratory symptoms (cough and dyspnea) in 4 patients. The least common clinical presentations were: proptosis, a visible lump or mass at the buttocks and vaginal bleeding found in a single patient each. None of our patients presented with manifestations of hormone overproduction or precocious puberty. For the whole group, the median lag time from first complaint to accurate diagnosis of our patients was 4 months (range 1-11 months). Table (1) shows patients clinical presentation (some patients presented with more than one symptom). Diagnosis and initial management: Diagnosis was confirmed after full preoperative evaluation and work up including imaging techniques which were used according to the suspected site either gonadal or extra gonadal, with 21 patients (61.8%) with gonadal involvement and only 13 patients (38.2%) at the extragonadal sites. Table (2) shows the anatomic distribution of GCTs in all groups. The most common site was the ovary in 12 patients (35%), followed by the pelvic in 10 patients (29.2%), then the testis in 9 patients (26%), mediastinal (Picture 1), sacrococcygeal and orbit in a single patient each (3%). Table (1): Clinical manifestations of pediatric germ cell tumors at presentation. Symptom

No.

(%)

Abdominal swelling/pain Testicular swelling Cough and dyspnea Proptosis Mass in the buttocks Vaginal bleeding

21 9 4 1 1 1

61.7 26.5 11.7 2.9 2.9 2.9

Table (2): Anatomic distribution of GCTs in all patients. Males

Site

No.

(%)

Females No.

(%)

Total No.

(%)

Gonadal: Ovary Testis

9 0 9

(26.4) 12 (35.4) 21 (61.8) (0.0) 12 (35.4) 12 (35.4) (26.4) 0 (0.0) 9 (26.4)

Extragonadal: Pelvic Sacrococcygeal Mediastinal Orbit

5 4 0 1 0

(14.6) (11.6) (0.0) (3.0) (0.0)

Total

14 (41.0) 20 (59.0) 34 (100.0)

8 6 1 0 1

(23.6) (17.6) (3.0) (0.0) (3.0)

13 10 1 1 1

(38.2) (29.2) (3.0) (3.0) (3.0)

The tumor markers, AFP and HCG, were measured in all cases prior, following surgery and during the whole period of follow-up. Histopathology: The most common histologic subtypes were yolk sac tumor and malignant teratoma, which were diagnosed in 13 patients (38.2%) each, followed by mature teratoma, dysgerminoma and embryonal carcinoma in two patients (5.9%) each. Lastly, benign mature cystic teratoma and mixed GCTs in only one patient (2.9%) each. All GCTs were malignant except for the one patient with benign mature teratoma (2.9%) (Table 3). Staging and surgical management: All patients, in our study, had surgery either in the form of curative complete resection (24 patients) or a biopsy after formal exploration with evidence of irresectability (9 patients). The outcomes of the different surgical procedures were listed in Table (4). Different surgical procedures were done based on the site of origin of GCTs being either gonadal or extragonadal and according to the gender of patients. Gonadal tumors: Twenty patients were included (one female patient with benign teratoma who had an ovariectomy was excluded from our statistical analysis). Eight male patients with testicular tumors underwent radical unilateral orchiectomy with high ligation and en bloc excision of the spermatic cord structures and testicle; alone in 4 patients (20.0%) and with additional procedures in the other 4 patients (20.0%). Hemi-scrotalectomy was added in 2 patients (10.0%) for suspected invasion. A paraaortic lymph node dissection (PALND) was performed in one patient (5.0%) with para-

The Role of Surgical Management in Pediatric Germ Cell Tumors (GCTs)

aortic lymph node enlargement. Lastly, a single patient (5.0%) had excision of brain metastases in addition to orchiectomy for solitary brain metastases. Only one patient (5.0%) had irresectable retroperitoneal nodal disease and orchiectomy was done as a biopsy.

73

my, a single (7.7%) patient with mediastinal mass who needed a lower lobectomy and a last patient who had an eye enucleation (7.7%). Eight patients (61.5%) had irrresectable lesions. Six patients (46.1%) with pelvic lesions among whom one patient needed a colostomy, one patient (7.9%) with a lesion at the sacrococcygeal area and lastly another patient with a pelvic lesion of no definite origin (7.9%). No significant surgical morbidity or mortality was encountered in our patients. Only one of the 13 patients (7.7%) with the extragonadal tumors had para-aortic lymph node metastases.

Regarding the 11 female patients, five patients (25.0%) were staged as FIGO Stage I and they all had ovariectomy. Two patients (10.0%) had additional appendectomies, for grossly abnormal appendices. Three patients (15.0%) had an additional infracolic omentectomy and one patient had a superadded formal panhystrectomy with the infracolic omentectomy (5.0%). Only one patient with ovarian GCTs had peritoneal metastases and ascites (5.0%). None of the 11 female patients had LN metastases.

Different factors that might be related to resectability were analyzed. Only extragonadal origin in the pelvis site was statistically significant for irresectability; as tumors of extragonadal origin in the pelvis proved to be less resectable (8/9 patients, 61.5%; 7/9 patients, 70%, respectively) than gonadal origin and nonpelvic tumor site at the other sites (p