Surgical Intervention

The emergence of effective chemotherapy for testicular GCT has changed the management of intracranial GCT; a review of the evolution of therapy is necessary to understand current trends. Prior to the 1970s, the attempted extirpation of a posterior III ventricular or pineal tumor carried a 25-70 per cent operative mortality risk. This naturally emphasized EBRT as the treatment of choice [21]. Takeuchi et al. [35] advocated exploiting the radiosensitivity of pineal germinomas as a diagnostic as well as therapeutic procedure. In Japan, it became common practice to first irradiate a pineal tumor thought to be a germinoma with 20 Gy ("the radiation test''), and then, if the tumor regressed, to continue EBRT. If there was no reduction in tumor size, surgical excision was then to be considered [35,36].

Chapman and Linggood [37] showed that early radiation response was not in itself diagnostic of tumor histology or curability. The postirradiation recurrence rate of large germinomas led Sano and Matsutani [38] to advocate direct surgery and postoperative EBRT for all such patients, with the exception of those with small or multiple germinomas. Over time, advances in radiographic imaging, neuroanesthesia, and microsurgical technique produced an operative mortality rate of less than 5 per cent among the patients with a pineal tumor. In fact, the extent of surgical resection has been proven to be an important determinant of survival in the single largest series of 153 patients with primary intracranial GCT [23]. Several recent retrospective studies have emphasized the importance of macroscopic total resection at the time of diagnosis, in addition to EBRT and chemotherapy, in order to achieve durable remissions among the NG-GCT patients with the poorest prognosis [39-41]. Another compelling justification for establishing a histological diagnosis was the identification of patients who could benefit from more intensive therapy specifically directed against the NG-GCT (vide infra).

The pendulum has swung back again in Asia. Due to the increasing effectiveness of nonsurgical therapies, Sawamura et al. [42] found that the attempted radical resection of primary intracranial germinomas offered no further survival advantage. Among the Japanese and Korean adolescents, the incidence of GCT among pineal region tumors has been so high that tissue diagnosis is no longer felt to be obligatory if the neuroradiographic findings and biomarker serologic studies favor this diagnosis. Japanese Society for Pediatric Neurosurgery and the Korean Society for Pediatric Neurosurgery have emphasized the radio- and chemo-sensitivity of pineal region GCT. Jointly these groups have published a position paper, which advocates minimally invasive surgical procedures, such as stereotactic biopsy, to be followed by platinum-based chemotherapy and/or targeted EBRT [36]. It is only among the cases demonstrating a poor response to ''trial therapy'' that surgical intervention is to be considered to clarify the pathologic diagnosis and/or to debulk the tumor in preparation for further treatment [17].

Neuroendoscopic procedures are increasingly explored as a means to select cases requiring a microsurgical approach, such as medium to large size NG-GCT, from cases better treated with EBRT and/or chemotherapy. Hydrocephalus is more often being controlled through the use of III ventricular fenestration to avoid the complications of ventriculoperitoneal shunting [43,44].

External Beam Radiotherapy

In the older literature, conventional EBRT achieved 5Y-OS rates of 60 per cent (range 25-88 per cent) among a heterogeneous group of pineal tumors, which were being empirically treated [45-47]. Higher radiation dosages (50-55 Gy) have reduced the local recurrence rate from 47 to 10 per cent [48]. Among the patients with known germinomas, combined surgical resection and radiotherapeutic intervention improved 10Y-0S rates from 69 to 93 per cent [23,38]. Radiotherapy prescriptions of a primary tumor dose of 50-55 Gy, with or without 18-36 Gy to the neuraxis, have been proven to be reliably effective in sustaining complete remissions; EBRT remains the standard of care for germinoma patients in a number of Western series [49-52]. Some investigators find that germinomas are not only curable with irradiation but advocate that cranio-spinal doses of < 25.5 Gy to the whole brain and < 22 Gy to the spine are comparable to higher doses of combined EBRT-chemotherapy [53].

Subsequent controversy regarding the treatment of ''pure'' germinomas, without known metastatic disease, has centered on the issues of dosimetry, treatment portals (involved field versus ventricular) and the indication for craniospinal irradiation. As experience accrued to prove that germinomas of the CNS were as radiosensitive as those of the testis, there has been increasing consensus that those patients without evident CSF dissemination could be controlled with involved-field EBRT alone [23,50,54-56]. Investigators at Kyoto University have prospectively studied the relationship between postoperative residual tumor volume and radiotherapy prescription among 35 patients with germinomas. Subjects with no evident disease were irradiated to 36 Gy, for those with less than 2.5 cm diameter residual disease the dose was 40 Gy. Among patients with a tumor diameter of 2.5-4 cm, doses of 45 Gy were administered, and therapy to > 50 Gy was reserved for dimensions larger than 4 cm. The decision regarding involved-field versus craniospinal treatment (20-24 Gy) was determined by the staging evaluation at diagnosis. The 10- year relapse-free survival has been 95 per cent with a lOY-OS of 91 per cent. Two suffered meningeal dissemination but there were no local recurrences [57].

Stereotactic radiotherapy has been suggested as a means of precisely administering the tumor boost following involved field, whole brain or craniospinal EBRT. Median doses of 25.2 Gy (range 15-36) to the whole brain and 21.6 Gy (range of 21-26) to the neuraxis were supplemented by a median boost of 26 Gy (range 21.6-36) delivered to the 95 per cent isodose line. At a median follow-up time of 40 months, there were no local or marginal recurrences among 13 germinoma patients. Two of the five subjects with mixed GCT experienced either relapse or inadequate control despite the addition of chemotherapy. Stereotactic radiotherapy was well tolerated in this group of children and is thought to have reduced long-term neurotoxicity [58].

There is no debate regarding the necessity of treating NG-GCT with EBRT to the tumor (50-55 Gy) and craniospinal axis (36-40 Gy) due to their significantly higher rate of metastasis and recurrence [12,17,54,59-61]. Unfortunately, the NG-GCTs, as a group, have shown limited sensitivity to radiotherapy doses of more than 50 Gy [62].

Multimodality Therapy including Chemotherapy

Among localized and metastatic testicular GCTs, investigational protocols have combined synergisti-cally cytotoxic agents, including vinblastine, actino-mycin D, bleomycin, adriamycin, cyclophosphamide, and/or cisplatin (the "VAB'' regimens), following surgery and irradiation to produce durable remission rates of 60-90 per cent. Such agents can be delivered systemically in sufficient concentration to cross the blood-brain barrier and significantly lengthen survival time in cases of GCT metastases to the brain [63]. This experience initially encouraged the use of chemotherapy among primary intracranial GCTs at relapse, and then adjunctively during the 1980s [64,65].

Several principles have emerged in the pharmaco-logic treatment of gonadal GCT:

(1) Cisplatin-based combination chemotherapy has significantly enhanced the survival rates among patients with bulky or metastatic disease,

(2) The dose-response relationship of cisplatin among GCT is such that intensive induction chemotherapy may obviate the need for maintenance treatment among patients who demonstrate a complete response (CR),

(3) Carboplatin has comparable efficacy to cisplatin, with less oto- and nephro-toxicity and may not elicit cross-resistance among patients previously treated with cisplatin, and

(4) Vinblastine has been largely replaced by etopo-side for synergism with the platinator, while the necessity for bleomycin remains uncertain.

To enhance the response to EBRT, investigators at Memorial Sloan-Kettering [66] used three cycles of high-dose cyclophosphamide or vinblastin-bleomycin-cyclophosphamide-cisplatin as induction therapy among 11 children with primary intracranial GCT. Patients with disease restricted to the primary site were treated with 30 Gy to the involved field. Children with disseminated disease received 30 Gy to the primary site and 20 Gy to the neuraxis. Of the original group, 10 remained disease-free for four years from diagnosis (Table 33.1) [66].

A second phase II trial by this group administered two courses of carboplatinum to 11 patients with primary CNS germinomas without known metastatic disease. Children achieving a CR were treated with reduced dose EBRT (30 Gy to the involved field with 21 Gy to the neuraxis). Those patients with lesser responses went on to two additional courses of chemotherapy followed by full-dose irradiation (50 Gy to the involved field with 36 Gy to the neuraxis). Five subjects (45 per cent) demonstrated a CR after two courses, an additional two (64 per cent) after the fourth course. Two patients experienced a partial response (PR; > 50 per cent tumor reduction) after two courses and one after a total of four cycles of chemotherapy. Ninety-one per cent of the participants maintained complete remission for a median of twenty-five months. There was one relapse and death in a child who had responded to the two courses of induction chemotherapy. His serum a-FP and y8-HCG were found to be elevated at that time suggesting the emergence of resistant NG-GCT elements [67].

Not all studies have segregated the germinoma and NG-GCT patients. A Japanese study prescribed two to three courses of cisplatin-etoposide prior to irradiation. Twelve patients were treated—five at initial diagnosis and seven at recurrence. Of the five patients with elevated biomarkers felt to indicate the diagnosis of a NG-GCT, all demonstrated normalization of these levels prior to EBRT. A CR was noted in seven, while the remaining five patients demonstrated a PR. One patient with recurrent disease progressed and died despite further treatment; there was one death from sepsis. The remaining ten patients were free of disease at a mean of thirteen months from diagnosis [68].

The University of Eppendorf group has brought the attention to the risk of iatrogenic dissemination following surgical biopsy and/or resection among biomarker-positive GCT of the pineal recess. These investigators have proposed using one course of bleomycin-etoposide-cisplatin to induce significant regression and better delineation of the tumor's margins. An infratentorial, supracerebellar surgical approach is then used for attempted gross total resection. This is to be followed by vinblastine-ifosfamide-cisplatin and craniospinal EBRT (50 Gy tumor boost with 30 Gy to the neuraxis). This protocol was used for the treatment of three boys, two with elevated ¿8-HCG and one with a-FP expression. Each of the children received surgical resection in toto. No one required a permanent ventriculoperitoneal shunt. Following induction therapy, pathologic analysis revealed benign, mature teratoma with derivatives from all three germinal layers. There was also hemor-rhagic necrosis with granulomatous and lymphocytic inflammation. In one case, there were small admixtures of immature teratoma with mitoses. None demonstrated malignant extraembryonal differentiation. The boys remained disease-free at 66, 71, and 78 months following surgery [69,70].

A series of 18 patients with known (14 cases) and presumed NG-GCT (four with elevated biomarkers) were treated with three to four courses of cisplatin-etoposide followed by EBRT. Of the twelve patients with evaluable disease, five demonstrated a CR and four patients exhibited PR, two patients had stable disease and one progressed. Patients were irradiated either to the involved field (eleven), craniospinal axis (four), or whole brain (two). Twelve subjects received an additional four cycles of postradiation chemotherapy with vinblastine-bleomycin-etoposide-carboplatin. The six patients with no evident postoperative disease remained in remission following additional chemotherapy. Four patients had died by the time of reporting, three were disease related. The 4Y-EFS and 4Y-OS rates were 67 and 74 per cent, respectively [71].

The French Society of Pediatric Oncology reported 29 patients with biopsy proven, localized germinomas. Induction chemotherapy with carboplatin-etoposide-ifosfamide was administered for two courses prior to irradiation of the initial tumor volume. Of the 26 evaluable patients, 58 per cent achieved a CR. Twenty-eight patients remained in their first full remission over a median follow-up period of 32 months (range 7-68 months). The 4Y-EFS was 93 per cent without any fatalities [72].

The Cooperative German/Italian Study accepted the diagnostic entry criteria to be the appropriate neuroradiographic findings and elevated biomarkers (y8-HCG and/or a-FP). Nineteen patients (16 males and 3 females) were placed into the study. The therapeutic design consisted of two induction courses of cisplatinum-etoposide-ifosfamide (PEI); patients responding to chemotherapy were to receive an additional two courses. Patients who did not respond and those with progressive disease were to be advanced to surgical resection, if feasible, prior to craniospinal EBRT (30 Gy with a tumor boost of 24 Gy). This cohort demonstrated an elevated a-FP and/or ^-HCG level in 16/19 patients at the time of diagnosis. Thirteen of the sixteen had normalization of biomarker levels following the second course of PEI induction chemotherapy, which paralleled objective neuroradiographic evidence of a cytoreductive effect in 10/13. Three children demonstrated a CR after two courses of chemotherapy. Eight demonstrated no evident disease following the fourth course, with two children showing lesser responses. Three patients with teratomas demonstrated progressive disease for which tumor resection was attempted; one child died postoperatively due to intratumoral hemorrhage. Seventeen of the patients survived; 81 per cent have remained in remission over a median follow-up of 11 months (range 7-39 months). The toxicity of therapy was reported to be tolerable [73].

The practice at the University of Tokyo has been to rank GCT patients by relative risk into three subgroups:

(1) mixed germinoma and teratoma,

(2) mixed GCT with predominance of germinoma or teratoma with some ''pure malignant tumor'' (embryonal carcinomas, endodermal sinus tumors. and choriocarcinomas), and

(3) mixed tumors with predominance of ''pure malignant tumor''.

Surgery and EBRT produced a 10Y-OS rate of 91.7 per cent among the germinoma patients. Combination chemotherapy (cisplatin-vinblastine-bleomycin, cisplatin-etoposide, or carboplatin-etoposide) and radiotherapy was shown to significantly reduce the risk of disease recurrence in the intermediate prognostic group when compared to irradiation alone (p = 0.049). Forty per cent of these patients experienced a CR following induction chemotherapy as well as a much lower recurrence rate (21.4 versus 45 per cent). The high-risk patients did better with chemotherapy (3Y-OS rate of 27.3 per cent) than with EBRT alone (3Y-OS of 10.2 per cent) although the difference did not reach statistical significance [23].

The Japanese Pediatric Brain Tumor Study Group has expanded upon these observations. Germinoma patients were treated with carboplatin-etoposide or cisplatin-etoposide for three courses and followed with EBRT. The NG-GCT subjects received ifosfamide-cisplatin-etoposide for three courses after which they underwent irradiation. Among those with germinomas, 84 per cent achieved a CR with induction chemotherapy. Over a median follow-up period of 2.9 years, recurrent disease developed in 12 per cent. In the majority of these, relapse occurred outside the radiation portal. Among the 10 patients with y8-HCG secreting germinomas, 78 per cent achieved a CR with chemotherapy. Following EBRT, 90 per cent were disease-free. Over a monitoring period of 3.4 years, none have recurred. The intermediate prognosis group consisted of 18 patients with malignant teratoma and mixed tumors. There were no CRs noted with chemotherapy. However, following the completion of radiotherapy, 56 per cent were tumor-free. Two patients relapsed during a median observation period of 3.7 years. The poor prognosis group was made up of nine patients, of whom two were still alive without recurrence more than two years after the treatment [74].

The Societe Francaise d'Oncologie Pediatrique (SFOP) initiated a study of induction chemotherapy with either limited-field irradiation (40 Gy) for germinoma patients with localized disease (51 children) or "low dose'' craniospinal EBRT for those with dissemination (six patients). Four alternating courses of etoposide-carboplatin and etoposide-ifosfamide were administered postoperatively. With a median follow-up of 42 months, the 3Y-EFS was 96.4 per cent and the 3Y-OS was 98 per cent. Of the four patients who relapsed, three achieved a second CR with chemotherapy with or without irradiation [75].

The investigators at Kumamoto University have proposed treating patients with NG-GCT with neoadjuvant chemotherapy and EBRT to sufficiently cytoreduce the neoplasm as to allow gross total resection. In a population of 11 such patients (five with yolk sac tumor, one with embryonal carcinoma, one immature teratoma, and four subjects with mixed malignant GCT), induction therapy produced two CR and six PR. Two patients demonstrated stable disease and one progressed. Nine patients underwent surgical removal of residual disease. Ten of the eleven are alive at a mean of 96 months (range 30-177) following the diagnosis [76].

The Hokkaido University protocol is to combine induction chemotherapy with ''low-dose'' involved-field EBRT for newly diagnosed GCT patients. Solitary, ''pure'' germinomas are initially treated with etoposide-cisplatin. Ifosfamide-cisplatin-etoposide (ICE) is used for the induction of those patients with NG-GCT, y8-HCG secreting germinomas and for those with disseminated disease. Patients with pineal germinomas are treated with attempted gross total resection while others were biopsied or partially removed. The clinical treatment volume for EBRT includes the tumor site for germinomas and immature teratomas. Multifocal germinomas received irradiation (24 Gy) to the ''whole ventricle'', which includes the III and lateral ventricles, for pineal region tumors the IV ventricle is also treated. The ^-HCG secreting germinoma patients are administered an additional 6 Gy to the neurohypophysis and 10 Gy to the pineal region. Highly malignant and disseminated tumors are treated with craniospinal EBRT with a tumor boost (50-54 Gy). This group reported their results with 16 germinoma patients, 11 with ¿8-HCG secreting germinomas, three with immature teratoma/germinoma, and three patients with either embryonal carcinoma or malignant teratoma. Eight patients had multifocal origin and three suffered from metastatic disease. Every germinoma patient achieved a CR by the third course of chemotherapy; the presence of ^-HCG did not affect the response to therapy. For the entire group of 33 patients, 5Y-OS was 93 per cent. Relapse-free survival rates at 5 years were 69 per cent for the whole population, 90 per cent for the germinoma patients, and 44 per cent for those with ^-HCG secreting germinomas (the difference between the last two being p = 0.025). Patients with relapsed disease did respond to further therapy. All the six patients with NG-GCT were alive at a median of 65 months (range of 23-92 months). There were no treatment or disease-related deaths. Furthermore, no cognitive deterioration was observed using Wechsler Intelligence Scale testing for children and adults, with the exception of one patient who died from hypothalamic dysfunction (Table 33.1) [30].

Chemotherapy Alone

A Taiwanese study evaluated 11 newly diagnosed cases of primary intracranial germinoma who were treated only with six courses of vinblastine-bleomycin-cisplatin-etoposide (VBPE). Every patient achieved a CR, however 55 per cent relapsed at a

TABLE 33.1 Complete Response Rates to Induction Chemotherapy among Newly Diagnosed Germinomas and NonGerminomatous Germ Cell Tumor Patients



VBL-BLM-CPM-cDDP (ll pts) CBDCA (ll pts) 1st IGCTS

CBDCA-VP16-BLM (45 pts) SFOP

VP16-CBDCA ! VP16-IFOS (29 pts) SFOP

VP16-CBDCA ! VP16-IFOS (57 pts) Hokkaido

VPl6-cDDP (16 pts) (ßHCG (+) germinoma (17 pts) VBL-BLM-VPl6-cDDP (ll pts) JPBTSG

CBDCA-VP16 or cDDP-VPl6 (75 pts) jß-HCG(+) germinoma (10 pts) 2ud IGCST


CBDCA-VP16-BLM (19 pts)

Course 2 Course 4 Course 6 EBRT 2Y-EFS 3Y-EFS 4Y-EFS 5Y-EFS Reference (%) (%) (%) (%) (%) (%) (%)

0 0

Post a comment