Management of Primary Malignant Germ Cell Tumor of the Mediastinum
Jpn J Clin Oncol
Management of Primary Malignant Germ Cell Tumor of the Mediastinum
Hiroyuki Sakurai 0 1
Hisao Asamura 1
Kenji Suzuki 1
Shun-ichi Watanabe 1
Ryosuke Tsuchiya 1
0 Second Department of Surgery, University of Yamanashi , Kitakoma-gun, Yamanashi , Japan
1 Division of Thoracic Surgery, National Cancer Center Hospital , Tokyo
Background: Primary mediastinal malignant germ cell tumors (GCTs) are rare and have a worse prognosis than their gonadal counterparts. Although multimodality treatment is a standard therapeutic strategy in mediastinal GCTs, the clinical implications of surgical intervention remain unclear. Methods: Forty-eight patients with primary mediastinal malignant GCT who were treated at the National Cancer Center Hospital, Tokyo, from 1962 to 2002 were studied retrospectively with regard to their histology and clinical profile. Results: Mediastinal GCT occurred predominantly in young males, with a mean age of 28.8 years at the time of diagnosis. There were 46 males (96%) and two females (4%). Histologically, seven patients (15%) were diagnosed as having pure seminoma and 41 (85%) had nonseminomatous GCT. Treatment consisted of surgery alone in nine patients, surgery followed by chemotherapy in two, and chemotherapy followed by surgery in 20. The other 17 patients received chemotherapy and/or radiotherapy without surgery. Of these latter 17 patients, 14 developed progressive disease and three were followed up with a sustained partial response. Among the 31 patients who underwent surgery, complete resection was performed in 27 (87%) and incomplete resection was performed in four (13%). Twelve (41%) patients had elevated serum tumor marker levels preoperatively. Among the 20 patients who received preoperative chemotherapy, viable cells were found in the resected specimen in six (30%). With regard to tumor recurrence in patients with surgical intervention, the preoperative serum tumor marker levels and the presence of viable cells in the resected specimen were significantly associated with recurrence. There was no significant association between surgical curability and recurrence. The 5-year overall survival rate in all 48 patients was 45.5%. Conclusions: Surgical intervention for mediastinal GCT may be needed to remove a chemotherapy-refractory tumor or to assess the pathological response to chemotherapy to determine the indications for further chemotherapy.
mediastinum - mediastinal neoplasms - germ cell tumor - tumor markers
Although the most common location for malignant germ cell
tumors (GCTs) is the gonads, they can also arise in
extragonadal regions such as the mediastinum, retroperitoneum and
pineal gland (
). It has been speculated that they occur in such
unusual locations due to the abnormal migration of germ cells
during embryogenesis (
). The histologic characteristics of
extragonadal tumors are similar to those of gonadal GCTs, and
both are chemosensitive. The mediastinum is the most
common site of primary extragonadal GCTs (4). Primary malignant
GCTs of the mediastinum account for 1–6% of all mediastinal
). Primary extragonadal GCTs, especially primary
mediastinal tumors, are considered to have a poor prognosis
). Although they have similar histologic features,
mediastinal GCTs are clinically and biologically distinct from their
Malignant GCTs are closely related to serum tumor markers,
especially alpha-fetoprotein (AFP) and/or the beta subunit of
human chorionic gonadotropin (HCG). Measurement of these
serum tumor markers is important in diagnosis of the disease,
and in the management and follow-up of patients with GCTs.
Chemotherapy for malignant GCTs has become standard
practice since the introduction of cisplatin-based combination
chemotherapy in the late 1970s (
chemotherapy with peripheral blood stem cell transplantation
(PBSCT) has also been used for refractory tumors (
Patients with these tumors have not usually been considered as
candidates for primary surgery because of the presumed
systemic nature of the disease. Thus, all resections for GCTs have
been performed as ‘salvage surgery’, which is considered to be
the resection of residual tumor after a partial response to
induction chemotherapy or an operation at the time of relapse after a
complete response to chemotherapy (23). Although salvage
surgery is performed with the intent to remove
chemotherapyresistant disease, to assess the pathological response to
chemotherapy in the resected specimen and to guide additional
chemotherapy, the clinical significance and indications of salvage
surgery in the treatment of mediastinal GCTs remain unclear.
Several issues must be addressed in the management of
malignant GCT: preoperative normalized serum tumor marker
levels, the relationship between prognosis and the presence of
persistent viable cells in post-chemotherapy resected
specimens, the significance of the completeness of resection and the
operative risk, especially after high-dose chemotherapy with
PBSCT. Due to the rarity of GCTs arising in the mediastinum,
there have been few reports on the clinical outcome after
therapeutic challenge for this type of tumor in a large population.
This report describes our experience with managing patients
with mediastinal GCT.
PATIENTS AND METHODS
In the 41-year period from 1962 through 2002, 48 patients with
malignant GCT of the mediastinum were treated at the
National Cancer Center Hospital, Tokyo. Three patients were
referred to us after surgical resection of the tumor as the initial
treatment of choice at another institute. The
clinicopathological characteristics of these 48 patients were examined in this
retrospective study. The patients were considered to have
primary mediastinal GCT when a bulky anterior mediastinal mass
was present in the absence of any clinically detectable
testicular or ovarian masses at any time during the course of the
). The diagnosis of mediastinal GCT was defined
clinicopathologically in all patients. Serum tumor markers,
especially AFP and/or HCG, were measured preoperatively in
The treatments used in the 48 patients with mediastinal GCTs
are summarized in Fig. 1. All patients who received
chemotherapy initially or postoperatively were treated with a regimen
containing cisplatin. The chemotherapy regimen used in each
case was one of those sequentially developed throughout the
study period at our institute for the management of germ cell
tumors. Mainly, the patient was treated with either a
combination of cisplatin and etoposide (PE) or a combination of
cisplatin, bleomycin and etoposide (BEP), whether for
seminomatous or non-seminomatous GCT. The PE chemotherapy
consisted of intravenous cisplatin (120 mg/m2) and intravenous
etoposide (100 mg/m2) given on days 1–5. The BEP
chemotherapy consisted of intravenous cisplatin (20 mg/m2 of
bodysurface area) given daily for 5 days, intravenous bleomycin (30
U) given on days 2, 9 and 16, and intravenous etoposide (100
mg/m2) given on days 1–5. The patients received two to four
courses of the chemotherapy given at 3-week intervals. Of
these 48 patients, 23 received multimodality treatment while
25 received only one form of treatment, such as chemotherapy,
surgery or radiotherapy. Thirty-four patients received
chemotherapy, 11 received radiotherapy and 31 underwent surgery.
Three patients received high-dose combination chemotherapy
with PBSCT: one underwent high-dose chemotherapy
following the initial chemotherapy and surgical resection, and the
others underwent the initial high-dose chemotherapy followed
by surgical resection.
Based on previous reports (
), the response criteria
were defined as follows: complete remission (CR) was
recorded when serum tumor markers normalized and complete
resolution of tumor masses occurred. CR was also documented
when resection of a residual mass revealed only necrosis.
Partial remission (PR) was defined as a >50% decrease in
bidimensional tumor measurements and ≥50% decline in serum
tumor markers. A radiological increase in tumor dimensions or
an elevation of marker levels indicated progressive disease
Operative reports were reviewed and the curability of
resection (complete resection or incomplete resection) and outcome
(mortality and morbidity) were recorded. Complete resection
was defined as no macroscopic or microscopic residual tumor,
and incomplete resection was defined as evident macroscopic
or microscopic residual tumor. Operative death was defined as
any death within 30 days of the operation or during
hospitalization. Preoperative serum tumor markers, AFP and HCG,
were defined as normal or elevated on the basis of marker
values collected before the operation and approximately 4–6
weeks from the date of the last chemotherapy.
Post-chemotherapy resected specimens were categorized as
follows. Specimens that showed the histological appearance of
GCTs such as embryonal carcinoma, yolk sac carcinoma,
immature or mature teratoma, choriocarcinoma or seminoma
were regarded as having viable cells. Specimens that included
only necrosis, based on histologic findings of necrotic debris
without any viable cells, were regarded as having no viable
A chi-square test was used to compare the frequencies among
different groups. Survival curves were estimated by the
Kaplan–Meier method using the date when treatment began
as the starting point and the date of recurrence, death or last
follow-up as the endpoint. A P-value of <0.05 was considered
Malignant GCT of the mediastinum occurred predominantly in
young adults, with a mean age of 28.8 years at the time of
diagnosis (range, 13–68). There was a male preponderance: 46
(96%) males and two (4%) females. Thirty-six (75%) of the 48
patients showed a symptom that could be related to the tumor
at the initial examination. These symptoms were non-specific
and resulted from the expanding tumor encroaching on
surrounding structures (Table 1). The remaining 12 (25%) patients
were asymptomatic and the tumors were found based on
evidence of an anterior mediastinal mass on a routine chest
radiograph. Klinefelter syndrome was also found in a 13-year-old
patient with an immature teratoma.
DIAGNOSIS AND HISTOLOGY
The histologic diagnosis of these 48 malignant GCTs was
confirmed in resected specimens taken by thoracotomy (13
patients, 27%), incisional biopsy of the tumor (six patients,
13%), percutaneous needle biopsy (27 patients, 56%), or
autopsy (two patients, 4%). Although biopsy specimens were
taken in the two autopsy cases, they were not of diagnostic
significance. The histologic subtype could not be determined in
one patient because of total necrosis of the resected tumor after
cisplatin-based chemotherapy, and, therefore, no histologic
type could be specified. However, since the pre-chemotherapy
serum tumor markers were markedly elevated, a diagnosis of
non-seminomatous GCT was made. Seven patients (15%) were
diagnosed as having pure seminoma and 41 (85%) had
nonseminomatous GCT, containing non-seminomatous elements.
TREATMENT AND RESPONSE
Among these 48 patients, nine underwent surgical resection
alone. Two patients underwent surgical resection followed by
chemotherapy and 20 patients received chemotherapy followed
by surgical resection. Seventeen patients received
chemotherapy and/or radiotherapy without surgery. Of these latter
17 patients, 14 developed PD and three were followed up with
sustained PR. With regard to the initial treatment, 11 patients
underwent surgical resection and 37 received chemotherapy
and/ or radiotherapy. Among these latter 37, 14 (38%), all of
whom had non-seminomatous lesions, developed PD, and 20
(54%) were followed by surgery with PR. The remaining
three patients (8%) (one was diagnosed as having
nonseminomatous lesion and two were seminoma) were followed
up without surgical resection because of sustained good PR
(tumor measured 3 cm or less in diameter on CT).
For the 31 patients who underwent surgical resection,
complete resection was performed in 27 patients (87%) and
incomplete resection was performed in four (13%). Twelve (41%) of
the 29 patients whose serum tumor markers were measured had
elevated tumor markers prior to resection. Nine of these 12
patients had elevated AFP and five had elevated HCG. Both
tumor markers were elevated in only two patients. In the
remaining 17 patients (59%), serum tumor markers were
within the normal range.
Among the 20 patients who received preoperative
chemotherapy, six (30%) had viable cells in the resected specimen
and 14 (70%) had no viable cells.
PROGNOSIS The median follow-up period was 6.5 years. The 5-year relapse-free survival rate in 34 patients, excluding the 14
Chemo, conventional chemotherapy; Viable cells, viable cells present in the resected specimen; High-dose, high-dose
chemotherapy with PBSCT; NED, no evidence of disease.
patients who developed PD after initial chemotherapy, was
63.2% (Fig. 2). Among the 29 patients whose preoperative
serum tumor markers were measured, 17 had normal serum
tumor marker levels and two (12%) of these had recurrence.
The remaining 12 had elevated levels and nine (75%) of these
had recurrence. Thus, the prevalence of recurrence in patients
with preoperative elevated serum tumor markers was
significantly higher than that in patients with preoperative normal
serum tumor markers (P = 0.001). In the three patients with
preoperative elevated serum tumor markers who had no
recurrence (Table 2), the histological diagnosis was
non-seminomatous GCT. Two of these three patients received chemotherapy
followed by surgery and had no viable cells in their resected
specimens, and one of these two received postoperative
highdose chemotherapy with PBSCT because of pulmonary
disseminated disease on CT. These two patients both survived
relapse-free for more than 5 years. The other patient underwent
initial surgery followed by chemotherapy. This patient was lost
on the 285th postoperative day due to secondary malignancy
Among the 31 surgical resections, there were 27 complete
resections and four incomplete resections. Among the 27
complete resections, there were 10 cases (37%) with recurrence
and 17 (63%) without recurrence. On the other hand, there
were two cases (50%) with recurrence and two (50%) without
recurrence in the four incomplete resections. There was no
significant difference between curability and recurrence (P =
With regard to the relationship between recurrence and the
presence of viable cells in the resected specimen (Table 3),
among the 20 patients who underwent preoperative treatment
(chemotherapy and/or radiotherapy), six had viable cells
and 14 had no viable cells. Four (67%) of the six patients with
viable cells, all of whom had been diagnosed as having
nonseminoma, had recurrence. Two (14%) of the 14 patients (five
seminomas, nine non-seminomas) with no viable cells had
recurrence. There was a significant difference between
recurrence and the presence of viable cells in the resected specimen
(P = 0.018).
The 5-year overall survival rate in the 48 patients was 45.5%
(Fig. 3). Among the 23 deaths following treatment, 21 were
caused by either recurrence or progression of the disease
(tumor-related death). The remaining two patients were lost on
the 7th and 285th postoperative days because of multiple organ
failure due to disseminated intravascular coagulation (DIC)
and secondary malignancy (acute leukemia), respectively.
Surgical mortality and morbidity are shown in Table 4. Among the
31 patients who underwent surgical resection, there was one
(3%) operative death caused by DIC. Postoperative morbidity
was found in 13% of patients.
Although the general histologic and serologic characteristics of
mediastinal malignant GCTs are similar to those of testicular
GCTs, mediastinal GCTs have been reported to have a worse
prognosis than gonadal GCTs (
). Furthermore, the clear
differences in clinical behavior suggest that these tumors are
biologically distinct. Some investigators have proposed that
mediastinal GCTs arise from cells of a different embryologic
origin than testicular primaries (10). Others have suggested
that the poor outcome of treatment for mediastinal GCTs
reflects the advanced stage of the disease usually present at the
time of diagnosis and that there is little intrinsic difference
between mediastinal and gonadal GCTs with respect to their
response to chemotherapy (
). However, the reasons for the
observed differences in the clinical characteristics and
behavior of mediastinal GCT are unknown. In addition, some reports
have suggested that patients with Klinefelter syndrome have a
slightly increased predisposition of having mediastinal GCT
). One patient (2%) in the present study had Klinefelter
syndrome. The pathogenesis of the association between
mediastinal GCT and Klinefelter syndrome remains unclear.
With regard to the treatment of mediastinal malignant GCTs,
the current consensus is that initial systemic chemotherapy
should be followed by aggressive complete resection of all
macroscopic residual tumor when necessary (
As in most series, patients treated with this approach had a
superior outcome to those treated with initial resection, even
though these tended to be smaller tumors. Our study also ended
in disaster for the prognosis. Nine of the 11 patients who
underwent surgery as initial treatment had recurrence (distant
organ) within one year after surgery, three of whom received
adjuvant chemotherapy; the remaining two were survivors for
more than 10 years, one of whom received adjuvant
chemotherapy and the other, diagnosed as having immature teratoma,
underwent only surgery for treatment. However, the clinical
implications of surgical resection in the multimodality
treatment of mediastinal GCTs are not yet clear.
The management of patients with elevated serum tumor
marker levels after chemotherapy is controversial. The degree
of the elevation of serum tumor markers, especially AFP
and HCG, plays a crucial role in defining the outlook
). In the present study, patients with preoperative
normal tumor marker levels had significantly better survival
than those with preoperative elevated tumor marker levels.
Surgical resection for patients whose markers have not normalized
with chemotherapy is usually futile (
). Dulmet and
colleagues also stated that normalizing serum tumor marker
levels before surgical resection was a significant favorable
prognostic factor (
). On the other hand, Vuky and associates
found a tendency for shorter survival in patients with elevated
and increasing tumor markers compared to patients with
normal or elevated-but-declining markers before surgical
). In our series, two patients with chemo-refractory
marker-elevated GCTs achieved a prolonged relapse-free
status with salvage surgery and they had no viable cells in
the resected specimens. These two patients had elevated, but
markedly declining, marker levels before surgical resection. In
addition, both these two patients had no viable cells in the
resected specimens pathologically. These might be due to a
half-life of tumor markers and the possibility of a few viable
cells besides sections examined microscopically in spite of the
diagnosis of no viable cells. Thus, selected patients, such as
patients with elevated but declining marker levels before
surgery, might be considered as candidates for surgery (
The histopathologic findings of post-chemotherapy resected
specimens are important prognostically (
presence of persistent GCT in the resected specimen is a poor
prognostic factor. In the present study, four (67%) of six patients
with viable cells in the resected specimen had recurrent disease
despite the resection of persistent GCT, whereas two (14%) of
14 patients with no viable cells had recurrence. One definite
purpose of surgery is careful evaluation of the tumor after
chemotherapy to see whether or not any viable cells remain.
Further chemotherapy in the adjuvant setting may be
considered in patients who show viable tumor. On the other hand,
patients with residual tumor such as mature teratoma are
commonly found because of the histological heterogeneity of
GCTs. Residual teratomas must be completely removed, since
they can compress or invade surrounding mediastinal
structures, and dedifferentiation into carcinoma is possible (
In the present study, histology influenced the treatment
outcome. Patients with pure seminoma achieved a high
pathological CR rate and an excellent prognosis (5-year survival rate,
100%). Several reports have confirmed the good prognosis of
mediastinal pure seminoma compared with non-seminomatous
Chemo, conventional chemotherapy; High-dose, high-dose chemotherapy with peripheral blood stem cell transplantation;
Viable cells, viable cells present in the resected specimen; NED, no evidence of disease; WD, with disease.
). The Southeastern Cancer
Study Group reported excellent sustained remissions with
initial chemotherapy in all patients with mediastinal seminomas
(19). In our study, five patients with seminoma, who received
chemotherapy followed by surgery, all had CR without any
viable cells in the resected specimen. Motzer and colleagues
proposed that close observation without surgery is possible for
seminoma patients with a post-chemotherapy residual tumor of
3 cm or less in diameter because none of the patients with a
residual tumor of this size had viable cells in the resected
). In the present study, two patients with pure
seminoma also had sustained post-chemotherapy good PR (tumor
size, 3 cm or less in diameter on CT) without tumor relapse. On
the other hand, GCTs could have a mixed histology, with
both seminomatous and non-seminomatous components (
Accordingly, elevated serum tumor markers in patients with
GCT might indicate a non-seminomatous component even if
biopsied specimens reveal only pure seminoma. A diagnosis of
seminoma based on small biopsy specimens should be
considered clinically as well as histopathologically.
High-dose chemotherapy was introduced to improve the
outcome in patients with refractory GCTs, based on the fact that
the response rate of non-seminomatous GCT increases with the
intensification of chemotherapy (
). Some reports have
suggested that high-dose chemotherapy with PBSCT may have
curative potential in refractory GCTs (
). In the present
study, three patients were treated with high-dose carboplatin
and etoposide with PBSCT (Table 5). All three patients
underwent surgical intervention. Two of them received two or more
regimens of conventional chemotherapy prior to high-dose
chemotherapy followed by surgery, with viable cells in the
resected specimens. Ultimately, both died of relapse disease in
less than 2 years. The other patient received high-dose
chemotherapy for disseminated disease (lung disease) after
conventional chemotherapy followed by reduction surgery, and
remains free of disease at 74 months. Some investigators have
proposed that the use of high-dose chemotherapy with PBSCT
does not produce durable remission in patients with refractory
mediastinal GCT with a poor prognosis (
). The efficacy
of high-dose chemotherapy, in conjunction with surgical
resection when necessary, in the treatment of refractory mediastinal
GCTs must be further evaluated.
In conclusion, surgical resection for mediastinal GCT may
be required to remove chemotherapy-refractory tumor for local
control or assess the pathological response to chemotherapy.
Although whether a gross chemotherapy refractory tumor
includes pathologically viable cells or not is unknown, if there
are any, it might lead to a hotbed of local relapse, which can
life-threateningly compress mediastinal structures such as the
heart, aorta and superior vena cava. The presence of viable
cells in the resected specimen may suggest a necessity for
additional chemotherapy because of the presumed systemic nature
of the disease. On the other hand, patients with preoperative
elevated increasing tumor marker levels and/or progressive
disease after initial chemotherapy are no longer candidates
for surgery for local control because of their extremely poor
prognosis. However, the clinical implications and indications
of surgical resection must be investigated further.
Additionally, more appropriate and effective therapeutic strategies are
needed in the treatment of patients with mediastinal GCT to
achieve better survival rates.
1. Hainsworth JD , Greco FA . Extragonadal germ cell tumors and unrecognized germ cell tumors . Semin Oncol 1992 ; 19 : 119 - 27 .
2. Fine G , Smith RW , Pachter MR . Primary extragenital choriocarcinoma in male subject . Case report and review of the literature. Am J Med 1962 ; 32 : 776 - 94 .
3. Nichols CR , Saxman S , Williams SD , Loehrer PJ , Miller ME , Wright C , et al. Primary mediastinal nonseminomatous germ cell tumors. A modern single institution experience . Cancer 1990 ; 65 : 1641 - 6 .
4. Walsh GL , Taylor GD , Nesbitt JC , Amato RJ . Intensive chemotherapy and radical resections for primary nonseminomatous mediastinal germ cell tumors . Ann Thorac Surg 2000 ; 69 : 337 - 44 .
5. Wychulis AR , Payne WS , Clagget OT , Woolner LB . Surgical treatment of mediastinal tumors: a 40-year experience . J Thorac Cardiovasc Surg 1971 ; 62 : 379 - 92 .
6. Wright CD , Kesler KA , Nichols CR , Mahomed Y , Einhorn LH , Miller ME , et al. Primary mediastinal nonseminomatous germ cell tumors. Results of a multimodality approach . J Thorac Cardiovasc Surg 1990 ; 99 : 210 - 7 .
7. Dulmet EM , Macchiarini P , Suc B , Verley JM . Germ cell tumors of the mediastinum. A 30-year experience . Cancer 1993 ; 72 : 1894 - 901 .
8. Jyothirmayi R , Ramadas K , Jacob R , Gangadharan VP , Nair MK . Primary malignant germ cell tumours of the mediastinum. Results of multimodality treatment . Acta Oncol 1997 ; 36 : 317 - 21 .
9. Bajorin D , Katz A , Chan E , Geller N , Vogelzang N , Bosl GJ . Comparison of criteria for assigning germ cell tumor patients to 'good risk' and 'poor risk' studies . J Clin Oncol 1988 ; 6 : 786 - 92 .
10. Toner GC , Geller NL , Lin SY , Bosl GJ . Extragonadal and poor risk nonseminomatous germ cell tumors . Cancer 1991 ; 67 : 2049 - 57 .
11. Childs WJ , Goldstraw P , Nicholls JE , Dearnaley DP , Horwich A . Primary malignant mediastinal germ cell tumours: improved prognosis with platinum-based chemotherapy and surgery . Br J Cancer 1993 ; 67 : 1098 - 101 .
12. Bukowski RM , Wolf M , Kulander BG , Montie J , Crawford ED , Blumenstein B . Alternating combination chemotherapy in patients with extragonadal germ cell tumors. A South West Oncology Group Study . Cancer 1993 ; 71 : 2631 - 8 .
13. Gerl A , Clemm C , Lamerz R , Wilmanns W. Cisplatin-based chemotherapy of primary extragonadal germ cell tumors. A single institution experience . Cancer 1996 ; 77 : 526 - 32 .
14. Moran CA , Suster S , Koss MN . Primary germ cell tumors of the mediastinum . III. Yolk sac tumor, embryonal carcinoma, choriocarcinoma, and combined nonteratomatous germ cell tumors of the mediastinum: a clinicopathologic and immunohistochemical study of 64 cases . Cancer 1997 ; 80 : 699 - 707 .
15. International Germ Cell Cancer Collaborative Group. International germ cell consensus classification: a prognostic factor-based staging system for metastatic germ cell cancers . J Clin Oncol 1997 ; 15 : 594 - 603 .
16. Toner GC , Motzer RJ . Poor prognosis germ cell tumors: current status and future directions . Semin Oncol 1998 ; 25 : 194 - 202 .
17. Logothetis CJ , Samuels ML , Selig DE , Dexeus FH , Johnson DE , Swanson DA , et al. Chemotherapy of extragonadal germ cell tumors . J Clin Oncol 1985 ; 3 : 316 - 25 .
18. Williams SD , Birch R , Einhorn LH , Irwin L , Greco FA , Loehrer PJ . Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide . N Engl J Med 1987 ; 316 : 1435 - 40 .
19. Loehrer P , Birch R , Williams SD , Greco FA , Einhorn LH . Chemotherapy of metastatic seminoma: The Southeastern Cancer Study Group Experience . J Clin Oncol 1987 ; 8 : 1212 - 20 .
20. Nichols CR , Andersen J , Lazarus HM , Fisher H , Greer J , Stadtmauer EA , et al. High-dose carboplatin and etoposide with autologous bone marrow transplantation in refractory germ cell cancer: An Eastern Cooperative Oncology Group Protocol . J Clin Oncol 1992 ; 10 : 558 - 63 .
21. Motzer RJ , Mazumdar M , Gulati SC , Bajorin DF , Lyn P , Vlamis V , et al. Phase II trial of high-dose carboplatin and etoposide with autologous bone marrow transplantation in first-line therapy for patients with poor-risk germ cell tumors . J Natl Cancer Inst 1993 ; 85 : 1828 - 35 .
22. Motzer RJ , Mazumdar M , Bosl GJ , Bajorin DF , Amsterdam A , Vlamis V . High-dose carboplatin, etoposide, and cyclophosphamide for patients with refractory germ cell tumors: treatment results and prognostic factors for survival and toxicity . J Clin Oncol 1996 ; 14 : 1098 - 105 .
23. Shepherd FA , Ginsberg R , Patterson GA , Feld R , Goss PE , Pearson FG , et al. Is there ever a role for salvage operations in limited small-cell lung cancer? J Thorac Cardiovasc Surg 1991 ; 101 : 196 - 200 .
24. Asamura H , Tsuchiya R , Goya T , Kondo H , Naruke T , Saijo N , et al. Malignant germ cell tumor of the mediastinum: a multimodality therapeutic approach . Surg Today 1994 ; 24 : 137 - 141 .
25. Goss PE , Schwertfeger L , Blackstein ME , Iscoe NA , Ginsberg RJ , Simpson WJ , et al. Extragonadal germ cell tumors. A 14-year Toronto experience . Cancer 1994 ; 73 : 1971 - 9 .
26. Hainsworth JD , Einhorn LH , Williams SD , Stewart M , Greco FA . Advanced extragonadal germ cell tumors. Successful treatment with combination chemotherapy . Ann Intern Med 1982 ; 97 : 7 - 11 .
27. Nichols CR , Heerema NA , Palmer C , Loehrer PJ , Williams SD , Einhorn LH . Klinefelter's syndrome associated with mediastinal germ cell neoplasms . J Clin Oncol 1987 ; 5 : 1290 - 4 .
28. Hasle H , Jacobsen BB , Asschenfeldt P , Andersen K. Mediastinal germ cell tumour associated with Klinefelter syndrome. A report of case and review of the literature . Eur J Pediatr 1992 ; 151 : 735 - 9 .
29. Jain KK , Bosl GJ , Bains MS , Whitmore WF , Golbey RB . The treatment of extragonadal seminoma . J Clin Oncol 1984 ; 7 : 820 - 7 .
30. Ganjoo KN , Rieger KM , Kesler KA , Sharma M , Heilman DK , Einhorn LH . Results of modern therapy for patients with mediastinal nonseminomatous germ cell tumors . Cancer 2000 ; 88 : 1051 - 6 .
31. Takeda S , Miyoshi S , Ohta M , Minami M , Masaoka A , Matsuda H . Primary germ cell tumors in the mediastinum. A 50-year experience at a single Japanese institution . Cancer 2003 ; 97 : 367 - 6 .
32. Bosl GJ , Geller NL , Cirrincione C , Vogelzang NJ , Kennedy BJ , Whitmore WF , et al. Multivariate analysis of prognostic variables in patients with metastatic testicular cancer . Cancer Res 1983 ; 43 : 3403 - 7 .
33. Kesler KA , Rieger KM , Ganjoo KN , Sharma M , Fineberg NS , Einhorn LH , et al. Primary mediastinal nonseminomatous germ cell tumors: the influence of postchemotherapy pathology on long-term survival after surgery . J Thorac Cardiovasc Surg 1999 ; 118 : 692 - 701 .
34. Tait D , Peckham MJ , Hendry WF , Goldstraw P . Post-chemotherapy surgery in advanced non-seminomatous germ-cell testicular tumours: the significance of histology with particular reference to differentiated (mature) teratoma . Br J Cancer 1984 ; 50 : 601 - 9 .
35. Motzer RJ , Bosl GJ . Chemotherapy for germ cell tumors . Urol Clin North Am 1987 ; 14 : 389 - 98 .
36. Motzer RJ , Geller NL , Tan CC , Herr H , Morse M , Fair W , et al. Salvage chemotherapy for patients with germ cell tumors. The Memorial SloanKettering Cancer Center experience ( 1979 - 1989 ). Cancer 1991 ; 67 : 1305 - 10 .
37. Vuky J , Bains M , Bacik J , Higgins G , Bajorin DF , Mazumdar M , et al. Role of postchemotherapy adjunctive surgery in the management of patients with nonseminoma arising from the mediastinum . J Clin Oncol 2001 ; 19 : 682 - 8 .
38. Murphy BR , Breeden ES , Donohue JP , Messemer J , Walsh W , Roth BJ , et al. Surgical salvage of chemorefractory germ cell tumors . J Clin Oncol 1993 ; 11 : 324 - 9 .
39. Afifi HY , Bosl GJ , Burt ME . Mediastinal growing teratoma syndrome . Ann Thorac Surg 1997 ; 64 : 359 - 62 .
40. Economou JS , Trump DL , Holmes EC , Eggleston JE . Management of primary germ cell tumors of the mediastinum . J Thorac Cardiovasc Surg 1982 ; 83 : 643 - 9 .
41. Knapp RH , Hurt RD , Payne WS , Farrow GM , Lewis BD , Hahn RG , et al. Malignant germ cell tumors of the mediastinum . J Thorac Cardiovasc Surg 1985 ; 89 : 82 - 9 .
42. Moran CA , Suster S , Przygodzki RM , Koss MN . Primary germ cell tumors of the mediastinum . II. Mediastinal seminomas: a clinicopathologic and immunohistochemical study of 120 cases . Cancer 1997 ; 80 : 691 - 8 .
43. Motzer R , Bosl G , Heelan R , Fair W , Whitmore W , Sogani P , et al. Residual mass: an indication for further therapy in patients with advanced seminoma following systemic chemotherapy . J Clin Oncol 1987 ; 7 : 1064 - 70 .
44. Ozols RF , Ihde DC , Linehan WM , Jacob J , Ostchega Y , Young RC . A randomized trial of standard chemotherapy v a high-dose chemotherapy regimen in the treatment of poor prognosis nonseminomatous germ-cell tumors . J Clin Oncol 1988 ; 6 : 1031 - 40 .
45. Broun ER , Nichols CR , Einhorn LH , Tricot GJK . Salvage therapy with high-dose chemotherapy and autologous bone marrow support in the treatment of primary nonseminomatous mediastinal germ cell tumors . Cancer 1991 ; 68 : 1513 - 5 .
46. Saxman SB , Nichols CR , Einhorn LH . Salvage chemotherapy in patients with extragonadal nonseminomatous germ cell tumors: the Indiana University experience . J Clin Oncol 1994 ; 12 : 1390 - 3 .