ORIGINAL ARTICLE

Spontaneous Regression of Testicular Germ Cell Tumors

An Analysis of 42 Cases
Bonnie L. Balzer, MD, PhD* and Thomas M. Ulbright, MDw

Abstract: Spontaneous regression of testicular germ cell tumors

(GCTs) is a well-recognized phenomenon but has been
incompletely characterized. Many pathologists are not familiar
with the ndings that support a diagnosis of a burnt-out
primary in a patient with metastatic GCT. We therefore report
the clinical, gross, and histologic ndings in 42 cases of testicular
GCT that showed either complete (26) or greater than 50%
scarring (16). Thirty-seven patients (88%) had either known
GCT metastasis or some residual testicular GCT, and none had
treatment before orchiectomy. The patients were 17 to 67 years
old, with a median of 32. Thirty presented with symptoms of
metastasis, 7 with a testicular mass, 2 with elevated human
chronic gonadotropin, and 1 with testicular pain. In 2 patients
the presentation was unknown. Two patients had prior
orchiopexy; another had an intraabdominal testis, and 2 others
had prior contralateral seminoma (20 and 42 years previously).
Gross descriptions in 37 cases identied white to tan scars, 0.6 to
2.4 cm, in 33. These were circumscribed in 16, with 15 of these
having nodular or multinodular congurations and 1 a bandlike appearance. In 9 cases the scar was ill dened or stellate,
and in 8 cases no further details concerning the scar conguration were available. In 4 cases no scar was apparent; 2 of these
had received intraoperative biopsy. Microscopically, all cases
showed circumscribed to irregular foci of scarring, distinct from
the adjacent parenchyma, in association with widespread
testicular atrophy. Other common features were lymphoplasmacytic inltrates in the scars (37/42) and ghost tubules in
scars (31/42). Less common features in the scars included
angiomatous foci (22/42), siderophages (15/42), and coarse
intratubular calcications (6/42); in the surrounding testis they
included intratubular germ cell neoplasia, unclassied (IGCNU)
(22/42), Leydig cell prominence (18/42), and necrosis (5/42).
Tubular microliths occurred in 13 cases, 12 peripheral to the scar
and 1 within it. Metastases in 31 cases were: pure seminoma (17,
3 with residual testicular seminoma), mixed GCT with
seminoma (4, 3 with residual testicular seminoma), mixed

From the *Department of Pathology, Stanford University Hospital and

Clinics, Stanford, CA; and wDepartment of Pathology, Indiana
University School of Medicine, Indianapolis, IN.
Presented in part at the 93rd annual meeting of the United States and
Canadian Academy of Pathology, Washington, DC, March 28,
2004.
Reprints: Thomas M. Ulbright, MD, Department of Pathology, Indiana
University Hospital, Room 3465, 550 North University Boulevard,
Indianapolis, IN 46202 (e-mail: tulbrigh@iupui.edu).
Copyright r 2006 by Lippincott Williams & Wilkins

858

nonseminomatous GCT (4, 3 with residual testicular GCT),

pure embryonal carcinoma (2), pure teratoma (2, 1 with residual
testicular teratoma), and pure yolk sac tumor (2). In 5 cases with
clinically diagnosed metastases, there was no histologic documentation of the nature of the metastatic tumor. Testicular
tumors in the remaining 6 cases having residual primaries
without concomitant metastases were pure seminoma (3), mixed
GCT with seminoma (2), and pure embryonal carcinoma (1).
The most specic histologic ndings of a regressed GCT are a
distinct scar in association with either IGCNU or coarse
intratubular calcications; however, many cases lack the latter
2 features. In such cases additional features supportive of
regressed GCT include testicular atrophy, microlithiasis and, in
the scar, lymphoplasmacytic inltrates and prominent vascularity. Ghost tubules in many scars are not evidence of a nonneoplastic process but likely reect regression of tumors with
intertubular growth. Intertubular growth is a common nding in
seminoma, which is the single most frequent type of regressed
GCT, occurring either in pure or mixed form in the metastases
of 68% (21/31) of the cases and identiable in 62% (10/16) of
persistent testicular tumors. We conclude that regression of
testicular GCTs shows a distinctive constellation of ndings that
usually permits its recognition. In contrast, nonspecic atrophy
lacks distinct scars, and scars from non-neoplastic causes lack
most of the associated ndings seen in our cases.
Key Words: testicular germ cell tumor regression, burnt-out
testicular germ cell tumor, intratubular germ cell neoplasia,
unclassied, metastatic testicular germ cell tumor, testicular scar
(Am J Surg Pathol 2006;30:858865)

eports from the early portion of the 20th century

suggested that testicular germ cell tumors (GCTs)
may undergo spontaneous regression. These documented
retroperitoneal GCTs in the absence of a known testicular
primary. Although some were considered to be primary
retroperitoneal neoplasms,1,29,37,40 others were felt to
represent retroperitoneal metastases from occult testicular tumors.20,22,39 Subsequently, evidence for spontaneous
GCT regression was rmly established by Azzopardi and
coworkers,3,4 who examined autopsy material from
patients who died of widely metastatic GCT but lacked
testicular masses. Histologic examination of the testes
revealed well-delineated brous scars and variable
persistence of seminomatous and nonseminomatous
residual testicular tumors. Other occasional ndings were
Am J Surg Pathol

Volume 30, Number 7, July 2006

Am J Surg Pathol

Volume 30, Number 7, July 2006

hematoxylin staining bodies (described as coarse,

purple, calcied material located within seminiferous
tubules), ghost neoplastic cells, siderophages, and residual
intratubular tumor. Similar ndings were subsequently
reported in other studies.7,33,36
Current chemotherapy is largely eective against all
types of metastatic, nonteratomatous testicular GCT,19,25
but often not for the primary testicular tumor itself,
perhaps because of a blood-testis barrier.21 Since the
testis is considered a sanctuary or protected anatomic
site, surgical excision is indicated for all such cases.28,42,43
Thus, if a patient with metastatic GCT and a regressed
testicular primary undergoes orchiectomy, it is crucial for
the pathologist to establish that the resected testis is or is
not the primary source of the GCT. If a rm diagnosis is
not established, the contralateral testis falls under
suspicion, and may be unnecessarily excised. Alternatively, if an unjustied diagnosis of regressed GCT is
made for the resected testis, whereas the contralateral
testis is the true primary site, it may give rise to a
recurrence. Although the phenomenon of testicular GCT
regression has been well documented, no single study has
systematically addressed the question of which histologic
features or combination of features constitute sucient
evidence of a regressed testicular GCT. We therefore
report the histologic ndings in 42 cases of regressed
testicular GCT, with an emphasis on features supporting
a diagnosis of spontaneous GCT regression.

MATERIALS AND METHODS

A search of the Indiana University Medical Center
surgical pathology archives was performed for cases
coded as regressed GCT, testicular scar, and GCT, or
testicular brosis and GCT. These cases were retrieved
and screened in context with the medical charts documenting clinical presentation, prior chemotherapy, or
metastasis of a GCT. Cases in which the patient had
received treatment before orchiectomy were excluded.
The 42 remaining cases form the basis for this study. This
study was approved by the Institutional Review Board of
Indiana University Purdue University at Indianapolis.
From 1 to 9 slides of each case, with an average of 3
slides of the orchiectomy specimen, and 1 slide of the
metastasis, when available, were reviewed independently
by the authors. Histologic subtyping of GCTs was
performed according to the 2004 World Health Organization classication system for testis GCTs and the AFIP
fascicle.45,48 Tumors were classied as pure if only 1
histologic subtype was present and mixed if more than 1
histologic subtype was identied in either the residual or
the metastasis. The following histologic features were
tabulated for each orchiectomy specimen: the presence of
a distinct testicular scar; testicular atrophy peripheral to
the scar; the histologic subtype of any residual GCT,
either invasive or intratubular; foci of necrosis; Leydig
cell prominence, dened as nodular clusters of Leydig
cells larger than the transverse diameter of 1 histologically
normal seminiferous tubule; and intratubular calcicar

2006 Lippincott Williams & Wilkins

Spontaneous Regression of Testicular Germ Cell Tumors

tions, either coarse or microlith type. The latter were

recognized as round, psammoma bodylike calcications
within seminiferous tubules that did not exceed the
diameter of a normal tubule. Histologic features unique
to the scarred portion were also catalogued and included:
calcications, ghost tubules, dened as eosinophilic
remnant outlines of tubules lacking cytologic detail;
inammatory inltrate; siderophages; and vascularity.
The histologic subtype of metastatic tumor was recorded
for each case in which metastasis was documented.
Additional pertinent clinical, ultrasonographic, and gross
information was obtained from the medical charts and
pathology reports.

RESULTS
Clinical Findings
Patient age ranged from 17 to 67 years, with a
median of 32. The majority of patients (30) presented with
symptoms related to metastases. Of these, 20 had retroperitoneal metastases, and 2 of these patients had widely
disseminated tumor, 1 with lung and liver metastases, and
1 with mediastinal metastases. The ipsilateral groin lymph
nodes were the exclusive sites of metastases in 2 other
patients. The other metastatic sites included thyroid,
neck, and thoracic cavity. Of the remaining patients, 7
presented with a testicular mass, 2 with elevated serum bsubunit of human chorionic gonadotropin (hCG) and 1
with testicular pain. Two patients had an unknown
presentation. Two patients had prior orchiopexy; another
had an intraabdominal testis, and 2 others had prior
contralateral seminoma (20 and 42 years previously).

Ultrasonographic Findings
Ultrasound ndings were available in 7 cases. In 4
there was a discrete nodule (3) or a solid mass (1). The
other 3 had ill-dened inhomogeneity, but no mass or
features of ischemia were identied. In 1 of these the
ultrasonographic ndings were interpreted as suggestive
of an inammatory lesion, which prompted initial
treatment with antibiotics.

Gross Findings
Gross descriptions were available in 37 orchiectomy
specimens. In 21 cases the primary site was the left testis,
and in the remaining 16 the right testis. There were
grossly apparent scars, ranging from 0.6 to 2.4 cm, in 33
cases. White to tan, well-delineated nodular or multinodular foci were described in 15 cases, only 4 of which
also showed grossly visible residual tumor. The lesion in 1
other case was described as well demarcated but band
like. Nine cases had an ill-dened white to tan lesion, 2
specically described as scarlike and 1 as stellate.
Three of these 9 also showed gross residual tumor. No
details concerning the scarred foci were available in 8
cases. In 2 cases no gross abnormalities were appreciated.
Two other specimens were obtained after intraoperative
biopsy, which compromised the nal gross examination,
but no scar was described.

859

Balzer and Ulbright

Am J Surg Pathol

Volume 30, Number 7, July 2006

Microscopic Findings
A well-demarcated, nodular scar in the testicular
parenchyma (Fig. 1) was found in half the cases. Of these
90% were single nodules, with the remainder showing a
multinodular pattern of scarring. In the other half of the
cases, the scars showed varying proportions of nodular
and stellate congurations, the latter characterized by
irregular dissection of brous tissue between seminiferous
tubules at the periphery (Fig. 2). In all cases, there was
associated testicular atrophy, as demonstrated by shrunken seminiferous tubules with decreased or absent
spermatogenesis, and thickened peritubular basement
membranes peripheral to the scar. Many tubules had a
Sertoli cell only pattern, and others were completely
hyalinized (Fig. 3). A lymphoplasmacytic inltrate, which
varied in intensity from scant to moderately dense, was
identied within the scar in 37 of 42 cases (Fig. 4). Ghost
tubules, consisting of densely collagenized remnants of
seminiferous tubules, were present within the hyalinized
scars in 31 cases (Fig. 5). Increased vascularity, manifested by collections of small, round to arcuate, nonarborizing vessels, was also identied in the scar of 22
cases (Fig. 4). Siderophages were identied within the scar
in 15 cases (Fig. 5). Coarse intratubular, irregularly
shaped dystrophic calcications were identied in 6 cases,
all within the scar (Fig. 6).
Histologic ndings peripheral to the area of scarring
included intratubular germ cell neoplasia of the unclassi-

FIGURE 1. A well-demarcated, nodular scar with large

aggregates of Leydig cells in the surrounding, atrophic testis.

860

FIGURE 2. An irregular scar with dissection of fibrous tissue

between seminiferous tubules.

ed type (IGCNU) in 22 specimens (Fig. 7). (Marked

tubular sclerosis in many of the cases likely contributed to
the absence of identiable IGCNU.) Prominent clusters
of Leydig cells were identied in the atrophic testis
peripheral to the scar in 18 cases (Fig. 1), and foci of
coagulative tumor cell necrosis in 5 cases.
Microlith type calcications were seen both within
and outside the scar. These consisted of small, round,
psammoma bodylike structures within the tubules in 13
cases (Fig. 3), 12 outside and 1 within the scar. In 1 case
both microliths and coarse intratubular calcications
were seen within the scar. The frequencies of the dierent
features of testicular GCT regression are shown in graph
form (Fig. 8).
The histologic subtypes of residual and metastatic
GCT are summarized in Table 1 and shown in graph form
(Fig. 9). Residual tumor in the testis consisted of pure
seminoma (6 cases), mixed GCT with seminoma (5 cases),
mixed GCT without seminoma (3 cases), and 1 case each
of pure embryonal carcinoma and pure teratoma.
Metastases in 31 cases were: pure seminoma in 17 cases
(including 3 with residual testicular seminoma), mixed
GCT with seminoma in 4 cases (including 3 with residual
testicular seminoma) and mixed or pure GCT lacking
seminoma in 10 cases. In 5 cases with clinically diagnosed
metastases, there was no histologic documentation of the
nature of the metastatic tumor. Ten cases had both
residual testicular and metastatic GCT, whereas 5
additional cases had only scarred testis, lacking residual
r

2006 Lippincott Williams & Wilkins

Am J Surg Pathol

Volume 30, Number 7, July 2006

Spontaneous Regression of Testicular Germ Cell Tumors

FIGURE 3. Testicular atrophy with shrunken and hyalinized

tubules. One (center) contains a microlith.

FIGURE 5. Outlines of hyalinized, ghost tubules in the scar,

with scattered siderophages.

FIGURE 4. A lymphoplasmacytic infiltrate within the scarred

zone, which also contains numerous round and arcuate
vessels.

FIGURE 6. Coarse intratubular calcification within a scar.

2006 Lippincott Williams & Wilkins

861

Am J Surg Pathol

Balzer and Ulbright

Volume 30, Number 7, July 2006

TABLE 1. Histology of Primary and Metastatic GCTs

Histology

Only in
Metastasis

In Metastasis
and Testis

Only in
Testis

Totals

14
1
1
2
2
1
N/A
21

3
3
3
0
0
1
N/A
10

3
2
0
1
0
0
0
6

20
6
4
3
2
2
5*
42

Pure seminoma
Mixed w/seminoma
Mixed w/o seminoma
Pure EC
Pure YST
Pure teratoma
Unknown
Total(s)

*Five cases were known to have metastatic tumor by clinical and radiographic
evaluation but no specimen was available from the metastatic site.
EC indicates embryonal carcinoma; N/A, not applicable; YST, yolk sac
tumor.

with the remaining 6 cases showing mixed histology, 3

with seminomatous components.

DISCUSSION
The results of this study indicate that reproducible
histologic features consistently present in regression
include a well-delineated to irregular scar and peripherally located testicular atrophy, the latter manifest by
hyalinized, shrunken or Sertoli cell-only lined tubules.
The uniform association of regression with atrophy is
FIGURE 7. Intratubular germ cell neoplasia, unclassified type
in the seminiferous tubules peripheral to a scar.

Relative Distribution of Histologic Subtype

60

Distribution of Histologic Features

Scar

50

Percentage of cases

tumor in the testis, and not known to have metastatic

GCT. Among the 10 cases with both testicular and
metastatic tumor, 7 had disparities between the histologic
components at the 2 sites, whereas 3 cases showed pure
seminoma in both the primary and metastatic deposits. In
1 of the 7 cases showing histologic discrepancy, teratoma
was the only remaining component in the primary tumor,

40

30

20

10

Histologic features

Hyalinized tubules
Lymphocytes

Ghost tubules

n
"
ma
ow
ma
kn
ino
no
i
n
m
U
e
m
se
-se
n-s
CT
on
no
re"
G
n
+
u
"
T
M
P
C
re
Pu
MG
no
mi

Vascularized scar
IGCNU
Microliths
Coarse calcs
Leydig cell clusters

"

i
sem

no

ma

Histologic subtype

Siderophages
Necrosis

10

20

30

40

50

Number of cases
Partial regression (16)

Complete regression (26)

FIGURE 8. Relative distribution of histologic features in

completely and partially regressed GCTs.

862

ma

Study cases (42)

Regressed lit cases (97)
Non-regressed lit cases (1356)

FIGURE 9. Comparison of the distribution of histologic

subtypes in regressed GCT among study cases and literature
cases. Data for regressed GCTs are summarized from Refs.
24,6,10,11,1315,26,30,32,40,41 Data for nonregressed
GCTs are summarized from 2 large series, Refs. 27,46.
r

2006 Lippincott Williams & Wilkins

Am J Surg Pathol

Volume 30, Number 7, July 2006

hardly surprising, given that it is the atrophic testis rather

than the normal-sized testis that is at most increased risk
for the development of testicular GCTs.8,22 Microlithiasis, identied in 13 of our cases, is also a recognized risk
factor and, in conjunction with atrophy, reects a
background upon which GCTs commonly develop.5,12,38
In the presence of demonstrable metastases, the combination of a scarred area in an atrophic background, in our
opinion, suces as evidence of a regressed primary
testicular GCT. However, although scarring and atrophy
were found in all cases, they are nonspecic features that
may also occur secondary to ischemia, trauma, and
infection. Therefore, in the absence of known metastatic
GCT these features alone are suggestive but not
necessarily diagnostic of a regressed GCT.
Among the features analyzed in this study, IGCNU
(apart from residual invasive tumor) in a scarred testis is
the single most specic one for GCT regression. We
accept the presence of IGCNU in combination with
testicular scarring and atrophy as diagnostic evidence of
GCT regression, even in the absence of known metastatic
GCT. However, IGCNU was evident in only 48% of our
cases, so its absence cannot be used to exclude a regressed
GCT. An additional relatively specic but even less
sensitive nding for GCT regression are the coarse,
irregular intratubular calcications that correspond to
regressed intratubular embryonal carcinoma; these were
seen in only 12% of our cases. It is important, because of
the diagnostic signicance of this nding, not to
misinterpret the smaller, round, often psammoma bodylike calcications within atrophic tubules that characterize microliths for the same phenomenon. Although it is
clear that such microliths are associated with GCTs and,
as evidenced by our study, regressed GCTs, they may also
be seen in testes that lack any tumor.
In those cases of regression that lack IGCNU,
coarse intratubular calcications and known metastasis,
some additional features apart from scarring and atrophy
may permit a greater degree of condence that what is
represented in the testis corresponds to a regressed GCT.
Numerous small vessels within the scarred focus are
common in regressed GCTs and likely reect prior tumor
angiogenesis, with persistence of the neoformed vessels
after tumor regression. Induction of blood vessel formation by testicular GCTs has been previously demonstrated.34 Prominent clusters of Leydig cells in a minority
of cases is another helpful feature in supporting a
diagnosis of regressed GCT. Leydig cell hyperplasia is
known to occur in response to elevated concentrations of
luteinizing hormone or, because of its luteinizing hormonelike activity, hCG. Thus, hyperplasia of Leydig cells is
common in nonregressed testicular GCTs with syncytiotrophoblast cells because of their hCG production.49 In
cases of GCT regression, Leydig cell hyperplasia therefore
likely reects, at least in part, hCG production by the
tumor before regression. It is also true that in testes with
marked tubular atrophy and sclerosis Leydig cells may
form prominent clusters secondary to parenchymal
shrinkage and therefore appear hyperplastic. Since testes
r

2006 Lippincott Williams & Wilkins

Spontaneous Regression of Testicular Germ Cell Tumors

showing regressed GCT are invariably also atrophic, in

these instances it is dicult if not impossible to separate
true hyperplasia secondary to tumor eects from pseudohyperplasia secondary to atrophy. The histologic
ndings may in fact reect a combination of both. A
variable lymphoplasmacytic inltrate occurred within the
scar of the great majority of cases and is another
supporting feature for GCT regression. It suggests that
an immune reaction may play a role in tumor regression.
This idea receives some support from the observation that
the outcome in seminoma has a positive correlation with
the density of the associated lymphocytic inltrate.17,35
Our experience with 42 cases indicates that regression of testicular GCTs is most frequently associated with
seminoma, which was identiable in 26 of 42 cases (62%)
within the testis and/or metastases. Among these 26 cases,
seminoma, either in pure form (14 cases) or mixed with
other GCT types (1 case), occurred only in the metastases
of 15 cases. In an additional 5 cases residual seminoma (3
pure, 2 mixed) was identied only in the testis; and in 6
other cases seminoma (3 pure, 3 mixed) occurred both in
metastatic deposits and in residual testicular tumor. Thus,
there were 20 cases (48%) where seminoma alone was
identied in either metastases or residual testicular tumor.
The relative proportion of seminoma as either a pure
GCT or in combination with other subtypes of GCT in
our series of regressed cases, therefore, essentially
paralleled its incidence among all GCTs in 2 large
combined studies of unselected testicular primary GCTs
(Fig. 9). Pure seminoma, thus, represents the greatest
number of cases of regressed GCTs, but this nding
reects the overall high frequency of seminoma among
testicular GCTs rather than a special proclivity of
seminoma relative to other GCTs to undergo spontaneous regression. Although Munro and coworkers33
also reported a high (77%) frequency of seminomas
among their 13 cases of complete and partial regression, it
is apparent that in many prior studies seminoma was
underrepresented (Fig. 9) for unclear reasons.
Although in this series, pure seminoma gave rise to
pure seminomatous metastases, histologic discrepancy
between the primary and metastatic testicular GCT
occurred in 7 of 10 other cases. It is therefore unlikely
that, in cases of regression, attempts to infer the histologic
subtype in the testis from the metastases will prove
accurate (and visa versa). It is important to realize that
scars in the testis with seminoma may correspond to a
regressed nonseminomatous component, making the
usual treatment with radiation problematic for those
cases consisting of scar with seminoma. Serum marker
studies may assist in some of these cases, permitting the
detection of a metastatic nonseminomatous component.
It was our preconception that teratoma would prove
more resistant to regression than nonteratomatous
elements, similar to the situation in postchemotherapy
resections of metastatic GCT where teratoma is disproportionately represented. This, however, was not borne
out by our study; pure teratoma represented only 1 of
16 (6%) cases with residual testicular GCT, a gure

863

Balzer and Ulbright

comparable to the cited gure of about 5% for the

frequency of pure teratoma among testicular GCTs.45
In contrast to seminoma, previous reports have
apparently overrepresented choriocarcinoma compared
to its overall frequency in nonselected testicular
GCTs,27,36 a conclusion supported by the absence of
any well-documented example of regressed choriocarcinoma in our study. This may, however, be an artifact of
our selection criteria since we excluded all cases where the
patient had received chemotherapy before orchiectomy,
and it is relatively common to have patients with
choriocarcinoma rst receive such treatment before
orchiectomy because they commonly have widespread
disease at presentation.
Although, as implied above, we think it is not a
realistic objective to attempt prediction of the subtype of
a regressed GCT from the remnant features in the testis,
there are 2 ndings that suggest its nature, namely ghost
tubules and intratubular coarse calcications. The presence of ghost tubules in many scars is likely secondary to
regression of intertubular tumor, which is a common
nding in seminoma24; we speculate that regression of
foci of intertubular seminoma is probably responsible for
many of the cases having densely collagenized outlines of
seminiferous tubules within scarred areas. The presence of
coarse, often irregularly-shaped, intratubular calcications, in our opinion, corresponds to regression of
intratubular embryonal carcinoma, as such foci are
commonly seen adjacent to cases of embryonal carcinoma
with an intratubular component. The calcications
represent a dystrophic phenomenon in the foci of
comedo-type necrosis that are seen in intratubular
embryonal carcinoma; they often exceed the diameter of
normal seminiferous tubules since they develop in tubules
that are expanded by intratubular tumor.
Clinically, the important key dierential diagnosis
in these cases is between regression of a GCT and a nonneoplastic cause of scarring, such as traumatic or
vasculopathic infarction, prior inguinal or testicular
surgery or previous infection. Obviously, important
information can be gained from the history, but in the
absence of a predisposing factor the clinical distinction
may not always be possible. It is well known that
testicular ischemia or infarction can clinically mimic a
testicular GCT.18,31 Increasingly, however, these lesions
are being diagnosed by ultrasonographic techniques,
often avoiding orchiectomy but also limiting the opportunity to compare their histologic features to those of a
regressed GCT.44 In our experience, such non-neoplastic
scars usually occur in testes that lack the diuse atrophy
seen in our cases, more often are multifocal, may be
associated with vascular lesions (thrombi, vasculitis), and
consistently lack the association with more specic
features of regression (IGCNU, coarse intratubular
calcications). If the features remain equivocal, the
patient deserves thorough evaluation by radiographic
methods and serum marker studies. If these prove
negative, ongoing follow-up is still indicated. Atrophy
with interstitial brosis should also be distinguished from

864

Am J Surg Pathol

Volume 30, Number 7, July 2006

a regressed GCT. All of our cases had distinct scars, both

nodular and stellate, that would not be expected in pure
atrophy.
Ultrasound is of diagnostic value in identifying
testicular abnormalities, and its use has proven of great
help in locating occult testicular GCTs in patients who
presented with metastatic disease.9,47 The experience in
our series, however, demonstrates a lack of specicity
regarding persistent tumor versus scar versus a nonneoplastic process; 2 of 4 cases with an ultrasonographically detected nodule, on pathologic examination, showed
only a scar without residual tumor. Conversely, 1 case
showing parenchymal inhomogeneity without a nodule
was interpreted ultrasonographically as probable infection but histologically showed a tumor nodule.
Our study also brings home another point. Some of
the scars were quite small and could have easily been
overlooked, yet the patients had metastatic GCT. This
implies that specimens with features associated with GCT
regression should be examined thoroughly for scarred
areas. Thorough sampling of a testis with IGCNU is
indicated to nd inconspicuous foci of scarring, with its
implication for regression of an invasive tumor. If such
sampling is negative but the patient has rm clinical
evidence of a GCT, the possibility of a regressed tumor in
the opposite testis or a true extragonadal primary merits
clinical consideration. Reasons for this recommendation
include that the testis opposite a GCT has IGCNU in
approximately 5% of cases,16 and that patients with
extragonadal GCTs have about a 4% frequency of having
testicular GCT.23
REFERENCES
1. Allen E, Vespignani PM. Active testicular epithelium in the
connective tissue surrounding a human suprarenal gland. Anat
Record. 1938;72:293296.
2. Asif S, Uehling DT. Microscopic tumor foci in testes. J Urol.
1968;99:776779.
3. Azzopardi JG, Hobrand AV. Retrogression in testicular seminoma
with viable metastases. J Clin Pathol. 1965;18:135141.
4. Azzopardi JG, Mosto FK, Theiss EA. Lesions of testes observed in
certain patients with widespread choriocarcinoma and related
tumors. Am J Pathol. 1961;38:207225.
5. Bach AM, Hann LE, Hadar O, et al. Testicular microlithiasis: what
is its association with testicular cancer? Radiology. 2001;220:7075.
6. Bar W, Hedinger C. Comparison of histologic types of primary
testicular germ cell tumors with their metastases: consequences for
the WHO and the British Nomenclatures? Virchows Arch [A].
1976;370:4154.
7. Bar W, Hedinger C. Ausgebrannte (okkulte) Hodentumoren:
Hodenlasionen bei klinisch scheinbar primar extratesticularen
malignen Keimzelltumoren. Virchows Arch [A]. 1977;377:6778.
8. Berthelsen JG, Skakkebaek NE, von der Maase H, et al. Screening
for carcinoma in situ of the contralateral testis in patients with
germinal testicular cancer. BMJ. 1982;285:16831686.
9. Bockrath JM, Schaeer AJ, Kies MS, et al. Ultrasound identication of impalpable testicle tumor. J Urol. 1983;130:355356.
10. Bohle A, Studer UE, Sonntag RW, et al. Primary or secondary
extragonadal germ cell tumors. J Urol. 1986;135:939943.
11. Burt ME, Javadpour N. Germ-cell tumors in patients with
apparently normal testes. Cancer. 1981;47:19111915.
12. Cast JE, Nelson WM, Early AS, et al. Testicular microlithiasis:
prevalence and tumor risk in a population referred for scrotal
sonography. AJR Am J Roentgenol. 2000;175:17031706.
r

2006 Lippincott Williams & Wilkins

Am J Surg Pathol

Volume 30, Number 7, July 2006

13. Comiter CV, Renshaw AA, Benson CB, et al. Burned-out primary
testicular cancer: sonographic and pathological characteristics. J
Urol. 1996;156:8588.
14. Crook JC. Morphogenesis of testicular tumours. J Clin Pathol.
1968;21:7174.
15. Daugaard G, von der Maase H, Olsen J, et al. Carcinoma-in-situ
testis in patients with assumed extragonadal germ-cell tumours.
Lancet. 1987;2:528530.
16. Dieckmann KP, Loy V. The value of the biopsy of the contralateral
testis in patients with testicular germ cell cancer: the recent German
experience. APMIS. 1998;106:1320.
17. Dixon FJ, Moore RA. Tumors of the Male Sex Organs. Atlas of
Tumor Pathology, 1st series, Fascicles 31b and 32. Washington, DC:
Armed Forces Institute of Pathology; 1952:59.
18. Doebler RW, Norbut AM. Localized testicular infarction masquerading as a testicular neoplasm. Urology. 1999;54:366.
19. Einhorn LH. Chemotherapeutic and surgical strategies for germ cell
tumors. Chest Surg Clin N Am. 2002;12:695706.
20. Friedman NB. The comparative morphogenesis of extragenital and
gonadal teratoid tumors. Cancer. 1951;4:265276.
21. Greist A, Einhorn LH, Williams SD, et al. Pathologic ndings at
orchiectomy following chemotherapy for disseminated testicular
cancer. J Clin Oncol. 1984;2:10251027.
22. Haines JS, Grabstald H. Tumor formation in atrophic testes. Arch
Surg. 1950;60:857860.
23. Hartmann JT, Fossa SD, Nichols CR, et al. Incidence of
metachronous testicular cancer in patients with extragonadal germ
cell tumors. J Natl Cancer Inst. 2001;93:17331738.
24. Henley JD, Young RH, Wade CL, et al. Seminomas with exclusive
intertubular growth: a report of 12 clinically and grossly inconspicuous tumors. Am J Surg Pathol. 2004;28:11631168.
25. Hinton S, Catalano PJ, Einhorn LH, et al. Cisplatin, etoposide and
either bleomycin or ifosfamide in the treatment of disseminated
germ cell tumors: nal analysis of an intergroup trial. Cancer.
2003;97:18691875.
26. Holmes AS, Klimberg IW, Stonesifer KJ, et al. Spontaneous
regression of testicular seminoma: case report. J Urol. 1986;135:
795796.
27. Jacobsen GK, Barlebo H, Olsen J, et al. Testicular germ cell
tumours in Denmark 1976-1980. Pathology of 1058 consecutive
cases. Acta Radiol Oncol. 1984;23:239247.
28. James PP, Mead GM. Sanctuary site relapse in chemotherapytreated testicular cancer. Ann Oncol. 1992;3:4143.
29. Kahle PJ. Retroperitoneal seminoma (germinoma): report of a case:
histogenetic and diagnostic considerations. Am Surg. 1954;20:
538548.
30. Lopez JI, Angulo JC. Burned-out tumour of the testis presenting as
retroperitoneal choriocarcinoma. Int Urol Nephrol. 1994;26:
549553.
31. McCabe J, Das S, Hamid B, et al. Localized traumatic infarction of
the testicle. Scand J Urol Nephrol. 2004;38:442443.
32. Meares EM Jr, Briggs EM. Occult seminoma of the testis
masquerading as primary extragonadal germinal neoplasm. Cancer.
1972;30:300306.

2006 Lippincott Williams & Wilkins

Spontaneous Regression of Testicular Germ Cell Tumors

33. Munro AJ, Duncan W, Webb JN. Extragonadal presentations of

germ cell tumours. Br J Urol. 1983;55:547554.
34. Olivarez D, Ulbright T, de Riese W, et al. Neovascularization in
clinical stage A testicular germ cell tumor: prediction of metastatic
disease. Cancer Res. 1994;54:28002802.
35. Parker C, Milosevic M, Panzarella T, et al. The prognostic
signicance of the tumour inltrating lymphocyte count in stage I
testicular seminoma managed by surveillance. Eur J Cancer. 2002;
38:20142019.
36. Powell S, Hendry WF, Peckham MJ. Occult germ-cell testicular
tumours. Br J Urol. 1983;55:440444.
37. Prym P. Spontanheilung eines bosartigen, wahrscheinlich chorionepitheliomatoses Gewachses im Hoden. Virchows Arch A Pathol Anat
Histol. 1927;265:239.
38. Rashid HH, Cos LR, Weinberg E, et al. Testicular microlithiasis: a
review and its association with testicular cancer. Urol Oncol.
2004;22:285289.
39. Rather LJ, Gardiner WR, Frerichs JB. Regression and maturation
of primary testicular tumors with progression of growth of
metastases: report of 6 new cases and review of the literature.
Stanford Med Bull. 1954;12:1225.
40. Rottino A, DeBellis H. Extragenital chorioma: its relation to
teratoid vestiges in the testicles. Arch Pathol. 1944;37:7880.
41. Scholz M, Zehender M, Thalmann GN, et al. Extragonadal
retroperitoneal germ cell tumor: evidence of origin in the testis.
Ann Oncol. 2002;13:121124.
42. Simmonds PD, Mead GM, Lee AH, et al. Orchiectomy after
chemotherapy in patients with metastatic testicular cancer. Is it
indicated? Cancer. 1995;75:10181025.
43. Snow BW, Rowland RG, Donohue JP, et al. Review of delayed
orchiectomy in patients with disseminated testis tumors. J Urol.
1983;129:522523.
44. Sriprisad S, Kooiman GG, Muir GH, et al. Acute segmental
testicular infarction: dierentiation from tumour using high
frequency colour Doppler ultrasound. Br J Radiol. 2001;74:
965967.
45. Ulbright TM, Amin MB, Young RH. Atlas of Tumor Pathology:
Tumors of the Testis, Adnexa, Spermatic Cord and Scrotum. 3rd
Series. Washington, DC: Armed Forces Institute of Pathology;
1999:41173.
46. von Hochstetter AR, Hedinger CE. The dierential diagnosis of
testicular germ cell tumors in theory and practice: a critical analysis
of two major systems of classication and review of 389 cases.
Virchows Arch [A]. 1982;396:247277.
47. Weaver DJ, Havey AD, Weinstein SH, et al. Nonpalpable occult
testis tumor. Urology. 1989;34:218220.
48. Woodward PJ, Mosto FK, Talerman A, et al. Germ cell tumours.
In: Eble JN, Sauter G, Epstein JI, et al, eds. Pathology and Genetics
of Tumours of the Urinary System and Male Genital Organs. Lyon:
IARC Press; 2004:221249.
49. Zimmerli UU, Hedinger CE. Hyperplasia and hypertrophy of
Leydig cells associated with testicular germ cell tumours containing
syncytiotrophoblastic giant cells. Virchows Arch A Pathol Anat
Histol. 1991;419:469474.

865