Percutaneous Needle Biopsy Preceding Preoperative Chemotherapy in the Management of Massive Renal Tumors in Children By Ulla M. Saarinen, Sakari Wikstrim, Olli Koskimies, and Hannu Sariola While the National Wilms' Tumor Study (NWTS) Group in the United States puts an emphasis on accurate staging and histology before any therapy is given for Wilms' tumor, the International Society of Pediatric Oncology (SIOP) in Europe focuses on preoperative therapy and safer surgery. Our current approach combines the benefits of both policies in the management of massive renal tumors in children. In seven consecutive patients we first obtained a percutaneous posterior needle biopsy to obtain adequate tissue for histology, and proceeded with preoperative chemotherapy with vincristine and dactinomycin until tumor shrinkage was sufficient. Tumor removals were feasible and uneventful. At the time of operation, two tumors were found to be totally or almost totally
WILMS'
TUMOR is one of the pediatric
malignancies for which there has been the most success in responding to therapy; overall survival rates have risen to over 80% within the last 10-year period.' Advanced tumor stage, unfavorable histology, and especially both factors together still carry a worse prognosis, and improvement of therapy in these subgroups is needed.' Despite the uniformly good results, the policies on how to approach a patient with Wilms' tumor differ significantly between the two most prominent study groups, ie, the National Wilms' Tumor 2 Study (NWTS) group in the United States, and the International Society of Pediatric Oncology 3 (SIOP) group in Europe. The NWTS emphasizes accurate staging and an in-depth, detailed histology, which constitute the basis for choice of consequent therapy. Therefore, the NWTS stresses that operative removal of the tumor should be 2 performed as the first step, whenever possible.' On the other hand, SIOP wants to emphasize
From the Children'sHospital, University of Helsinki, Helsinki, Finland. Submitted April 11, 1990; accepted September 4, 1990. Address reprint requests to Ulla M. Saarinen, MD, Children's Hospital, University of Helsinki, Stenbiickinkatu 11, 00290 Helsinki, Finland. C 1991 by American Society of Clinical Oncology. 0732-183X/91/0903-0011$3.00/0
406
necrotic. In the others, which still included viable tumor, the histology corresponded well to the needle biopsy findings. One case with unfavorable histology and one with rhabdoid sarcoma would have been missed and given suboptimal therapy without the primary needle biopsy. As possible biopsy-related complications, subcapsular intratumoral bleeding was recognized in two patients. We conclude that percutaneous posterior needle biopsy is safe and yields definite, detailed histology in massive renal tumors in children. Preoperative chemotherapy facilitates surgery in these patients. J Clin Oncol 9:406-415. o 1991 by American Society of Clinical Oncology.
safety and feasibility of the operation. Wilms' tumors are soft and fragile, and the risk of operative rupture and abdominal tumor spillage is reduced if preoperative chemotherapy and/or radiotherapy has been administered. 4-9 Because abdominal spillage must be avoided, biopsy alone has also been strongly discouraged. The ideology of preoperative chemotherapy has been adopted in the treatment of several pediatric solid tumors, eg, urogenital rhabdomyosarcoma, osteosarcoma, and Ewing's sarcoma,' 0 • and has been found to be helpful in reducing tumor size, turning inoperable to operable, and in avoiding mutilating operative procedures. Chemotherapy is probably also more effective in the control of micrometastases in the preoperative setting, ie, while the primary tumor still is present.' 4 Regarding Wilms' tumor, preoperative chemotherapy, possibly combined with radiotherapy, seems beneficial in many aspects. However, the histologic picture is mostly destroyed after preoperative therapy is given, and importantly, it has been shown in both SIOP and NWTS materials that as many as 5% to 10% of preoperative Wilms' tumor diagnoses without histology turn out to be 5 15,16
wrong.,,6 In this report, we present an approach that combines the benefits of the above mentioned policies. First, by obtaining initially a percutane-
Journalof Clinical Oncology, Vol 9, No 3 (March), 1991: pp 406-415
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS ous posterior needle biopsy, a good specimen for
histology is obtained without creating abdominal spillage. Second, by performing preoperative chemotherapy, and radiotherapy when necessary, the tumor bulk is reduced in size thus facilitating operation. This approach applies only to massive and/or inoperable tumors. Small- or medium-sized tumors are still removed by primary operation.
407
amount of tumor shrinkage is, in our experience, consistent with tumor becoming more condensed and less fragile. Postoperative therapy included chemotherapy and radiotherapy according to stage and histology, as on the NWTS-3
protocols.'• Method of PercutaneousNeedle Biopsy
Between August 1986 and December 1989, we had a total of 11 new cases of renal tumors in children at the Children's Hospital, University of Helsinki, Helsinki, Finland. Seven of them, four boys and three girls, presented with massive primary renal tumors and were chosen for the study. The (median, 2½z to 88 years years (median, months to ranged from from 88 months ages ranged ages usedyears) anomalies. congenital (Table 1). None had associated
The needle biopsy of the tumor was obtained in general anesthesia or in heavy sedation with local anesthesia, while the child was lying down in ventral position. The specimens were obtained by percutaneous technique by using a TruCut disposable needle (Travenol Laboratories, Inc, Deerfield, IL), similar to regular renal biopsies in our nephrology unit. This technique allows for 0.6 to 1.0 mm thick and 10 to 20 mm long biopsy cores. Posterior direction was used to avoid peritoneal tumor contamination, and three to four biopsy samples were usually cut from the lower pole of the renal tumor in which it was palpable and easily accessible. CT scans were and contrast-enhanced guidance Ultrasound to localize the biopsy site. used to localize the biopsy site.
Outline for Clinical Approach
HistologicEvaluationof Biopsy Samples and Tumors
The initial work-up included abdominal film; abdominal ultrasound; computed tomographic (CT) scan of the abdomen, chest, and head; chest x-ray; skeletal survey; bone scan;h and bone marrow aspirate and biopsy. Intravenous
with a lancet; one for frozen section, one for electron
PATIENTS AND METHODS Patients
scan; and bone marrow aspirate and biopsy. Intravenous
Tumors
microscopy, and one to be embedded in paraffin. The frozen section served for immediate diagnosis, as well as for
pyelography pictures were taken by using the CT contrast immunohistochemistry. secretion. Laboratory tests were performed to evaluate For the paraffin sections, the samples were fixed with hematologic status, liver and kidney function, fluid and formalin, embedded in paraffin, phosphate-buffered 10% stained status. balance, electrolyte and infection and with hematoxylin-eosin, van Gieson, and periand stained with hematoxylin-eosin, van Gieson, and periAfter the initial work-up, the sequence of events was as odic acid Schiff. follows. A team including a pediatric surgeon, a pediatric For immunohistochemistry, frozen tissue was sectioned frozen tissue was sectioned Fora immunohistochemistry, a radiologist and evaluated the case. oncologist, and oncologist, case. Those Those with Reichert-Jung cryomicrotome at 7 lm on polylysinwithout or massive with children children who had massive renal tumors, with whohad or without renaltumors, coated glass slides, fixed with cold acetone, and stained by metastases, in whom a primary operation was considered to the indirect streptavidin-biotin staining method (Zymed, San Francisco, CA). Commercial antibodies effective against involve hazard and risk of tumor spillage, were selected for San Francisco, CA). Commercial antibodies effective against preoperative chemotherapy and preceding needle biopsy.neurofilaments (Progen, Heidelberg, Our definition for "massive" was a tumor size close to or 68, 160, a 200ndkd neurof ent s (Progen, Heidelberg, (Progen), desmin (Sanbio Uden The NetherGermany), synaptophysin over half of the cross-sectional left-to-right body diameter lands), smooth muscle actin (Dacopats, Copenhagen, Den(or of horizontal body cross-sectional area) at the tumor lands), smooth muscle actin (Dacopats, Copenhagen, Densite, as measured on the CT scans. In contrast, children with mark), and leukocyte common antigen (Dacopats) were small- or medium-sized renal tumors (not exceeding 40% of used. All specimens were analyzed by using an Olympus body cross-section) underwent prompt operation and were light microscope. not included in the study. For the electron microscopy, pieces of 1 mm' of tissue The children with massive tumors first underwent a were fixed in 2.5% phosphate-buffered glutaraldehyde. The percutaneous 0 needle biopsy from the tumor for percutaneous posterior samples were then stored at C and and were were only processed +4°C at +4 samples were then stored histology. They then received preoperative preoperative chemotherapy. chemotherapy. further if the regular histology and immunohistochemistry The regimen consisted of vincristine and dactinomycin in were not conclusive enough for the final diagnosis. the form of 5-day courses every second week, with dactinomycin at 0.015 mg/kg for 5 days and vincristine 1.5 mg/m2 on Sugical Methods days 1 and 5. Preoperative radiotherapy was reserved for those who did not have sufficient tumor response by A transverse upper abdominal transperitoneal incision chemotherapy alone. The children were followed-up by was used. The extent of the tumor was assessed by inspecabdominal ultrasounds and CT scans, and after sufficient tion and palpation of the peritoneal cavity and surfaces, the tumor shrinkage, they had surgery. Sufficient tumor reliver, the paraaortic areas, and the vena cava. The contralatsponse was, by our definition, a decrease by 20% or more of eral kidney was exposed and carefully inspected and palthe horizontal diameter of the tumor, or a decrease of pated, before proceeding with removal of the tumor and the tumor diameter to less than 40% of the cross-sectional involved kidney. A simple nephrectomy was performed. An left-to-right body diameter. Even if these guidelines were ipsilateral adrenalectomy was performed when necessary to arbitrary and somewhat variable on an individual basis, this ensure complete resection of the tumor. If the tumor was
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
408
SAARINEN ET AL
BI-
E
.
E
E
-.
•o
o
O -0
E0
E
co
E
o
0
,,uuUu
0
.
o
g
.
E-
>3 ..
>
E
E E
'ED
-
E
0
.
o
>>
00.
E
0
U
o0
E
1,
0
0.
.-
_
o
>
>.
>
>0
> -
o
0
Su
>
-,-
0
>
C0O
•o a..
u
0
"t
"
~Eu
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS fixed to the surrounding structures, part of the involved tissues, eg, the diaphragm, the mesocolon, the fasciae, and muscles of the back, were excised. The regional lymph nodes were sampled in all cases for staging purposes, along with the removal of suspected juxtaregional paracaval and paraaortic nodes. However, no attempt was made to perform a complete retroperitoneal lymph node dissection.
RESULTS
Histopathologyof Needle Biopsy Samples and Tumors To avoid nonrepresentative sampling, the needle biopsy cores were first checked by using a
409
dissection microscope. The needle biopsy specimens turned out to be representative in all seven cases, and of adequate size for frozen section, immunohistochemistry, and paraffin histology (Fig 1A). Histology showed classic Wilms' tumor with favorable histology in five patients (Fig 1B), Wilms tumor with diffuse anaplasia and unfavorable histology in one patient, and renal sarcoma of rhabdoid type in one patient (Fig 1C, Table 2). Only in the case of renal rhabdoid sarcoma did immunohistochemical analysis play a significant role in the final diagnosis. The sarcomatous cells
Fig 1. Photomicrographs of needle biopsy samples of massive renal tumors. (A) Low-power (magnification x 160) micrograph of a needle biopsy core; (B) classic, mixed-type Wilms' tumor (magnification x 300); (C) rhabdoid renal sarcoma (magnification x 700).
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
410
SAARINEN ET AL
expressed both keratin and vimentin in abundance. In addition, scattered foci of cells containing 68 kd neurofilament were found. In all other cases, immunohistochemistry was confirmatory to the histology finding of Wilms' tumor, with cytokeratins in the tubules, and vimentin in the blastemal cells. Electron microscopy was helpful only in the diagnosis of the renal rhabdoid sarcoma, showing typical perinuclear aggregates of intermediatesized filaments. When the specimens from the subsequent operation after preoperative chemotherapy were studied, two of the tumors were totally necrotic with no viable tumor tissue preserved (Fig 2, Table 2); four were widely necrotic but included small islets of tumor cells appearing similar to those found in the previous needle biopsy; one, the renal sarcoma, included a large amount of viable tumor (after a single course of chemotherapy), and the histology was consistent with the needle biopsy finding. Accordingly, the analysis of the removed whole tumors did not change the histopathologic diagnoses obtained in the needle biopsy. In con-
trast, the operation specimens were not diagnostic because of total necrosis in two cases, and a reliable analysis of anaplasia was not possible in four cases because the vast majority of the tumor cells were necrotic (Table 2). Complications of PercutaneousNeedle Biopsy The events during preoperative therapy, in reference to possible biopsy-related complications, are summarized in Table 3.
Fig 2.
Specimen of a Wilms' tumor taken at the time of
same as in Fig 1A, which was removed after operation, preoperative chemotherapy. The tumor, indicated by a star, is necrotic (magnification x 160), and no evaluation of anaplasia is possible.
To detect intratumoral bleeding, as well as to follow tumor response to therapy, all patients were followed-up by abdominal ultrasound and CT scan. Hematomas were confirmed at operation. Three of the seven patients had tumor hematomas (Table 3). Patients no. 3 and 5 (Fig 3) had subcapsular, sickle-shaped hematomas posterolaterally in the tumor. Although the correlation in
Table 2. Histology of the Primary Needle Biopsy Samples and the Operation Specimens After Preoperative Chemotherapy in Seven Children With Massive Renal Tumors Patient No.
Needle Biopsy Sample
Operation Specimen
1 2
Wilms' tumor, mixed type, FH Wilms' tumor, diffuse anaplasia, UH
3
Wilms' tumor, mixed type, FH
4
Renal rhabdoid sarcoma
5
Wilms' tumor, mixed type, FH
6
Wilms' tumor, mixed type, FH
7
Wilms' tumor, mixed type, FH
Totally necrotic, histology not assessable Nearly totally necrotic, histology not assessable Widely necrotic, small islets of tumor cells; Wilms' tumor, mixed type Widely necrotic, lots of viable tumor; renal rhabdoid sarcoma Widely necrotic, some viable tumor; Wilms' tumor, mixed type Nearly totally necrotic, small foci of viable tumor; Wilms' tumor, mixed type Widely necrotic, some viable tumor; Wilms' tumor, mixed type
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
411
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS
0
Iz
Z
0.
0
0
a
Z
Z
0 .2
.2 -0 f ff
.
-
0
0
Z
I
0
o
P
a
o
-o
Euo.
--
pr-
-•
"to
-0
--
-"
t
0
E- E
-v V
w
0
- • u
-
zz
z
)
a
0E 0
".r?
oi I oE
oL
~
0
" -t-~2
E 0-
aso ~
"~
I
.E
,)
U
*
., CR
I
-0
00
00
00n
t~lc
•
•EE
ij
0 0'
0
-~ r A
-
o
0
.;~
I. o
2
CL
0
@ 000
.2 w0
0
ca
20
0
C.
X
-
C.sE
ri
0
CAr
CC
o
S
0.p
a
F
~ ~E2 *
u
0
C
CC 0
0
U
0
aa 0)~0~
So
m
0
C5
h.
C
0
ue,
a -
0
x
W NO-t
- a
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
SAARINEN ET AL
412
Fig 3. Contrast-enhanced CT scan of an infant 8 months old with massive Wilms' tumor (patient no. 5, Table 1). (A) At diagnosis, (B) preoperative status showing laterally a sickle-shaped subcapsular hematoma.
time was not good, the posterolateral location raises the possibility that the needle biopsy might have had a role in causing these hematomas. Patient no. 3 bled slowly and gradually, whereas patient no. 5 bled acutely, as indicated by a drop in hemoglobin concentration, probably facilitated by the start of a septic episode. Patient no. 6 had multiple intratumoral bleeds occurring in several small episodes associated with a drop in platelet count after each course of dactinomycin. Postoperatively, after being given a course of dactinomycin and after being discharged home in good condition, her platelet count dropped again and a massive intraperitoneal bleed was verified to be of diffuse character in a relaparotomy 11 days postoperation. She recovered after 1 week's aggressive blood component and antifibrinolytic therapy. We concluded that both her intratumoral and postoperative bleeds were spontaneous, most likely related to dactinomycin-induced thrombocytopenia, but unrelated to the needle biopsy (Table 3). Six of the seven patients had fever, and five received antibiotics during the preoperative period (Table 3). However, only four were considered to have suspected sepsis, based on the clinical picture and high C-reactive protein levels; none were blood culture-positive. The start of these septic episodes had no correlation with the time of the needle biopsy (Table 3). Patient no. 3 first had a laparotomy performed for suspected appendicitis at a local hospital; at operation, a large retroperitoneal tumor was detected. This patient remained febrile and septic throughout the following preoperative chemotherapy period. Indium granulocyte scan did not show any abscess. At the second operation and tumor extirpation, no pus was
detected. Bacterial cultures taken during the operation proved negative. One week after tumor removal he was well; therefore, in part, the fever was attributed to tumor (Table 3). Accordingly, none of the infectious episodes originated in the tumor biopsy. Moreover, none of the patients had significant neutropenia during the vincristinedactinomycin chemotherapy that would have rendered them particularly susceptible for sepsis. Effect of PreoperativeTherapy All patients received preoperative chemotherapy for 4 to 13 weeks (median, 7 weeks) except for the renal sarcoma patient who was operated on after 2 weeks. Three or four vincristine-dactinomycin courses were required for desired tumor response (see Methods). One patient with metastatic disease also required tumor bed irradiation with 1,600 cGy. The largest horizontal tumor diameter ranged from 8 to 15 cm (median, 13 cm) in the primary CT scan (Table 1). Tumor shrinkage achieved by preoperative therapy ranged from 1.1 to 5.4 cm (median, 2.9 cm) or 13% to 41% (median, 22%), when evaluated as the change in the largest horizontal tumor diameter in the CT scans (Fig 4). The best responses were seen in the largest tumors, and the lowest figures were those of the renal sarcoma patient. The percentage of tumor in the body left-to-right diameter at diagnosis was a median of 57% (range, 44% to 68%), and it was reduced preoperatively to a median of 40% (range, 35% to 52%). Feasibilityof SurgeryAfter Preoperative Therapy Operative tumor removal was performed by the same surgeon (S.W.) after he felt comfortable with
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS
413
Fig 4. Contrast-enhanced CT scans of two children with Wilms' tumor, demonstrating the effect of preoperative therapy. (A) Patient no. 1 (Table 1) at diagnosis, (B) patient no. 1 preoperatively. (C) Patient no. 2 at diagnosis, (D) patient no. 2 preoperatively.
the tumor response achieved by preoperative therapy. In other words, four of the seven patients (those with the largest tumors initially) had reduction of the horizontal diameter of the tumors by 20% or more (41%, 37%, 26%, 22%), and the remaining three had the horizontal diameters of their tumors reduced to < 40% of the left-to-right body diameter. The tumors still ranged 300 to 1,000 g in weight and 6 to 11 cm in diameter. No operative tumor rupture or minor spillage occurred. Two tumors (patients no. 2 and 6, Table 1) infiltrated the diaphragm and the muscles of the back extensively; one of the two also infiltrated the mesocolon. In two other cases, there was infiltration beyond the tumor pseudocapsule, in the psoas muscle in one (patient no. 1) and in the mesocolon in the other (patient no. 3). Adrenalectomy was performed in one patient only (no. 2). In none of the patients was there tumor extension into the vena cava or beyond. Two tumors contained a flat subcapsular hematoma (patients no. 3 and 5). One patient had an enlarged, congested liver with ascites, but no evidence of metastasis or infectious
foci were found (patient no. 5). In conclusion, radical tumor extirpation in all these patients was feasible and uneventful. Outcome ofthe Patients Table 1 presents the patients in detail, including age, sex, stage, histology, preoperative and postoperative therapy, and event-free survival. All are well and disease-free 7 to 47 months from the diagnosis, and three are off therapy for 28, 3, and 2 months. DISCUSSION The natural stage distribution of Wilms' tumor is known to be different in America and in Europe, with a higher proportional presentation of low stages in the United States.3 NWTS reports only 5% (17) of the tumors to be considered primarily inoperable because of massive size. Finland, like other European countries, seems to have an overpresentation of massive tumors. During the past 5 years, Children's Hospital, University of Helsinki, Finland, had 15 Wilms' tumors of which nine were
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
414
SAARINEN ET AL
massive, and three metastatic; only six of 15 (40%) were of stage I and II tumors according to the NWTS staging, compared with about 70% of stages I and II in NWTS-2.' 7 This trend has prompted us to develop the present protocol that combines the benefits of both NWTS and SIOP policies. Our approach with an initial percutaneous posterior needle biopsy from the tumor yields accurate diagnosis with sufficient samples for good quality histology. On the other hand, preoperative chemotherapy, and radiotherapy when necessary, allows tumor shrinkage and facilitates surgery with a reduced risk of rupture or tumor spillage. Chemotherapy in the preoperative setting will theoretically also offer better control of metastases.14 Tumor staging between II and III, according to the NWTS-4 guidelines, may sometimes be complicated; if staging at diagnosis is chosen, it has to be based on the CT scan. Staging at surgery is giving different information, since after effective chemotherapy, eg, enlarged, affected lymph nodes may not show any more viable tumor. Because lymph node involvement has significant negative effect on survival, 17 we did choose to upstage massive tumors with enlarged nodes to stage III, whether or not viable tumor was seen in the nodes removed at operation (Table 1). Concerns of complications are involved in biopsy procedures regarding Wilms' tumor. The most serious concern is that of tumor soilage. However, if a needle biopsy is performed posterolaterally so that peritoneal contamination is avoided, only local soilage may occur. According to the current NWTS-4 protocol, this will place the patient in stage II category. If percutaneous needle biopsy is applied to massive stage II, III, and IV tumors only, local tumor soilage loses significance. Another concern is the risk of intratumoral bleeding, but this may either be spontaneous or biopsy-related. In our material, three of seven had intratumoral hematomas, two of which were subcapsular and might have been biopsy-related, one being significant while requiring RBC transfusions. These risks must be weighed against the risks of primary operation of a massive, fragile tumor. In our experience, surgical excision of the primary tumor as the first step of treatment would have been hazardous in all our cases, at least regarding operative tumor rupture, and mutilating operative procedures might have been required in
some patients. In contrast, preoperative therapy without definite diagnosis would have included a real possibility of obscuring the histology forever. We want to emphasize that in our small series of seven patients, one case of unfavorable histology (diffuse anaplasia) and one of rhabdoid sarcoma would have been missed without the initial needle biopsy, possibly resulting in suboptimal therapy. The risk of biopsy-related infection also exists, although we believe that it is small, particularly if the patient is not driven into profound neutropenia by the preoperative chemotherapy regimen. Fever and infection may occur during the preoperative period, as it did in five of seven of our patients. However, none of these episodes was considered biopsy-related. Percutaneous fine-needle aspiration biopsy has been advocated as a means of obtaining histology, 18 21 but it is suboptimal in recognizing the definite histologic type. For example, Verdeguer et a121 reported 12 unsatisfactory specimens out of 70 fine-needle aspirations, and both false-positive and false-negative results occurred. Moreover, the risks of bleeding or infection may not be smaller in using fine-needle aspiration, as opposed to our method. Bray et al"9 describe extensive hemorrhages within each Wilms' tumor after fine-needle aspiration biopsy; it is not stated whether they were considered spontaneous or biopsy-related. The approach of combining percutaneous needle biopsy and preoperative chemotherapy appears beneficial in the management of massive and/or inoperable renal tumors in children. It might also be suitable for bilateral Wilms' tumor patients who benefit from preoperative chemotherapy. Our approach, like NWTS,2 focuses on the individual and his or her needs, while still being able to facilitate surgery by preoperative therapy. We believe, however, that primary operation is optimal in small- and medium-size tumors. Combined radiotherapy and chemotherapy has produced responses similar to those obtained by preoperative chemotherapy alone. 5 We would like to preserve preoperative radiotherapy exclusively for treating patients for whom radiotherapy is designed: stage IVs and clear stage IIIs, on condition that prior chemotherapy treatment had not been effective. In this study, the combination of dactinomycin and vincristine was sufficient in six of seven patients. The preoperative chemotherapy
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS
might be developed in the future, possibly by prolonging the duration, and by including more active chemotherapeutic agents in selected cases. In conclusion, percutaneous posterior needle biopsy is a feasible and safe procedure that yields adequate samples for detailed histology in chil-
415
dren with massive renal tumors. An initial needle biopsy allows administration of preoperative therapy without the risk of losing information regarding histology. Chemotherapy in the neoadjuvant setting is very helpful in reducing tumor size and facilitating surgery in massive Wilms' tumors.
REFERENCES 1. D'Angio GJ, Beckwith JB, Breslow N, et al: Wilms' tumor (nephroblastoma, renal embryoma), in Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology. Philadelphia, PA, Lippincott, 1989, pp 583-606 2. Lanzkowsky P: Wilms' tumor, in Lanzkowsky P (ed): Pediatric Oncology. New York, NY, McGraw-Hill, 1983, pp 232-266 3. Tournade M-F, Lemerle J, Sarrazin D, et al: Tumours of the kidney, in Voute PA, Barrett A, Bloom HJG (eds): Cancer in Children. New York, NY, Springer Verlag, 1986, pp 252-264 4. Wagget J, Koop CE: Wilms' tumor: Preoperative radiotherapy and chemotherapy in the management of massive tumors. Cancer 26:338-340, 1970 5. Lemerle J, Voute PA, Tournade MF, et al: Preoperative versus postoperative radiotherapy, single versus multiple courses of actinomycin D, in the treatment of Wilms' tumor. Cancer 38:647-654, 1976 6. Bracken RB, Sutow WW, Jaffe N, et al: Preoperative chemotherapy for Wilms' tumor. Urology 19:55-60, 1982 7. Kalifa C, Tournade MF, Patte C, et al: Pediatric solid tumors, in Pinedo HM (ed): Cancer Chemotherapy. Amsterdam, The Netherlands, Excerpta Medica, 1982, p 412 8. Lemerle J, Voute PA, Tournade MF, et al: Effectiveness of preoperative chemotherapy in Wilms' tumor: Results of an International Society of Pediatric Oncology (SIOP) clinical trial. J Clin Oncol 1:604-609, 1983 9. Broecker BH, Perlmutter AD: Management of unresectable Wilms' tumor. Urology 24:170-174, 1984 10. Raney R, Hays D, Maurer H, et al: Primary chemotherapy, radiation therapy and/or surgery for children with sarcoma of the prostate, bladder, or vagina: Preliminary results of the Intergroup Rhabdomyosarcoma Study (IRSII), 1978-1982. Proc Am Soc Clin Onc 2:75, 1983 (abstr) 11. Etcubanas E, Rao B, Kumar M, et al: Delayed surgery and local tumor control in childhood rhabdomyosarcoma. Proc Am Soc Clin Onc 3:78, 1984 (abstr) 12. Rosen G, Caparros B, Huvos AG, et al: Preoperative
therapy for osteogenic sarcoma: Selection of postoperative adjuvant chemotherapy based on the response of the primary tumor to preoperative therapy. Cancer 49:12211230, 1982 13. Miser JS, Triche TJ, Pritchard DJ, et al: Ewing's sarcoma and the nonrhabdomyosarcoma soft tissue sarcomas of childhood, in Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology. Philadelphia, PA, Lippincott, 1989, pp 659-688 14. Fisher B, Gunduz N, Saffer EA: Influence of the interval between primary tumor removal and chemotherapy on kinetics and growth of metastases. Cancer Res 43:14881492, 1983 15. D'Angio GJ, Evans AE, Breslow N, et al: The treatment of Wilms' tumor. Results of the National Wilms' tumor study. Cancer 38:633-646, 1976 16. Lanzkowsky P: Wilms' tumor, in Lanzkowsky P (ed): Pediatric Oncology. New York, NY, McGraw-Hill, 1983, p 247 17. D'Angio GJ, Evans A, Breslow N, et al: The treatment of Wilms' tumor: Results of the second national Wilms' tumor study. Cancer 47:2302-2311, 1981 18. Schaller RT, Schaller JF, Bushmann C: The usefulness of percutaneous fine-needle aspiration biopsy in infants and children. J Pediatr Surg 18:398-401, 1983 19. Bray GL, Pendergrass TW, Schaller RT Jr, et al: Preoperative chemotherapy in the treatment of Wilms' tumor diagnosed with the aid of fine needle aspiration biopsy. Am J Pediatr Hematol Oncol 8:75-78, 1986 20. van Sonnenberg E, Wittich GR, Edwards DK, et al: Percutaneous diagnostic and therapeutic interventional radiologic procedures in children: Experience in 100 patients. Radiology 162:601-605, 1987 21. Verdeguer A, Castel V, Torres V, et al: Fine-needle aspiration biopsy in children: Experience in 70 cases. Med Pediatr Oncol 16:98-100, 1988 22. D'Angio GJ: Editorial: SIOP and the management of Wilms' tumor. J Clin Oncol 1:595-596, 1983
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
WILMS'
TUMOR is one of the pediatric
malignancies for which there has been the most success in responding to therapy; overall survival rates have risen to over 80% within the last 10-year period.' Advanced tumor stage, unfavorable histology, and especially both factors together still carry a worse prognosis, and improvement of therapy in these subgroups is needed.' Despite the uniformly good results, the policies on how to approach a patient with Wilms' tumor differ significantly between the two most prominent study groups, ie, the National Wilms' Tumor 2 Study (NWTS) group in the United States, and the International Society of Pediatric Oncology 3 (SIOP) group in Europe. The NWTS emphasizes accurate staging and an in-depth, detailed histology, which constitute the basis for choice of consequent therapy. Therefore, the NWTS stresses that operative removal of the tumor should be 2 performed as the first step, whenever possible.' On the other hand, SIOP wants to emphasize
From the Children'sHospital, University of Helsinki, Helsinki, Finland. Submitted April 11, 1990; accepted September 4, 1990. Address reprint requests to Ulla M. Saarinen, MD, Children's Hospital, University of Helsinki, Stenbiickinkatu 11, 00290 Helsinki, Finland. C 1991 by American Society of Clinical Oncology. 0732-183X/91/0903-0011$3.00/0
406
necrotic. In the others, which still included viable tumor, the histology corresponded well to the needle biopsy findings. One case with unfavorable histology and one with rhabdoid sarcoma would have been missed and given suboptimal therapy without the primary needle biopsy. As possible biopsy-related complications, subcapsular intratumoral bleeding was recognized in two patients. We conclude that percutaneous posterior needle biopsy is safe and yields definite, detailed histology in massive renal tumors in children. Preoperative chemotherapy facilitates surgery in these patients. J Clin Oncol 9:406-415. o 1991 by American Society of Clinical Oncology.
safety and feasibility of the operation. Wilms' tumors are soft and fragile, and the risk of operative rupture and abdominal tumor spillage is reduced if preoperative chemotherapy and/or radiotherapy has been administered. 4-9 Because abdominal spillage must be avoided, biopsy alone has also been strongly discouraged. The ideology of preoperative chemotherapy has been adopted in the treatment of several pediatric solid tumors, eg, urogenital rhabdomyosarcoma, osteosarcoma, and Ewing's sarcoma,' 0 • and has been found to be helpful in reducing tumor size, turning inoperable to operable, and in avoiding mutilating operative procedures. Chemotherapy is probably also more effective in the control of micrometastases in the preoperative setting, ie, while the primary tumor still is present.' 4 Regarding Wilms' tumor, preoperative chemotherapy, possibly combined with radiotherapy, seems beneficial in many aspects. However, the histologic picture is mostly destroyed after preoperative therapy is given, and importantly, it has been shown in both SIOP and NWTS materials that as many as 5% to 10% of preoperative Wilms' tumor diagnoses without histology turn out to be 5 15,16
wrong.,,6 In this report, we present an approach that combines the benefits of the above mentioned policies. First, by obtaining initially a percutane-
Journalof Clinical Oncology, Vol 9, No 3 (March), 1991: pp 406-415
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS ous posterior needle biopsy, a good specimen for
histology is obtained without creating abdominal spillage. Second, by performing preoperative chemotherapy, and radiotherapy when necessary, the tumor bulk is reduced in size thus facilitating operation. This approach applies only to massive and/or inoperable tumors. Small- or medium-sized tumors are still removed by primary operation.
407
amount of tumor shrinkage is, in our experience, consistent with tumor becoming more condensed and less fragile. Postoperative therapy included chemotherapy and radiotherapy according to stage and histology, as on the NWTS-3
protocols.'• Method of PercutaneousNeedle Biopsy
Between August 1986 and December 1989, we had a total of 11 new cases of renal tumors in children at the Children's Hospital, University of Helsinki, Helsinki, Finland. Seven of them, four boys and three girls, presented with massive primary renal tumors and were chosen for the study. The (median, 2½z to 88 years years (median, months to ranged from from 88 months ages ranged ages usedyears) anomalies. congenital (Table 1). None had associated
The needle biopsy of the tumor was obtained in general anesthesia or in heavy sedation with local anesthesia, while the child was lying down in ventral position. The specimens were obtained by percutaneous technique by using a TruCut disposable needle (Travenol Laboratories, Inc, Deerfield, IL), similar to regular renal biopsies in our nephrology unit. This technique allows for 0.6 to 1.0 mm thick and 10 to 20 mm long biopsy cores. Posterior direction was used to avoid peritoneal tumor contamination, and three to four biopsy samples were usually cut from the lower pole of the renal tumor in which it was palpable and easily accessible. CT scans were and contrast-enhanced guidance Ultrasound to localize the biopsy site. used to localize the biopsy site.
Outline for Clinical Approach
HistologicEvaluationof Biopsy Samples and Tumors
The initial work-up included abdominal film; abdominal ultrasound; computed tomographic (CT) scan of the abdomen, chest, and head; chest x-ray; skeletal survey; bone scan;h and bone marrow aspirate and biopsy. Intravenous
with a lancet; one for frozen section, one for electron
PATIENTS AND METHODS Patients
scan; and bone marrow aspirate and biopsy. Intravenous
Tumors
microscopy, and one to be embedded in paraffin. The frozen section served for immediate diagnosis, as well as for
pyelography pictures were taken by using the CT contrast immunohistochemistry. secretion. Laboratory tests were performed to evaluate For the paraffin sections, the samples were fixed with hematologic status, liver and kidney function, fluid and formalin, embedded in paraffin, phosphate-buffered 10% stained status. balance, electrolyte and infection and with hematoxylin-eosin, van Gieson, and periand stained with hematoxylin-eosin, van Gieson, and periAfter the initial work-up, the sequence of events was as odic acid Schiff. follows. A team including a pediatric surgeon, a pediatric For immunohistochemistry, frozen tissue was sectioned frozen tissue was sectioned Fora immunohistochemistry, a radiologist and evaluated the case. oncologist, and oncologist, case. Those Those with Reichert-Jung cryomicrotome at 7 lm on polylysinwithout or massive with children children who had massive renal tumors, with whohad or without renaltumors, coated glass slides, fixed with cold acetone, and stained by metastases, in whom a primary operation was considered to the indirect streptavidin-biotin staining method (Zymed, San Francisco, CA). Commercial antibodies effective against involve hazard and risk of tumor spillage, were selected for San Francisco, CA). Commercial antibodies effective against preoperative chemotherapy and preceding needle biopsy.neurofilaments (Progen, Heidelberg, Our definition for "massive" was a tumor size close to or 68, 160, a 200ndkd neurof ent s (Progen, Heidelberg, (Progen), desmin (Sanbio Uden The NetherGermany), synaptophysin over half of the cross-sectional left-to-right body diameter lands), smooth muscle actin (Dacopats, Copenhagen, Den(or of horizontal body cross-sectional area) at the tumor lands), smooth muscle actin (Dacopats, Copenhagen, Densite, as measured on the CT scans. In contrast, children with mark), and leukocyte common antigen (Dacopats) were small- or medium-sized renal tumors (not exceeding 40% of used. All specimens were analyzed by using an Olympus body cross-section) underwent prompt operation and were light microscope. not included in the study. For the electron microscopy, pieces of 1 mm' of tissue The children with massive tumors first underwent a were fixed in 2.5% phosphate-buffered glutaraldehyde. The percutaneous 0 needle biopsy from the tumor for percutaneous posterior samples were then stored at C and and were were only processed +4°C at +4 samples were then stored histology. They then received preoperative preoperative chemotherapy. chemotherapy. further if the regular histology and immunohistochemistry The regimen consisted of vincristine and dactinomycin in were not conclusive enough for the final diagnosis. the form of 5-day courses every second week, with dactinomycin at 0.015 mg/kg for 5 days and vincristine 1.5 mg/m2 on Sugical Methods days 1 and 5. Preoperative radiotherapy was reserved for those who did not have sufficient tumor response by A transverse upper abdominal transperitoneal incision chemotherapy alone. The children were followed-up by was used. The extent of the tumor was assessed by inspecabdominal ultrasounds and CT scans, and after sufficient tion and palpation of the peritoneal cavity and surfaces, the tumor shrinkage, they had surgery. Sufficient tumor reliver, the paraaortic areas, and the vena cava. The contralatsponse was, by our definition, a decrease by 20% or more of eral kidney was exposed and carefully inspected and palthe horizontal diameter of the tumor, or a decrease of pated, before proceeding with removal of the tumor and the tumor diameter to less than 40% of the cross-sectional involved kidney. A simple nephrectomy was performed. An left-to-right body diameter. Even if these guidelines were ipsilateral adrenalectomy was performed when necessary to arbitrary and somewhat variable on an individual basis, this ensure complete resection of the tumor. If the tumor was
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
408
SAARINEN ET AL
BI-
E
.
E
E
-.
•o
o
O -0
E0
E
co
E
o
0
,,uuUu
0
.
o
g
.
E-
>3 ..
>
E
E E
'ED
-
E
0
.
o
>>
00.
E
0
U
o0
E
1,
0
0.
.-
_
o
>
>.
>
>0
> -
o
0
Su
>
-,-
0
>
C0O
•o a..
u
0
"t
"
~Eu
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS fixed to the surrounding structures, part of the involved tissues, eg, the diaphragm, the mesocolon, the fasciae, and muscles of the back, were excised. The regional lymph nodes were sampled in all cases for staging purposes, along with the removal of suspected juxtaregional paracaval and paraaortic nodes. However, no attempt was made to perform a complete retroperitoneal lymph node dissection.
RESULTS
Histopathologyof Needle Biopsy Samples and Tumors To avoid nonrepresentative sampling, the needle biopsy cores were first checked by using a
409
dissection microscope. The needle biopsy specimens turned out to be representative in all seven cases, and of adequate size for frozen section, immunohistochemistry, and paraffin histology (Fig 1A). Histology showed classic Wilms' tumor with favorable histology in five patients (Fig 1B), Wilms tumor with diffuse anaplasia and unfavorable histology in one patient, and renal sarcoma of rhabdoid type in one patient (Fig 1C, Table 2). Only in the case of renal rhabdoid sarcoma did immunohistochemical analysis play a significant role in the final diagnosis. The sarcomatous cells
Fig 1. Photomicrographs of needle biopsy samples of massive renal tumors. (A) Low-power (magnification x 160) micrograph of a needle biopsy core; (B) classic, mixed-type Wilms' tumor (magnification x 300); (C) rhabdoid renal sarcoma (magnification x 700).
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
410
SAARINEN ET AL
expressed both keratin and vimentin in abundance. In addition, scattered foci of cells containing 68 kd neurofilament were found. In all other cases, immunohistochemistry was confirmatory to the histology finding of Wilms' tumor, with cytokeratins in the tubules, and vimentin in the blastemal cells. Electron microscopy was helpful only in the diagnosis of the renal rhabdoid sarcoma, showing typical perinuclear aggregates of intermediatesized filaments. When the specimens from the subsequent operation after preoperative chemotherapy were studied, two of the tumors were totally necrotic with no viable tumor tissue preserved (Fig 2, Table 2); four were widely necrotic but included small islets of tumor cells appearing similar to those found in the previous needle biopsy; one, the renal sarcoma, included a large amount of viable tumor (after a single course of chemotherapy), and the histology was consistent with the needle biopsy finding. Accordingly, the analysis of the removed whole tumors did not change the histopathologic diagnoses obtained in the needle biopsy. In con-
trast, the operation specimens were not diagnostic because of total necrosis in two cases, and a reliable analysis of anaplasia was not possible in four cases because the vast majority of the tumor cells were necrotic (Table 2). Complications of PercutaneousNeedle Biopsy The events during preoperative therapy, in reference to possible biopsy-related complications, are summarized in Table 3.
Fig 2.
Specimen of a Wilms' tumor taken at the time of
same as in Fig 1A, which was removed after operation, preoperative chemotherapy. The tumor, indicated by a star, is necrotic (magnification x 160), and no evaluation of anaplasia is possible.
To detect intratumoral bleeding, as well as to follow tumor response to therapy, all patients were followed-up by abdominal ultrasound and CT scan. Hematomas were confirmed at operation. Three of the seven patients had tumor hematomas (Table 3). Patients no. 3 and 5 (Fig 3) had subcapsular, sickle-shaped hematomas posterolaterally in the tumor. Although the correlation in
Table 2. Histology of the Primary Needle Biopsy Samples and the Operation Specimens After Preoperative Chemotherapy in Seven Children With Massive Renal Tumors Patient No.
Needle Biopsy Sample
Operation Specimen
1 2
Wilms' tumor, mixed type, FH Wilms' tumor, diffuse anaplasia, UH
3
Wilms' tumor, mixed type, FH
4
Renal rhabdoid sarcoma
5
Wilms' tumor, mixed type, FH
6
Wilms' tumor, mixed type, FH
7
Wilms' tumor, mixed type, FH
Totally necrotic, histology not assessable Nearly totally necrotic, histology not assessable Widely necrotic, small islets of tumor cells; Wilms' tumor, mixed type Widely necrotic, lots of viable tumor; renal rhabdoid sarcoma Widely necrotic, some viable tumor; Wilms' tumor, mixed type Nearly totally necrotic, small foci of viable tumor; Wilms' tumor, mixed type Widely necrotic, some viable tumor; Wilms' tumor, mixed type
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
411
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS
0
Iz
Z
0.
0
0
a
Z
Z
0 .2
.2 -0 f ff
.
-
0
0
Z
I
0
o
P
a
o
-o
Euo.
--
pr-
-•
"to
-0
--
-"
t
0
E- E
-v V
w
0
- • u
-
zz
z
)
a
0E 0
".r?
oi I oE
oL
~
0
" -t-~2
E 0-
aso ~
"~
I
.E
,)
U
*
., CR
I
-0
00
00
00n
t~lc
•
•EE
ij
0 0'
0
-~ r A
-
o
0
.;~
I. o
2
CL
0
@ 000
.2 w0
0
ca
20
0
C.
X
-
C.sE
ri
0
CAr
CC
o
S
0.p
a
F
~ ~E2 *
u
0
C
CC 0
0
U
0
aa 0)~0~
So
m
0
C5
h.
C
0
ue,
a -
0
x
W NO-t
- a
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
SAARINEN ET AL
412
Fig 3. Contrast-enhanced CT scan of an infant 8 months old with massive Wilms' tumor (patient no. 5, Table 1). (A) At diagnosis, (B) preoperative status showing laterally a sickle-shaped subcapsular hematoma.
time was not good, the posterolateral location raises the possibility that the needle biopsy might have had a role in causing these hematomas. Patient no. 3 bled slowly and gradually, whereas patient no. 5 bled acutely, as indicated by a drop in hemoglobin concentration, probably facilitated by the start of a septic episode. Patient no. 6 had multiple intratumoral bleeds occurring in several small episodes associated with a drop in platelet count after each course of dactinomycin. Postoperatively, after being given a course of dactinomycin and after being discharged home in good condition, her platelet count dropped again and a massive intraperitoneal bleed was verified to be of diffuse character in a relaparotomy 11 days postoperation. She recovered after 1 week's aggressive blood component and antifibrinolytic therapy. We concluded that both her intratumoral and postoperative bleeds were spontaneous, most likely related to dactinomycin-induced thrombocytopenia, but unrelated to the needle biopsy (Table 3). Six of the seven patients had fever, and five received antibiotics during the preoperative period (Table 3). However, only four were considered to have suspected sepsis, based on the clinical picture and high C-reactive protein levels; none were blood culture-positive. The start of these septic episodes had no correlation with the time of the needle biopsy (Table 3). Patient no. 3 first had a laparotomy performed for suspected appendicitis at a local hospital; at operation, a large retroperitoneal tumor was detected. This patient remained febrile and septic throughout the following preoperative chemotherapy period. Indium granulocyte scan did not show any abscess. At the second operation and tumor extirpation, no pus was
detected. Bacterial cultures taken during the operation proved negative. One week after tumor removal he was well; therefore, in part, the fever was attributed to tumor (Table 3). Accordingly, none of the infectious episodes originated in the tumor biopsy. Moreover, none of the patients had significant neutropenia during the vincristinedactinomycin chemotherapy that would have rendered them particularly susceptible for sepsis. Effect of PreoperativeTherapy All patients received preoperative chemotherapy for 4 to 13 weeks (median, 7 weeks) except for the renal sarcoma patient who was operated on after 2 weeks. Three or four vincristine-dactinomycin courses were required for desired tumor response (see Methods). One patient with metastatic disease also required tumor bed irradiation with 1,600 cGy. The largest horizontal tumor diameter ranged from 8 to 15 cm (median, 13 cm) in the primary CT scan (Table 1). Tumor shrinkage achieved by preoperative therapy ranged from 1.1 to 5.4 cm (median, 2.9 cm) or 13% to 41% (median, 22%), when evaluated as the change in the largest horizontal tumor diameter in the CT scans (Fig 4). The best responses were seen in the largest tumors, and the lowest figures were those of the renal sarcoma patient. The percentage of tumor in the body left-to-right diameter at diagnosis was a median of 57% (range, 44% to 68%), and it was reduced preoperatively to a median of 40% (range, 35% to 52%). Feasibilityof SurgeryAfter Preoperative Therapy Operative tumor removal was performed by the same surgeon (S.W.) after he felt comfortable with
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS
413
Fig 4. Contrast-enhanced CT scans of two children with Wilms' tumor, demonstrating the effect of preoperative therapy. (A) Patient no. 1 (Table 1) at diagnosis, (B) patient no. 1 preoperatively. (C) Patient no. 2 at diagnosis, (D) patient no. 2 preoperatively.
the tumor response achieved by preoperative therapy. In other words, four of the seven patients (those with the largest tumors initially) had reduction of the horizontal diameter of the tumors by 20% or more (41%, 37%, 26%, 22%), and the remaining three had the horizontal diameters of their tumors reduced to < 40% of the left-to-right body diameter. The tumors still ranged 300 to 1,000 g in weight and 6 to 11 cm in diameter. No operative tumor rupture or minor spillage occurred. Two tumors (patients no. 2 and 6, Table 1) infiltrated the diaphragm and the muscles of the back extensively; one of the two also infiltrated the mesocolon. In two other cases, there was infiltration beyond the tumor pseudocapsule, in the psoas muscle in one (patient no. 1) and in the mesocolon in the other (patient no. 3). Adrenalectomy was performed in one patient only (no. 2). In none of the patients was there tumor extension into the vena cava or beyond. Two tumors contained a flat subcapsular hematoma (patients no. 3 and 5). One patient had an enlarged, congested liver with ascites, but no evidence of metastasis or infectious
foci were found (patient no. 5). In conclusion, radical tumor extirpation in all these patients was feasible and uneventful. Outcome ofthe Patients Table 1 presents the patients in detail, including age, sex, stage, histology, preoperative and postoperative therapy, and event-free survival. All are well and disease-free 7 to 47 months from the diagnosis, and three are off therapy for 28, 3, and 2 months. DISCUSSION The natural stage distribution of Wilms' tumor is known to be different in America and in Europe, with a higher proportional presentation of low stages in the United States.3 NWTS reports only 5% (17) of the tumors to be considered primarily inoperable because of massive size. Finland, like other European countries, seems to have an overpresentation of massive tumors. During the past 5 years, Children's Hospital, University of Helsinki, Finland, had 15 Wilms' tumors of which nine were
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
414
SAARINEN ET AL
massive, and three metastatic; only six of 15 (40%) were of stage I and II tumors according to the NWTS staging, compared with about 70% of stages I and II in NWTS-2.' 7 This trend has prompted us to develop the present protocol that combines the benefits of both NWTS and SIOP policies. Our approach with an initial percutaneous posterior needle biopsy from the tumor yields accurate diagnosis with sufficient samples for good quality histology. On the other hand, preoperative chemotherapy, and radiotherapy when necessary, allows tumor shrinkage and facilitates surgery with a reduced risk of rupture or tumor spillage. Chemotherapy in the preoperative setting will theoretically also offer better control of metastases.14 Tumor staging between II and III, according to the NWTS-4 guidelines, may sometimes be complicated; if staging at diagnosis is chosen, it has to be based on the CT scan. Staging at surgery is giving different information, since after effective chemotherapy, eg, enlarged, affected lymph nodes may not show any more viable tumor. Because lymph node involvement has significant negative effect on survival, 17 we did choose to upstage massive tumors with enlarged nodes to stage III, whether or not viable tumor was seen in the nodes removed at operation (Table 1). Concerns of complications are involved in biopsy procedures regarding Wilms' tumor. The most serious concern is that of tumor soilage. However, if a needle biopsy is performed posterolaterally so that peritoneal contamination is avoided, only local soilage may occur. According to the current NWTS-4 protocol, this will place the patient in stage II category. If percutaneous needle biopsy is applied to massive stage II, III, and IV tumors only, local tumor soilage loses significance. Another concern is the risk of intratumoral bleeding, but this may either be spontaneous or biopsy-related. In our material, three of seven had intratumoral hematomas, two of which were subcapsular and might have been biopsy-related, one being significant while requiring RBC transfusions. These risks must be weighed against the risks of primary operation of a massive, fragile tumor. In our experience, surgical excision of the primary tumor as the first step of treatment would have been hazardous in all our cases, at least regarding operative tumor rupture, and mutilating operative procedures might have been required in
some patients. In contrast, preoperative therapy without definite diagnosis would have included a real possibility of obscuring the histology forever. We want to emphasize that in our small series of seven patients, one case of unfavorable histology (diffuse anaplasia) and one of rhabdoid sarcoma would have been missed without the initial needle biopsy, possibly resulting in suboptimal therapy. The risk of biopsy-related infection also exists, although we believe that it is small, particularly if the patient is not driven into profound neutropenia by the preoperative chemotherapy regimen. Fever and infection may occur during the preoperative period, as it did in five of seven of our patients. However, none of these episodes was considered biopsy-related. Percutaneous fine-needle aspiration biopsy has been advocated as a means of obtaining histology, 18 21 but it is suboptimal in recognizing the definite histologic type. For example, Verdeguer et a121 reported 12 unsatisfactory specimens out of 70 fine-needle aspirations, and both false-positive and false-negative results occurred. Moreover, the risks of bleeding or infection may not be smaller in using fine-needle aspiration, as opposed to our method. Bray et al"9 describe extensive hemorrhages within each Wilms' tumor after fine-needle aspiration biopsy; it is not stated whether they were considered spontaneous or biopsy-related. The approach of combining percutaneous needle biopsy and preoperative chemotherapy appears beneficial in the management of massive and/or inoperable renal tumors in children. It might also be suitable for bilateral Wilms' tumor patients who benefit from preoperative chemotherapy. Our approach, like NWTS,2 focuses on the individual and his or her needs, while still being able to facilitate surgery by preoperative therapy. We believe, however, that primary operation is optimal in small- and medium-size tumors. Combined radiotherapy and chemotherapy has produced responses similar to those obtained by preoperative chemotherapy alone. 5 We would like to preserve preoperative radiotherapy exclusively for treating patients for whom radiotherapy is designed: stage IVs and clear stage IIIs, on condition that prior chemotherapy treatment had not been effective. In this study, the combination of dactinomycin and vincristine was sufficient in six of seven patients. The preoperative chemotherapy
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
PERCUTANEOUS NEEDLE BIOPSY IN MASSIVE RENAL TUMORS
might be developed in the future, possibly by prolonging the duration, and by including more active chemotherapeutic agents in selected cases. In conclusion, percutaneous posterior needle biopsy is a feasible and safe procedure that yields adequate samples for detailed histology in chil-
415
dren with massive renal tumors. An initial needle biopsy allows administration of preoperative therapy without the risk of losing information regarding histology. Chemotherapy in the neoadjuvant setting is very helpful in reducing tumor size and facilitating surgery in massive Wilms' tumors.
REFERENCES 1. D'Angio GJ, Beckwith JB, Breslow N, et al: Wilms' tumor (nephroblastoma, renal embryoma), in Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology. Philadelphia, PA, Lippincott, 1989, pp 583-606 2. Lanzkowsky P: Wilms' tumor, in Lanzkowsky P (ed): Pediatric Oncology. New York, NY, McGraw-Hill, 1983, pp 232-266 3. Tournade M-F, Lemerle J, Sarrazin D, et al: Tumours of the kidney, in Voute PA, Barrett A, Bloom HJG (eds): Cancer in Children. New York, NY, Springer Verlag, 1986, pp 252-264 4. Wagget J, Koop CE: Wilms' tumor: Preoperative radiotherapy and chemotherapy in the management of massive tumors. Cancer 26:338-340, 1970 5. Lemerle J, Voute PA, Tournade MF, et al: Preoperative versus postoperative radiotherapy, single versus multiple courses of actinomycin D, in the treatment of Wilms' tumor. Cancer 38:647-654, 1976 6. Bracken RB, Sutow WW, Jaffe N, et al: Preoperative chemotherapy for Wilms' tumor. Urology 19:55-60, 1982 7. Kalifa C, Tournade MF, Patte C, et al: Pediatric solid tumors, in Pinedo HM (ed): Cancer Chemotherapy. Amsterdam, The Netherlands, Excerpta Medica, 1982, p 412 8. Lemerle J, Voute PA, Tournade MF, et al: Effectiveness of preoperative chemotherapy in Wilms' tumor: Results of an International Society of Pediatric Oncology (SIOP) clinical trial. J Clin Oncol 1:604-609, 1983 9. Broecker BH, Perlmutter AD: Management of unresectable Wilms' tumor. Urology 24:170-174, 1984 10. Raney R, Hays D, Maurer H, et al: Primary chemotherapy, radiation therapy and/or surgery for children with sarcoma of the prostate, bladder, or vagina: Preliminary results of the Intergroup Rhabdomyosarcoma Study (IRSII), 1978-1982. Proc Am Soc Clin Onc 2:75, 1983 (abstr) 11. Etcubanas E, Rao B, Kumar M, et al: Delayed surgery and local tumor control in childhood rhabdomyosarcoma. Proc Am Soc Clin Onc 3:78, 1984 (abstr) 12. Rosen G, Caparros B, Huvos AG, et al: Preoperative
therapy for osteogenic sarcoma: Selection of postoperative adjuvant chemotherapy based on the response of the primary tumor to preoperative therapy. Cancer 49:12211230, 1982 13. Miser JS, Triche TJ, Pritchard DJ, et al: Ewing's sarcoma and the nonrhabdomyosarcoma soft tissue sarcomas of childhood, in Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology. Philadelphia, PA, Lippincott, 1989, pp 659-688 14. Fisher B, Gunduz N, Saffer EA: Influence of the interval between primary tumor removal and chemotherapy on kinetics and growth of metastases. Cancer Res 43:14881492, 1983 15. D'Angio GJ, Evans AE, Breslow N, et al: The treatment of Wilms' tumor. Results of the National Wilms' tumor study. Cancer 38:633-646, 1976 16. Lanzkowsky P: Wilms' tumor, in Lanzkowsky P (ed): Pediatric Oncology. New York, NY, McGraw-Hill, 1983, p 247 17. D'Angio GJ, Evans A, Breslow N, et al: The treatment of Wilms' tumor: Results of the second national Wilms' tumor study. Cancer 47:2302-2311, 1981 18. Schaller RT, Schaller JF, Bushmann C: The usefulness of percutaneous fine-needle aspiration biopsy in infants and children. J Pediatr Surg 18:398-401, 1983 19. Bray GL, Pendergrass TW, Schaller RT Jr, et al: Preoperative chemotherapy in the treatment of Wilms' tumor diagnosed with the aid of fine needle aspiration biopsy. Am J Pediatr Hematol Oncol 8:75-78, 1986 20. van Sonnenberg E, Wittich GR, Edwards DK, et al: Percutaneous diagnostic and therapeutic interventional radiologic procedures in children: Experience in 100 patients. Radiology 162:601-605, 1987 21. Verdeguer A, Castel V, Torres V, et al: Fine-needle aspiration biopsy in children: Experience in 70 cases. Med Pediatr Oncol 16:98-100, 1988 22. D'Angio GJ: Editorial: SIOP and the management of Wilms' tumor. J Clin Oncol 1:595-596, 1983
Downloaded from ascopubs.org by EAST CAROLINA UNIVERSITY on August 20, 2019 from 150.216.068.200 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
