The role of surgical excision in the management of relapsed Wilms' tumor patients with pulmonary metastases: A report from the National Wilms' Tumor Study

The Role of Surgical Excision in the Management of Relapsed Wilms’ Tumor Patients With Pulmonary Metastases: A Report From the National Wilms’ Tumor Study By Daniel M. Green, Norman E. Breslow, Yoichi Ii, Paul E. Grundy, Stephen J. Shochat, Janice Takashima, and Giulio J. D’Angio Buffalo, New York; Seattle, Washington; Edmonton, Alberta; Stanford, California; and Philadelphia, Pennsylvania l To determine the effect of surgical excision of pulmonary metastases from Wilms’ tumor on postrelapse survival, we retrospectively analyzed the clinical courses of 211 patients with stages I to Ill, favorable or unfavorable histology Wilms’ tumor entered on National Wilms’ Tumor Study-l, -2, or -3 whose first recurrence was limited to the lungs. There was no difference in the 4-year postrelapse survival percentage of favorable-histology patients with a solitary pulmonary metastasis who did or did not undergo surgical removal of the metastasis in addition to pulmonary irradiation and chemotherapy. Although histological confirmation of pulmonary relapse is frequently indicated, the present data suggest that therapeutic removal of pulmonary metastases from patients with relapsed Wilms’ tumor does not increase the percentage of patients who survive for 4 years postrelapse, compared with treatment with whole-lung irradiation and chemotherapy. Copyright o 1991 by W.B. Saunders Company INDEX WORDS: Wilms’ tumor; pulmonary metastases, cal removal.

surgi-

T

HE PROGNOSIS FOR relapse-free survival for most children with Wilms’ tumor is excellent following treatment that includes nephrectomy, combination chemotherapy, and, when indicated, radiation therapy.’ Unfortunately, some patients with Wilms’ tumor, especially those with unfavorable histology, do develop recurrent disease, most frequently in the lung.‘x3 Several investigators have reported prolonged freedom from second recurrence for patients with pulmonary metastases from Wilms’ tumor whose treatment

From the Department of Pediatrics, Roswell Park Memorial Institute, and the School of Medicine, State University of New York at Buffalo, Buffalo, m; the Depatiment of Biostatistics, University of Washington, Seattle, WA; the Department of Pediatrics, University of Alberta, Edmonton, Alberta; the Department of Pediatric Surgery, Stanford University Medical Center, Stanford, CA; the National Wilms’ Tumor Study, Data and Statistical Center, Seattle, WA; and the Department of Radiation Oncology, Hospital of the University of Pennsylvania, Philadelphia, PA. Date accepted: May 9, 1990. Supported in part by USPHS Grant CA-42326. Principal investigators at participating institutions also receive supportfrom the National Cancer Institute. Address reprint requests to Daniel M. Green, MD, Roswell Park Cancer Institute, Elm and Carlton Sts, Buffalo, NY14263. Copyright o 1991 by WB. Saunders Company 0022-3468191/2606-0021$03.OOlO 728

included excision of the metastases.4-6 There are several reports of prolonged survival of patients whose only treatment was surgical removal of the metastases.‘.lo However, no study has included enough cases to analyze the separate effects of prerelapse chemotherapy and histology, and postrelapse surgery, radiation therapy, and chemotherapy on the postrelapse survival of patients with recurrent disease limited to the lungs. Therefore, this study was undertaken to evaluate the role of surgical excision of pulmonary metastases in the management of patients with relapsed Wilms’ tumor. MATERIALS AND METHODS There were 3,675 children registered on the first, second, and third National Wilms’ Tumor Studies (NWTS-1, -2, and -3), in the randomized and followed categories as defined elsewhere.’ Children were excluded from this analysis of relapse if they had received preoperative therapy (142), were older than 15 years of age at diagnosis (19), had bilateral Wilms’ tumor (228), had a tumor in a horseshoe or solitary kidney or extrarenal site (lo), or had stage IV disease at diagnosis (416) (Table 1). The site(s) of first relapse were determined in the remaining 2,860 patients, 387 of whom had relapsed (Table 2) prior to January 1, 1985. The present analysis is restricted to the 211 patients with known histology of the 227 patients with stage I to III Wilms’ tumor whose first recurrence was limited to the lung(s) (Table 3). During the period under review, a standard protocol for the management of patients who developed pulmonary metastases did not exist. To facilitate the categorization of therapy received for the pulmonary recurrence, patients were considered to have been treated surgically as part of the initial relapse management only if surgery was performed within 30 days from the diagnosis of relapse. Likewise, patients were considered to have been treated with drugs or radiation only if chemotherapy was initiated within 30 days and radiation therapy was initiated within 60 days of the diagnosis of relapse. The data were analyzed using standard statistical methods including product limit estimates of survival curves and the log rank test.“.” Standard errors were computed using Greenwood’s formula.” Postrelapse survival and its standard error were estimated at 4 years postdiagnosis of relapse. Treatment The treatments for pulmonary relapse in known histology patients are summarized in Table 3. Some patients whose treatment is shown in Table 3 to have been radiation therapy or chemotherapy only, actually received combined modality therapy, but one modality was not initiated within 30 or 60 days of the Journal of Pediatric Surgery, Vol 26, No 6 (June), 1991:

pp728-733

PULMONARY METASTASES

729

IN WILMS’ TUMOR

Table 1. Summary of Patient Population NW-K-1

NWTS-2

NWTS-3

Total

Randomized

353

545

1,489

2,387

Followed

217

240

831

1,288

Total

570

785

2,320

3,675

Excluded

119

133

563

815

Preoperative therapy

3

2

137

142

> 15 yr old at diagnosis

3

3

13

19

33

37

158

228

5

6

Bilateral Wilms’ tumor Tumor in horseshoe kidney

0

1

Extrarenal primary

0

0

Mononephric at diagnosis Stage IV at diagnosis Total Included

0

0

3

3

80

90

246

416

451

652

1,757

2,860

of relapse. More detailed analyses by histology and number of metastases were performed in three groups of patients: those treated with whole-lung irradiation and chemotherapy, those treated with surgical excision followed by whole-lung irradiation and chemotherapy, and those treated with surgical excision and chemotherapy. diagnosis

RESULTS

Favorable Histology The distribution of favorable histology patients with a solitary metastasis by stage at diagnosis, number of drugs received in initial treatment regimen, and months from diagnosis to recurrence is shown in Table 4. More patients with stage II or III tumors were treated with surgical excision of the metastasis, followed by radiation therapy and chemotherapy. Thirteen patients with stages I to III favorable histology Wilms’ tumor who developed a solitary pulmonary metastasis underwent surgical excision of the lesion followed by treatment with whole-lung irradiation and chemotherapy (Table 5). The surgical procedures performed were wedge excision (10 patients), segmental removal (l), and lobectomy (2). Radiation therapy data were available for 11 of these 13 patients. All received whole-lung irradiation with Table 2. Distribution of Patients by Site of First Relapse NW-B-1

NWTS-2

NW-W%*

Site of relapse Lung only

79

66

114 (82)

259 (227)

Opposite kidney only

8

8

12 (5)

28 (21)

Bone only

3

3

10 (5)

16 (11)

0

6 (6)

7 (7)

3 12

4 (2) 33 (23)

7 (5) 56 (46)

Brain only Other only Lung + other Other combinations Subtotal Died (toxicity/other)

0 11

17

20

119

112

53 (33)

90 (70)

232 (156)

463 (387)

9

15

19

43

No relapse

323

525

1,506

2,354

Total

451

652

1,757

2,860

*Relapses prior to January 1,1985 are given in parentheses.

doses of 1,050 to 1,500 cGy. The median dose was 1,344 cGy. One patient received an additional 1,400 cGy to the area from which the metastasis was removed. The chemotherapeutic agents administered included vincristine (VCR) in 1 patient, Adriamycin (ADR) in 3 patients, VCR + actinomycin D (AMD) in 4 patients, VCR + AMD + ADR in 4 patients, and ADR + other in 1 patient. Eight patients (61%) were treated with ADR, alone or in combination. Three of the 13 patients (23%) treated with surgical excision, radiation therapy, and chemotherapy developed a second recurrence of disease. The sites were lung (same location as excised metastasis) in 1 patient, contralateral lung in 1 patient, and abdomen in 1 patient. The median survival time from relapse for the 10 surviving patients is 3.633 days (range, 1,962 to 5,986 days). Seventeen patients with stages I to III favorable histology Wilms’ tumor who developed a solitary pulmonary metastasis were treated with radiation therapy and chemotherapy, but no surgery. Fourteen patients received whole-lung irradiation with doses of 1,169 to 1,500 cGy. The median dose was 1,200 cGy. Two patients received additional radiation therapy localized to the pulmonary metastasis. Two patients received radiation therapy only to the involved lung (1,900 and 2,000 cGy). Additional radiation therapy was given to the pulmonary metastasis. The chemotherapeutic agents administered included VCR (2 patients), AMD (2) ADR (1) VCR + AMD (6) VCR + AMD + ADR (5) and ADR + cis-platinum (1). Seven patients (41%) were treated with ADR, alone or in combination. Nine of 17 patients (53%) treated with radiation therapy and chemotherapy either failed to achieve a complete response (4 patients) or developed a second recurrence of disease (5 patients). Of the four patients who failed to achieve a complete response, two had progression to bilateral pulmonary metastases, one died secondary to varicella pneumonia, and one underwent surgical excision of a persistent metastasis. Of the five patients who developed a second recurrence after achieving a complete response, two were successfully treated for a brain metastasis. Two patients developed disease recurrence in the site of the original metastasis, one despite treatment with a local radiation therapy boost. One patient recurred with bilateral pulmonary metastases. The median survival time for the 13 surviving patients is 3,115 days (range, 1,389 to 6,434 days). Five children with stages I to III favorable histology Wilms’ tumor who developed a solitary pulmonary metastasis underwent surgical excision of the lesion

730

GREEN ET AL

Table 3. Treatment

of Pulmonary Metastases

Received by Stage I to III NWTS Patients With Isolated Pulmonary Recurrence Whose Histology Was Known NW-E-1

NW-E-2

NW-E-3

Total

Favorable histology Surgery* only

1

Radiation therapyt only

5

Chemotherapy*

1

only

-

3

4

6

5

16

9

13

23

7

8

Surgery + radiation therapy

1

Surgery + chemotherapy

1

3

4

8

30

17

25

72 17

Radiation therapy + chemotherapy

-

Surgery + radiation therapy + chemotherapy

5

9

3

Other or unknown

3

0

3

6

47

44

63

154

-

-

-

-

Total Unfavorable histology Surgery* only Radiation therapyt only

1

1

1

3

Chemotherapy*

1

4

8

13

1

2

only

Surgery + radiation therapy

-

Surgery + chemotherapy Radiation therapy

t

chemotherapy

1 -

-

18

11

3

32

2

2

1

5

1

1

2

20

15

57

Surgery + radiation therapy + chemotherapy Other or unknown

-

Total

22

Unknown histology

8

1

3

12

No treatment data for relapse

2

1

1

4

79

66

82

227

Total *Surgery and chemotherapy within 30 days after diagnosis. tRadiation therapy within 60 days after diagnosis of metastases.

followed by treatment with chemotherapy. The surgical procedures performed were wedge resection (3), pneumonectomy (l), and unknown (1). Four of these five patients (80%) developed a second recurrence of disease, in the same lung from which the initial metastasis was removed. The chemotherapeutic agents administered included VCR + AMD (2 patients), VCR + ADR (l), and VCR + AMD + ADR (2). Two patients are surviving 504 and 1,627 days from the date of first recurrence. In spite of the substantial differences in the frequency of disease progression or recurrence among the three treatment groups (23% v 53% v 80%), the Table 4. Comparison of Prerelapse Characteristics

S + XRT + C

of Pulmonary

Unfavorable Histology

Metastases

XRT + C

stc

Stage

I II and Ill

3

8

1

10

9

4

11

15

4

2

2

1 0

Initial treatment 1 or 2 drugs 3 drugs

5). Fifty-five patients with stages I to III favorable histology Wilms’ tumor who developed multiple pulmonary metastases were treated with whole-lung radiation therapy and chemotherapy. The 4-year survival percentage was 44% (Table 5).

of Patients With

Favorable Histology and a Solitary Pulmonary Metastasis Treatment

statistical significance was only borderline due to the small number of patients (P = .07 for the three-way comparison and P = .05 for the pairwise comparison of 23% v 80%, both by the Fisher exact test). The 4-year survival percentage was 76% for those treated with radiation therapy and chemotherapy, compared with 77% for those treated with surgical excision, radiation therapy, and chemotherapy (Table

Months after diagnosis o-5

1

2

6-11

6

8

1

12+

6

7

4

Abbreviations: S, surgery; XRT, radiation therapy; C, chemotherapy.

The treatments for pulmonary relapse in unfavorable histology patients are summarized in Table 3. More detailed analyses were performed to evaluate the survival rate of such patients with solitary or multiple metastases. Five patients with unfavorable histology Wilms’ tumor developed a solitary lesion. Four of these were treated with surgical excision, radiation therapy, and chemotherapy, three of whom died (Table 5). Thirty-one patients with stages I to III unfavorable histology Wilms’ tumor who developed multiple pul-

PULMONARY METASTASES

731

IN WILMS’ TUMOR

Table 5. Survival Results for Patients With Relapse to Lung Only: Effect of Surgery in Those Treated Also With Radiation and Chemotherapy No. of Deaths

Extent of Lung Disease

Histology

Treatment

Solitary lesion

FH

Multiple lesions

FH

Group

No. of Patients

Solitary lesion

17

4

3.9

76%

13

3

3.1

77%

R+C

55

32

30.5

44%

4

1

2.8

75%

1

0.8

0

4

3

3.2

25%

31

28

26.0

10%

1

0

2.0

100%

104

65

61.0

22

7

11.0

R+C

Multiple lesions

R+C S+R+C

Totals

4.Year Survival

s+Fi+c

S+R+C UH

Actuarial

Expected*

R+C

S+R+C UH

Observed

R+C S+R+C

(adjusted P = .12)

Abbreviations: FH, favorable histology: UH. unfavorable histology; R, radiation therapy; C, chemotherapy; S, surgery. *“Expected”

number of deaths under null hypothesis using log-rank procedure,

monary metastases were given lung radiation therapy and chemotherapy. The 4-year survival percentage was 10% (Table 5). Comparisons

There was no statistically significant improvement in the 4-year survival percentages of patients with stage I to III, favorable histology Wilms’ tumor who developed a solitary metastasis, and did or did not have them surgically removed prior to initiation of whole-lung irradiation. There was no statistically significant difference when the analyses were stratified for treatment (P = .89) or primary tumor stage (P = S9) (Table 6). Table 6. Survival Results for Favorable Histology Patients With Solitary Pulmonary Lesion: Effect of Surgery in Those Treated Also With Radiation and Chemotherapy No. of Deaths Treatment

Group

No. of Patients

Observed

Expected+

Study NWTS-1 NWTS-2 NWTS-3

Total

R+C

8

1

1.3

S+R+C

4

1

0.7 0.5

R+ C

2

1

S+R+C

6

1

1.5

R+ C

7

2

2.0

S+R+C

3

1

1.0

R+C

17

4

3.8

S+R+C

13

3

3.2

(adjusted P = .89) Stage I II and Ill

Total

R+C

a

0

0.7

S+R+C

3

1

0.3

9

4

2.6

S+R+C

10

2

2.4

R+C

17

4

3.3

S+R+C

13

3

3.7

R+C

(adjusted P = .59) Abbreviations: R, radiation therapy; C, chemotherapy; S, surgery. *“Expected” procedure.

number of deaths under null hypothesis using log-rank

The prognosis for patients with stage I to III, unfavorable histology Wilms’ tumor who developed pulmonary metastases was poor regardless of treatment. The number of children with solitary metastases and/or whose treatment was initiated with excision of the metastases was too small to permit meaningful statistical analysis of the effect of either a solitary metastasis or surgical excision on 4-year survival probability. DISCUSSION

Wilms’ tumor is the most frequent malignant renal tumor in children and adolescents.‘4 Most patients with stage I to III, favorable histology Wilms’ tumor survive, relapse-free, following treatment with combination chemotherapy and abdominal irradiation when indicated.’ Wilms’ tumor can recur in a variety of sites including the lungs, liver, opposite kidney, bone(s), and brain.15 Review of the site of first recurrence for patients entered on NWTS-1, -2, and -3 demonstrated that the lungs were the most frequent site of first recurrence. The factors of importance in predicting survival after relapse included tumor histology (favorable v unfavorable), length of time from diagnosis to relapse (less than or more than 12 months), initial treatment with two drugs, and relapse limited to the lungs.* The distribution of these prognostic factors was examined in the patients included in the present study. More patients with stage II or III tumors were treated at relapse with surgical excision of the metastasis, followed by radiation therapy and chemotherapy. There was no statistically significant difference in the distribution of these factors between patients whose treatment did or did not include surgical excision of the lesion(s). Previous investigators have reported prolonged survival of patients with Wilms’ tumor whose only

732

GREEN ET AL

treatment for a pulmonary metastasis was surgical excision.5-1oTreatment prior to relapse for most of these children included only abdominal irradiation,5,7-‘obut two survivors were also given combination chemotherapy which included VCR, AMD, and ADR after nephrectomy.6 In the present study, the outcomes of children entered on NWTS-1 to -3 were analyzed with respect to several factors that might influence the 4-year survival of patients with stage I to III Wilms’ tumor who develop only pulmonary metastases. Surgical excision of the metastases within 30 days after diagnosis of recurrence did not improve the 4-year survival percentage of patients with favorable histology who had a solitary metastasis. The present report is a retrospective review of outcome rather than a prospective evaluation of the role of surgery in such patients. Surgical excision may have been used preferentially for patients with a larger, solitary metastasis. The data available precluded precise estimation of lesion diameter. The median time from nephrectomy to diagnosis of a solitary metastasis was not significantly different for those who did or did not undergo surgical resection, suggesting that patients with more rapidly recurrent disease were not referred selectively for surgical treatment. Others, seeking to avoid the toxicity of whole-lung irradiation, have reported successful treatment of patients who develop a solitary pulmonary metastasis with only surgical excision and chemotherapy.16 Diffuse interstitial pneumonitis occurred in 13% of NWTS-3 patients with stage IV, favorable histology tumors. The mortality rate for pneumonitis of unknown etiology was 73%. I7In addition, abnormalities of pulmonary function are well documented in longterm survivors of Wilms’ tumor whose treatment included whole-lung irradiation.1s-zo The significantly poorer outcome of the five NWTS patients treated with surgical excision and chemotherapy only, and the tendency for these patients to have recurrence in the same lung as that from which the initial lesion was removed, suggest that the risks of whole-lung irradiation are outweighed by the benefit of surviving the episode of recurrence. The results of the present study suggest that surgical excision of a solitary pulmonary metastasis

from a patient with stage I to III, favorable histology Wilms’ tumor does not improve the 4-year survival percentage of such patients. Surgical excision of a solitary metastasis should be undertaken to establish the nature of the lesion unless the diagnosis of malignancy can be established by needle biopsy or, unequivocally, by radiographic evaluation. The role of surgical excision in the management of patients with multiple metastases is not clear. Goorin et al*’ reported that 82% of patients with pulmonary metastases from osteosarcoma who had complete resection of the metastases were surviving, compared with 13% of those who did not have complete excision of all pulmonary disease. Seven of the nine surviving patients who had complete removal of metastatic disease had a solitary metastasis. Complete removal of metastatic disease was possible for only three (23%) patients with 2 to 5 metastases.” Roth et al reported that the presence of more than four metastases on preoperative whole-lung tomograms was the only factor significantly correlated with the longterm survival of osteosarcoma patients who developed pulmonary metastases, all of which were removed at thoracotomy.** The failure to demonstrate an improvement in survival percentage of patients following resection of solitary metastases from Wilms’ tumor and the data reported for patients with osteosarcoma suggest that complete surgical removal of all pulmonary metastases from patients with multiple, bilateral metastases is unlikely to improve the 4-year survival percentage of such patients and should not be used routinely prior to treatment with whole-lung irradiation and chemotherapy. The prognosis for patients with multiple putmonary metastases of unfavorable histology is poor regardless of the therapeutic strategy used. Future studies in the poor-prognosis group of patients might include evaluation of novel single agents or combinations, followed by whole-lung irradiation, with or without planned surgical excision of residual disease after an initial trial of chemotherapy. ACKNOWLEDGMENT The authors thank the many pathologists, surgeons, pediatricians, radiation therapists, and other health professionals who managed these children, without whom this study would have been impossible, and Diane Piacente for preparation of the manuscript.

REFERENCES 1. D’Angio GJ, Breslow N, Beckwith JB, et al: The treatment of Wilms’ tumor: Results of the third National Wilms’ Tumor Study. Cancer 64:349-360,1989 2. Grundy P, Breslow N, Green DM, et al: Prognostic factors for children with recurrent Wilms’ tumor: Results from the second and third National Wilms’ Tumor Study (NWTS -2 and -3). J Clin Oncol7:638-647, 1989

3. Sutow WW, Breslow NE, Palmer NF: Prognosis in children with Wilms’ tumor metastases prior to or following primary treatment: Results from the first National Wilms’ Tumor Study. Am J Clin Oncol5:339-347, 1982 4. Ballantine TVN, Wiseman NE, Filler RM: Assessment of pulmonary wedge resection for the treatment of lung metastases. J Pediatr Surg 10:671-675,1975

PULMONARY

METASTASES

IN WILMS’ TUMOR

5. Bond JV, Martin EC: Pulmonary metastases in Wilms’ tumor. Clin Radio1 27:191-195,1976 6. Willimas JA, Douglass EC, Hammond E, et al: Relapsed Wilms’ tumor. Factors affecting survival and cure. Am J Clin Oncol 8~324-328. 1985 7. White JG, Krivit W: Surgical excision of pulmonary metastases. Pediatrics 29:927-932, 1962 8. Lent MH, Staubitz WJ, Magoss IV, et al: Surgical treatment of pulmonary metastases from malignancies of the genitourinary organs. J Urol81:746-752, 1960 9. Soper RT: Management of recurrent or metastatic Wilms’ tumor. Surgery 50:555-559, 1961 10. Albers DD, Bell AH, Kalmon EH, et al: Pulmonary excision for solitary metastasis from a Wilms’ tumor with apparent cure. J Urol86:43-45, 1961 11. Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Statist Assoc 53:457-481,1958 12. Peto R, Peto J: Asymptotically efficient rank invariant test procedures. J R Statist Sot A 135:185-192,1972 13. Greenwood M: The natural duration of cancer, in Reports on Public Health and Medical Subjects 33. London, England, Her Majesty’s Stationery Office, 1926, pp l-26 14. D’Angio GJ, Beckwith JB, Breslow N, et al: Wilms’ tumor (nephroblastoma, renal embryoma), in Pizzo PS, Poplack DG (eds): Principles and Practice of Pediatric Oncology. Philadelphia, PA, Lippincott, 1989

733

15. Green DM: Diagnosis and Management of Malignant Solid Tumors in Infants and Children. Boston, MA, Martinus Nijhoff, 1985 16. Khalek ERA, Toumade MF, Lemerle J: Primary chemotherapy (PC) and primary surgery (PS) in the treatment of pulmonary metastases (PM) in Wilms’ tumor. Proc Int Sot Pediatr Oncol 13:27, 1981 17. Green DM, Finkelstein JZ, Tefft ME, et al: Diffuse interstitial pneumonitis after pulmonary irradiation for metastatic Wilms’ tumor. Cancer 63:450-453, 1989 18. Wohl MEB, Griscom NT, Traggis DG, et al: Effects of therapeutic irradiation delivered in early childhood upon subsequent lung function. Pediatrics 55:507-516, 1975 19. Littman P, Meadows AT, Pulgar G, et al: Pulmonary function in survivors of Wilms’ tumor. Cancer 37:2773-2776,1976 20. Benoist MR, Lemerle J, Jean R, et al: Effects on pulmonary function of whole lung irradiation for Wilms’ tumor in children. Thorax 37:175-180, 1982 21. Goorin AM, Delorey MJ, Lack EE, et al: Prognostic significance of complete surgical resection of pulmonary metastases in patients with osteogenic sarcoma: Analysis of 32 patients. J Clin Oncol2:425-431,1984 22. Roth JA, Putnam JB, Wesley MN, et al: Differing determinants of prognosis following resection of pulmonary metastases from osteogenic and soft tissue sarcoma patients. Cancer 2:13611366,1985