Outcome of Children with Stage IV Wilms Tumor - Our Experience of 15 Years.

CONTEXT: Stage IV Wilms tumor is associated with poor prognosis, and recent changes in management have been suggested based on genetic markers and response to chemotherapy in this subgroup of patients.
OBJECTIVE: The objective was to evaluate the outcomes of children with Stage IV Wilms tumor who were managed with the AIIMS-WT-99 protocol.
MATERIALS AND METHODS: All the children with Stage IV Wilms tumor who were managed by us from October 2000 to December 2012 were included in the study. All the patients who had received primary treatment elsewhere were excluded from the study. All patients were managed as per the AIIMS-WT-99 protocol. After appropriate investigations, tumors that were deemed resectable underwent an upfront surgery. Unresectable and inoperable tumors received chemotherapy after cytological confirmation of the diagnosis. Chemotherapy was administered as per the NWTS-5 study. Pulmonary and flank radiotherapy was advised to all patients. Patients with poor response to chemotherapy or with recurrence were managed with an alternative chemotherapy regimen. The outcomes that were assessed the 4-year overall survival (OS) and the 4-year event-free survival (EFS). STATISTICAL ANALYSIS USED: Kaplan-Meier survival estimates.
RESULTS: During the study period, 219 patients with Wilms tumor were treated. Of these, 36 (16.4%) had Stage IV disease, and they formed the study group. The 4-year OS was 48% with a mean survival time of 59 months limited to 115 months (95% confidence interval: 41.3-75.9 months). The 4-year EFS was 42.4%. Patients with liver metastases had a poor outcome, whereas patients with good response to chemotherapy had a good outcome.
CONCLUSION: Stage IV Wilms had a poor prognosis, and the survival rates in the index study are lower than those quoted in the literature. Although the exact reason for this poor result eludes us, these patients may benefit from the intensification of chemotherapy. Copyright:

INTRODUCTION

Wilms tumor is the most common renal tumor in children with a reported incidence of 7.1 cases per million in children younger than 15 years. The use of multimodality treatment has achieved excellent survival rates. Efforts are now being made to decrease the treatment-related morbidity and mortality while maintaining or improving the outcome.

The presence of distant metastases is an independent poor prognostic factor in patients with Wilms tumor. The reported 4-year recurrence-free survival (RFS) for metastatic Wilms tumor is reported to be 76% as compared to 85% in localized favorable histology tumors.[1] Recently, in the management of Stage IV Wilms tumor, the omission of radiotherapy has been proposed if there is a complete response (CR) of the metastatic lesions to the neoadjuvant chemotherapy, provided the genetic markers which are known to be associated with poor prognosis (Loss of heterozygosity (LOH) at 1p and 16q) are absent.[2] Conversely, intensification of therapy has been proposed in patients in whom the metastatic lesions persist despite the neoadjuvant chemotherapy.[3] Due to these recent recommendations, we felt the need to evaluate our results. We conducted the study to evaluate the outcomes of children with Stage IV Wilms tumor who were managed with the AIIMS-WT-99 protocol, which was prior to these recommendations.

MATERIALS AND METHODS

A retrospective study was conducted, and all the children with Stage IV Wilms tumor who were managed by us from October 2000 to December 2012 were included in the study. All the patients who had received primary treatment elsewhere were excluded from the study. Ethics approval was obtained from the Institute Ethics committee (IESC-T-252-01.06.2012).

Management: All patients were managed as per the AIIMS-WT-99 protocol. All patients with a suspected Wilms tumor had been investigated with abdominal ultrasonography, including a Doppler study and contrast-enhanced computed tomogram of the chest and the abdomen. Tumors that were deemed resectable clinically and radiologically underwent an upfront nephroureterectomy with lymph node sampling. Unresectable and inoperable tumors underwent a fine-needle aspiration cytology to confirm the diagnosis. After confirmation of the diagnosis of Wilms tumor, these patients received neoadjuvant chemotherapy with vincristine, dactinomycin, and adriamycin as per the DD-4A regimen of the NWTS-5 study.[1] The tumor response was assessed clinically and radiologically after 6 weeks of neoadjuvant chemotherapy. Patients with an adequate response to chemotherapy underwent surgery. Adjuvant chemotherapy after the surgery was dictated by the histopathological examination (HPE) of the tumor specimen. Patients with favorable histology tumors were continued on the DD4A regimen, whereas in the tumors with anaplasia, the chemotherapy was intensified to regimen I.[1234]

Patients with poor response to chemotherapy or with recurrence were managed with an alternative chemotherapy regimen. The alternate chemotherapy consisted of ifosfamide, cisplatin, and etoposide regimen administered every 3 weeks for six courses.[567] The further treatment was planned according to the response to alternate chemotherapy.

Pulmonary (12 Gy) and flank radiotherapy (10.8 Gy) was advised to all patients. Patients with tumor spill received whole abdomen irradiation in addition to flank irradiation. Pulmonary nodules that persisted after chemotherapy and radiotherapy underwent metastasectomy if feasible.

Clinical and radiological follow-ups were performed 3 months for the 1st year after the completion of treatment, 6 months for the next 2 years, and yearly after that.

Outcome: The outcomes that were assessed the 4-year overall survival (OS) and the 4-year event-free survival (EFS) using Kaplan–Meier survival estimates. The OS was calculated from the date of diagnosis till death, while the EFS was calculated from the date of diagnosis till the first event. The events were defined as death, recurrence, or disease progression.

RESULTS

During the study period, 219 patients with Wilms tumor were treated. Of these, 36 (16.4%) had Stage IV disease, and they formed the study group. The mean age at presentation was 56.2 months (range: 24–108 months). There was a male preponderance in the study group with 23 males and 13 females (M:F = 1.76:1). The HPE of the tumors in these children revealed favorable histology in 33 (92%) and diffuse anaplasia in 1 (2.7%) patient, respectively. The HPE was not available in 2 (5.4%) patients, as they had not been operated. The follow-up ranged from 2 to 115 months with a mean follow-up of 61.1 months.

Location of the metastases

The metastases were limited only to lungs in 28 (78%) patients. In 3 (8%) patients, the metastases were confined to the liver, whereas the remaining 5 (14%) patients had metastases in both the liver and lungs.

Surgical treatment

The treatment and outcome of these patients are presented in Figure 1. Among the nine patients who underwent upfront surgery, six patients died because of recurrent or progressive pulmonary disease. Of the 25 patients who were operated following neoadjuvant chemotherapy, ten died, and of these, two had a disease-free status and died because of other reasons. In the same group, 15 children were alive, but of these, three children with progressive disease or recurrence discontinued treatment and were alive at the last follow-up.

Figure 1

A flowchart depicting the interventions and the outcome of the patients included in the study

Metastatectomy

In the present study, only pulmonary metastasectomy had been performed in five patients. Unilateral metastasectomy was performed in three patients and serial bilateral metastasectomy in two patients. Of these five patients, three children had a disease-free survival (DFS). The biopsy was available in only two of these three patients with DFS, and it showed an absent or a nonviable tumor on HPE of the tumor. Two patients who died had been operated for the recurrent pulmonary metastases, and both of them had viable tumor cells on HPE.

Chemotherapy

Initial chemotherapy consisting vincristine, adriamycin, and dactinomycin was administered to all patients. In 16 patients (44%), alternate chemotherapy was started. In seven patients, it was started due to poor response to the initial chemotherapy, and in the remaining 9, because of the recurrent disease.

Radiotherapy

In our study, only 29 (80%) patients had received radiotherapy (RT). Seven patients did not receive radiotherapy, as six patients died before RT could be initiated, and one patient discontinued treatment. Of the patients who received RT, 23 (80%) received both local and lung RT, three to local, lung and liver, and three to local area only.

The outcome in relation to the site of metastases and response to chemotherapy: overall, 16/28 (57%) of these patients became disease free after the completion of therapy. Eight patients had liver metastases, of which seven died because of progressive or recurrent disease, and only 1 DFS (12.5%) could be achieved. The outcome in children with only lung metastases is summarized in Figure 2.

Figure 2

Flowchart of outcome in children with lung only metastases

The outcome in relation to the response of metastatic lesions to chemotherapy is shown in Figure 3. The lesions which remained stable or progressed despite chemotherapy had the worst prognosis, and only 12.5% of these children achieved DFS. In contrast, 60%–73% of children with a good or partial response to chemotherapy achieved DFS, respectively.

Figure 3

Flowchart showing the outcome in relation to the response of metastasis lesions to chemotherapy. (DF: Disease free)

Events

Events were documented in 21 (58.3%) of 36 patients. The first event to occur in these patients was a recurrence in 12 children, progressive disease in eight children, and death in one patient. Local recurrence with or without pulmonary or hepatic recurrence was noted in nine of these 12 patients. Two patients had recurrence limited to the lung, whereas the remaining one patient had recurrence limited to the liver. These patients had been managed with alternate chemotherapy with additional RT if indicated. In patients who developed local recurrence, only one patient had a tumor spill. In three patients, HPE of the lymph nodes sampled during the tumor removal had shown tumor metastases. All these patients had received adequate local RT as per protocol. Of these 12 patients with recurrence, there were only 1 DFS, whereas three children discontinued treatment. One patient became disease free, but died later due to sepsis secondary to a dental abscess. The remaining seven patients died because of progressive disease. All eight patients with progressive disease (without having achieved CR) died. Of the 21 patients who had an event, 17 died, three discontinued treatment, and only one had a DFS.

Survival estimates

The 4-year OS was 48% with a mean survival time of 59 months limited to 115 months (95% confidence interval: 41.3–75.9 months). [Figure 4a] The 4-year EFS was 42.4%, with a mean survival time of 50.49 months limited to 115 months (95% confidence interval: 33.4–67.6 months) [Figure 4b].

Figure 4

(a) Kaplan–Meir survival estimate of the 4-year overall survival of Stage IV disease. (b) Survival estimates of 4-year event-free survival of Stage IV disease. (MST: Mean survival time, LT: Limited to, CI: Confidence interval)

DISCUSSION

Wilms tumor is an excellent example of pediatric solid organ malignancy, wherein the use of multimodality treatment has achieved satisfactory outcomes. The 5-year survival of favorable histology Wilms tumors is reported to be ≥90%.[8] Encouraged by these results, there is an attempt to limit the therapy in patients with good outcomes to decrease the treatment-related morbidity. Conversely, intensification of therapy is proposed in children with poor prognostic factors to improve the outcomes.[23]

The SIOP/GPOH studies have reported a 5-year OS of 76.3%, whereas NWTS has reported a 5-year EFS of 75%.[910] However, in our study, despite adequate treatment, a 4-year OS of 48% and a 4-year EFS of 42% only could be achieved. The survival in the current study is inferior as compared to those reported in the literature. Hence, we analyzed all the variables that affected the outcome of our Wilms tumor patients.

Wilms tumor is a progressive disease, so children who present late are likely to present with larger tumors and metastases. The mean age of the presentation in our study was 56.2 months and was similar to 57.7 months reported from SIOP.[11] Another reason for poor results can be attributed to poor tumor biology in our patients. In the index study, no genetic evaluation was performed, so the status of LOH at 1p and 16q is not known. Only one patient had diffuse anaplasia, and focal anaplasia was not seen in any patient. Thus, most of our patients had favorable histology as per NWTS/COG guidelines.

Response to chemotherapy is also a prognostic factor, and children who have complete resolution of metastases at 6 weeks of chemotherapy have a better outcome, as is evident in the current study. In the index study, there was a complete resolution of metastases in 42% of children, which is similar to 39%, as reported in COG.[12] SIOP studies have reported a higher incidence of complete resolution (60.9%) of metastases on chemotherapy alone.[11] With a comparable rate of complete resolution of metastases, poor response to chemotherapy is not a likely cause for our results. The role of metastasectomy is still not clear; a recent study from SIOP reported that complete resolution of pulmonary metastases could be achieved in 84% of children with the combined use of chemotherapy and metastasectomy.[13] We were unable to analyze how many children in our study who had an incomplete response to chemotherapy were amenable to metastasectomy to achieve a CR in the lungs.

Radiotherapy, if indicated, is recommended and usually administered within 10 days of surgery in COG studies. In our patients, there was a significant delay in receiving radiotherapy, and most patients RT after 4–6 weeks of surgery. Although Kalapurakal et al. reported that the delay in radiotherapy does not affect local recurrence rates, most of the children in their study RT between 8 and12 days after the surgery.[14] Hence, their study was unable to test for a meaningful difference in recurrence rates.

The effect of the site of metastases on the prognosis is still controversial. Berger et al. reported that liver metastases, only when associated with pulmonary metastases, had a poor outcome.[15] Ehrlich et al. reported that liver metastases, whether or not associated with lung metastases do not affect the outcome.[10] In our study, patients with liver metastases, irrespective of whether or not associated with lung metastases had a poor outcome with seven deaths in eight patients.

It can be argued that our poor results may be due ours being a developing country with inherent problems such as malnutrition and later presentation with a large tumor mass. However, a recent study from South Africa reported a 5-year OS of 58.5% and 5-year EFS of 54% in metastatic Wilms tumor, which, even though less compared to those reported from developed countries, is still better than our results.[16]

Another reason for our poorer results is the dismal rates of salvage in children who had progressive or recurrent disease. Recurrence has been reported to occur in approximately 10%–20% of the cases and depends on the stage and the histology of the tumor.[51718] A recent study reported a recurrence rate of 22.6% in the patients with Stage IV Wilms tumor.[15] The recurrence rate in our study was high at 30%. Moreover, of the 11 patients in our study who presented with recurrence, we were able to achieve DFS in only one patient. Most studies have reported a salvage rate of 40%–60%.[51920] In the index study, our patients had a higher recurrence rate with an abysmal salvage rate among those with recurrence. We have noticed that our patients tolerated salvage chemotherapy regimens poorly and frequently defaulted on treatment. This may be partly explained as patients come from faraway places, and the treatment requires them to visit the hospital frequently. Although the treatment per se is free of cost, these frequent visits to the hospital and absence from work put a significant financial burden on the parents. Therefore, after recurrence, when the treatment schedule and the prognosis are explained, the parents lose hope early and default on treatment. In the index study, 3 of 11 patients with recurrence refused further treatment. We promote the interaction of these parents with our other survivors to improve their morale and increase treatment compliance.

The results of the AREN0533 study conducted by Children's Oncology Group were recently published.[3] In patients without LOH and the lung nodule CR who received DD4A chemotherapy only without RT, the authors noted a decrease in 4-year EFS from 85% to 79.5%, with a statistically insignificant but borderline value of P = 0.52. However, the 4-year OS was 96.1%, due to high salvage rates in patients who had a recurrence. The same study had shown significant improvement in EFS in patients with lung metastases with an incomplete response by the addition of four cycles of cyclophosphamide and etoposide. In our scenario, where our salvage rates are still poor, the omission of radiotherapy in patients who have completed the resolution of pulmonary metastases on initial chemotherapy may not be justified. Intensification of chemotherapy in patients who show persistent pulmonary metastases may help us to improve our outcomes in the future.

CONCLUSION

Stage IV Wilms had a poor prognosis with 4-year EFS of 42.9% and 4-year OS of 48%. Patients with good response of metastases to chemotherapy have a good prognosis, whereas patients with liver metastases have an inferior prognosis. Recurrence and progressive disease occurred in more than half of the patients who could not be salvaged. Although the exact reason for this poor result eludes us, these patients may benefit from intensification of chemotherapy, especially in patients who did not have a CR of metastases after the initial 6 weeks of chemotherapy.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.