Profile and Clinical Outcome of Children with Wilms' Tumor treated at a Tertiary Care Centre, India

Introduction

Wilms' tumor (WT) is the most common genitourinary malignancy in the pediatric age group. The WT affects approximately one child per 10,000 worldwide before the age of 15 years. Incidence rates are slightly elevated for the US and African Blacks in comparison with Whites but are only half as compared with Asians.1 The mean age of presentation is 3 to 4 years. The male to female ratio in unilateral Wilms' tumor is 0.92:1.00 and for bilateral cases is 0.60:1.00.2 Most patients are asymptomatic wherein abdominal lump is an incidental finding, however, patients may also present with abdominal pain, hematuria, and fever. Patients often have underlying congenital anomalies associated with such as Wilms' tumor, aniridia, genitourinary malformations, and mental retardation (WAGR), Denys–Drash, and Beckwith–Wiedemann syndrome.3 In a developing country, the patients usually present late with advanced disease.4 The National Wilms' Tumour Study Group (NWTSG) and the International Society of Pediatric Oncology (SIOP) have laid down structured protocols for the management of these tumors that include chemotherapy, surgery, and radiation.5 The main difference between the two groups is that the SIOP insists on neoadjuvant chemotherapy prior to nephrectomy as a rule. The advantages of giving neoadjuvant chemotherapy include 1) reduction in tumor mass, 2) preventing spillage, 3) decreasing chances of distant spread. The multimodality approach has indeed improved the prognosis in the past few decades. The clinical outcome however depends on the stage and the biology of the disease and the economic setting.6 We at our center follow the SIOP protocol 9301.7 We conducted this study to analyze the outcome of patients treated at our center from January 2015 to December 2019.

Materials and Methods

Results

Patient characteristics and management: Twenty-eight (n = 28) patients were analyzed in this study. The median age at presentation was 31 months (range: 4–96 months); the sex ratio (male/female) was 1.5. Also, 42.9% of the patients were < 24 months old, whereas 57.1% of them were ≥ 24 months. No associated congenital syndrome was found. The median duration of symptoms was 2 months (range: 1–8 months). The median time to diagnosis was 10 (2–22) days from the first visit at AIIMS, Patna. Other clinical characteristics are listed in Tables 1 and 2. Clinical features were dominated by abdominal lump (50%), followed by abdominal lump with fever (35.7%), vomiting (7.1%), and hematuria (7.1%). The diagnosis was based mainly on clinical, histopathological, and radiological evidence. Core needle biopsy, CT scan of the abdomen and thorax were done in all cases. Right-sided and left-sided disease were seen in 39.2% and 57.1%, respectively. Only one patient was found to have bilateral disease.

A CT scan was used for the assessment of response. The median size of the sum of diameters at diagnosis was 27 cm (14.1–46 cm). The median volume of the disease at diagnosis was 359.48 mL (52.67–1805.76). Renal vein thrombosis was found in 32.1% of cases, infiltration of the renal pelvis was found in 35.7% of patients, and locoregional lymph node involvement was seen in 42.9% of patients. Liver metastasis was found in 10.7% (3) patients, i.e., among them two patients with lung metastasis and one patient with only liver metastasis. CT scan thorax showed 10.7% (3) patients with lung metastasis, i.e., among them two patients with liver metastasis and one patient with only lung metastasis at diagnosis. There was an overlap of two patients with both liver and lung metastasis. Radiological staging workup showed that 7.1% in stage I, 39.3% in stage II, 39.3% in stage III, 10.7% in stage IV, and 3.6% in stage V.

All patients received preoperative chemotherapy according to the SIOP protocol 9301. Also, 89.3% of patients received regimen “VA (vincristine 1.5 mg/m² intravenous bolus at weeks 1, 2, 3, and 4; actinomycin-D 45 µg/kg intravenous bolus at weeks 1 and 3),” and 10.7% received regimen “VAA” (vincristine 1.5 mg/m² intravenous bolus at weeks 1, 2, 3, 4, 5, and 6; actinomycin-D 45 μg /kg intravenous bolus at weeks 1, 3, and 5; doxorubicin 50 mg/m² intravenous over 4 hours at weeks 1, 5, and 9). Preoperative chemotherapy included weekly chemotherapy for 4 weeks followed by surgical intervention, week 5 chemotherapy was given when planned surgery falls after week 5.

Preoperative chemotherapy toxicity (grade III) was seen in 25% (7) of patients. The most common toxicity was grade III diarrhea, seen in 10.7% of patients. Grade III neutropenia and hepatic toxicity were seen in 7.14% and 7.14% of patients, respectively (Table 3). Only one patient received the “VAA” regimen among the patients who developed grade III toxicity and the rest six were received the “VA” regimen. Dose reduction was done in 39.3% (11) of patients. Eight out of 11 patients were under 12 months of age group (dose reduction of all drugs of the chemotherapy regimen by 50% was done from week 1 and onward),8 further dose reduction (by 50%) in the remaining three patients was done due to chemotherapy-related grade III toxicities. Toxicity-related delay in the chemotherapy schedule was a median of 9 days.

Following neoadjuvant chemotherapy, all patients underwent radiological assessment using a CT scan (RECIST 1.1). The partial response was observed in 35.7% of patients, while 64.3% of patients showed stable disease. None of the patients showed a complete response. More than 50% of the tumor volume reduction was seen in 71.4% of patients, while 28.6% of patients showed < 50% of tumor mass reduction.

Twenty-four (85.7%) patients underwent surgical intervention. Surgical treatment consisted of radical nephrectomy in all cases. Histological subtypes and tumor stage were defined according to pathological assessment, which revealed 7.1% low risk, 67.9% intermediate-risk, and 25% high risk, according to the SIOP WT 2001 staging criteria for renal tumors of childhood.11 On histopathological subtype classification, mesoblastic nephroma, favorable histology, and diffuse anaplasia were 7.1%, 67.9%, and 25%, respectively. Postoperative staging showed 17.9% in stage I, 46.4% in stage II, 21.4% in stage III, 10.7% in stage IV, and 3.6% in stage V. Postoperative chemotherapy was indicated in 75% of cases. Postoperative radiotherapy was indicated in 28.6% of patients. The median time to start radiotherapy following surgical intervention was 18 (9–25) days. Radiotherapy was delivered with a daily fraction of 1.5 Gy (for whole abdomen irradiation) and 1.8 Gy (for flank irradiation). The dose of radiotherapy ranged from 10.8 Gy to 21.6 Gy. Flank radiotherapy was given to patients with stage II high risk (two patients), stage III intermediate risk (three patients), and stage III high risk (one patient). Whole abdomen radiotherapy was given to stage III intermediate risk with per operative bleeding (one patient), and stage III high risk with tumor spillage during surgery (one patient).

Correlation (bivariate) studies: Recurrence correlated statistically significantly with age < 24 months (p = 0.049), locoregional lymph nodes (p = 0.008), histopathological subtypes (p < 0.001), stage of the disease (p = 0.003), and risk groups (p < 0.001). There was statistically significant reduction in the tumor volume following neoadjuvant chemotherapy while comparing the means of the tumor volume (p < 0.001). Preoperative and postoperative comparisons of stages using paired t-test revealed that there was a significant down staging following chemotherapy (p = 0.018).

Univariate analysis using Cox regression model showed that DFS was statistically significantly dependent on locoregional lymph nodes (p = 0.018; confidence interval [CI], 1.562–111.854); percentage tumor volume shrinkage > 50% (p = 0.043; CI, 1.052–22.374); histopathological subtype of tumor (p = 0.045; CI, 1.024–7.449), risk group (p = 0.008; CI, 2.147–146.540), and postoperative stage (p = 0.002; CI, 1.981–20.828). Multivariate analysis using Cox regression model showed that postoperative stage (p = 0.019; CI, 1.344–26.823) was the only independent factor for DFS (Table 4).

Survival analysis: At the time of analysis, 25% of patients developed recurrence during the median follow-up of 25 months. The median DFS was not reached, the mean DFS was 48 months (95% CI; 36.892–61.038). The median OS was also not reached, the mean OS was 59.13 months (95% CI: 49.738–68.527), shown in Figs. 1 and 2. The 1- and 3-year DFS were 100% and 64.1%, respectively. Three-year DFS of stages I and II was 100%, 40% in stage III, 0% in stages IV and V (Fig. 3). Three-year DFS was 100%, 89%, and 33% for low risk, intermediate risk, and high risk, respectively (Fig. 4). The 1- and 3-year OS was 100% and 75% respectively. Three-year OS of stages I and II was 100%, 60% in stage III, 0% in stages IV and V.

Close

Fig. 1: The median disease-free survival (DFS).

Export to PPT

Close

Fig. 2: The median overall survival (OS).

Export to PPT

Close

Fig. 3: DFS was significantly differing according to risk group (Log rank Mantel–Cox; p < 0.001).

Export to PPT

Close

Fig. 4: DFS was significantly differing according to stage of the disease (Log rank Mantel–Cox; p < 0.001).

Export to PPT

Discussion

This analysis provides an outcome of WT cases at a tertiary care center. The total number of patients enrolled during this period was 28. All were histologically confirmed cases of WT. Bilateral WT was seen only in one patient. The age and sex distribution of this study were similar to another large Indian study.12 The male to female ratio was 1.5:1, which was similar to other studies.10 In our study, the most frequent presenting complaint was abdominal mass (50%), and 35% of the patients presented with an abdominal lump with fever. Western studies reported ∼74% of patients presenting with an abdominal mass.13 A study by Guruprasad et al reported that 90% of their patients presented with abdominal mass,12 which was almost similar to our findings. Only 17.9% of patients were in stage I in our study. This is in contrast to other larger studies showing a larger number of patients in the early stages.14 This may be due to the late presentation of the majority of the patients possible reason includes poor access to the health care facility. These advanced stage presentations of our study patients caused most of them in upfront inoperable disease. Our cohort showed 25% of patients with high-risk histopathological features, while in other Indian studies this varied from 7% to 37.5%. The high-risk histopathological features in Western studies ranged from 7% to 28.8% (Table 5).10121415161718 Total surgical excision is the standard of the treatment of unilateral WT.19 In this study, we adopted the SIOP study group protocol, which included neoadjuvant chemotherapy followed by surgical intervention. Most of our patients presented with advanced disease with a median tumor volume of 359.48 mL and 37.5% of patients had renal pelvis involvement as interpolated from the radiological information. Delay in presentation and advanced disease at diagnosis SIOP study group approach seemed to be more feasible for most of the patients presented to us.

Preoperative chemotherapy was mainly associated with hematological and gastrointestinal toxicity. In this study, we observed grade III neutropenia in 7.14% of patients and grade III diarrheas in 10.7% of patients. Israels et al in their study reported 27% of grade III neutropenia.20 This low incidence of neutropenia in our cohort of patients may be due to reduction of dose by 50% in patients with age < 12 months.21 Preoperative response to neoadjuvant chemotherapy in this study showed partial response in 35.7% and stable disease in 64.5% of patients. In our study, 71.4% of patients showed > 50% reduction in the tumor mass. Other studies reported > 50% tumor mass in 82.7% of cases.18 This difference in response may be due to dose reduction in patients with age < 12 months (9 of 28 patients).

We observed in this study that there was significant downstaging following neoadjuvant chemotherapy. Twenty-five percent of our patients had anaplastic histology. The presence of anaplasia had a significant impact on the DFS and OS and this finding was supported by larger studies including the NWTS group and SIOP.22232425 In this study, recurrence was statistically significantly found to be associated with age > 24 months, the presence of locoregional lymph nodes, histopathological subtypes, risk groups, and stage of the disease. The finding of our study is supported by larger studies NWTS-3.26 Univariate analysis showed tumor volume shrinkage > 50%, histopathological subtypes, risk group, and postoperative stage were the dependent risk factors for DFS, and the postoperative stage was the only independent risk factor of DFS on multivariate analysis. The present study showed a 3-year DFS of 100% in stages I and II and 40% in stage III. Findings in the SIOP-9 study showed that almost 70% of DFS in stage III,27 this difference might be due to a small number of patients in our study, genetic or ethnic factors, and health care access facility or delay in diagnosis due to late presentation at the health care facility.

Conclusion

Presentation of the pediatric WT dominated with advanced stage probably due to difficulty in access of health care facility or ignorance by parents. This advanced stage leads to a large voluminous tumor and renders most of the cases upfront inoperable. Neoadjuvant chemotherapy followed by surgical intervention seems to be a good and feasible option for these subsets of patients with comparable outcomes and toxicities in comparison with western studies.