Indian Journal of Cancer, Vol. 42, No. 4, October-December, 2005, pp. 173-177
Pattern of metastases in renal cell carcinoma: A single institution study
Sivaramakrishna B, Gupta NarmadaP, Wadhwa Pankaj, Hemal AshokK, Dogra PremN, Seth Amlesh, Aron Monish, Kumar Rajeev
Department of Urology, All India Institute of Medical Sciences, New Delhi
Code Number: cn05031
AbstractBackground: Increasing numbers of patients with renal cell carcinoma (RCC) are incidentally detected and can be potentially cured by surgery alone. In treating metastatic RCC, worthwhile survival rates are achieved in cases of low burden recurrences. This necessitates a rational follow up protocol, which picks up early recurrences and avoids costly surveillance for those with a favorable prognosis.
Aims: We studied the patterns of metastases occurring in patients operated for localized or locally advanced renal cell carcinoma in the Indian setting and try to evolve a suitable follow up protocol.
Setting and Design: Institution based, retrospective data.
Method and Materials: Records of patients from January 1988 to December 2003, operated for initially localized RCC were reviewed. Follow up was performed using an established protocol. Occurrence of metastases and their patterns were studied.
Statistical analysis used: Comparison of the different survival times was performed using the one-way analysis method. Multiple comparisons (post hoc test) were performed using the Bonferroni method.
Result: Follow up was available on 209 patients. Mean survival was 43.75 months (SD ± 28.72). Thirty-nine patients developed 59 metastases. Lungs were the commonest site of metastases (37%), followed by bone (22%), liver (19%) and brain (8%). Relapse and survival showed significant correlation with pathological stage (p < 0.001), with higher stage being associated with greater relapses and lesser survival. There was no correlation between site of recurrence and stage of disease.
Conclusions: Occurrence of metastases correlate with the pathological stage of the disease at primary presentation. Tailored, stage-based follow up protocols allow adequate surveillance for disease activity and progression without escalating the overall costs.
Keywords: Renal cancer, metastases, survival, follow up, protocol
Renal cell carcinoma (RCC) is known to have a myriad pattern of presentation and variable clinical course. Due to widespread application of imaging, patients of RCC are now more often incidentally detected, have low stage disease and can be cured by surgery alone. Although improvements have been made in treating metastatic disease with immunotherapy, the results are at best modest. However, 5 year survival rates up to 75% have been reported following metastatectomy in select cases. This necessitates a rational follow up protocol to be able to detect early recurrences or metastases, albeit avoid the burden of intensive surveillance on those patients with a favorable prognosis and unlikely recurrences. Recently, several groups had analyzed relapse patterns in RCC after nephrectomy and proposed specific guidelines on surveillance. ,,,, However, Indian data in this regard is lacking.
We studied the pattern of metastases in our patients operated for localized renal cell carcinoma with an aim to evolve a protocol for follow up for such patients with non-metastatic disease at presentation [UICC 1997, stage I to stage IV (M0 disease)].
Material and MethodsRecords of patients operated for non-metastatic RCC from January 1988 to December 2003, obtained from our outpatient follow up clinic, maintained in a hand written filing system with subsequent computed data base filing were reviewed by a single observer. The patients were staged according to the UICC 1997 staging.
We had followed an established protocol of reviewing patients three monthly for the first year, six monthly in the second year and yearly thereafter. At each visit the patient had been subjected to a clinical evaluation, laboratory investigations which included urine routine and microscopy, renal and liver functions tests, including transaminases and alkaline phosphatase levels. A skiagram of the chest and an ultrasonography (USG) of the abdomen were also performed in all cases. Computerized tomography (CT) and bone scan had been performed as indicated by a specific localizing symptom or abnormality in the aforementioned investigations. Time to metastasis had been recorded as the time of detection of the first metastases in a given patient. Metastatic patterns, time to occurrence, and TNM staging (UICC 1997, stage I, II, III, and M0 stage IV) at presentation were noted.
Statistical methodStatistical analysis was performed using the SPSS statistical package (version 7, SPSS Inc, Chicago, Illinois). The comparison of the different survival times was performed using the one-way analysis method (anova). Multiple comparisons ( post hoc test) were also carried out to find the significant pair contributing toward overall significance by the Bonferroni method. Kaplan-Meier plot was constructed to analyze the time to metastases. The different survival curves were compared using the log rank test. The overall significant was observed to be significant if ′ P ′ value was 0.05.
Three hundred and forty three patients with localized RCC were operated upon, during this period. Of these, follow up was available on 209 patients, which included 162 men and 47 women. The mean age at presentation was 53 years. The patient characteristics and stage-wise distribution of patients with relapse are provided in [Table - 1]. Cases were histologically classified as 174 (83.2%) clear cell RCC, 24 (11.4%) chromophobe and 11 (5.2%) papillary RCC. Overall mean survival was 43.75 (± SD 28.72) months. The overall survival in stage I was significantly greater as compared to other stages. Values (P < 0.001). Similarly, significantly greater survival was noted in stage II in comparison to stage III or IV ( P < 0.001). However no significant difference was noted in the survival between stage III and IV ( P = 1.0) [Table - 2] and [Figure - 1].
During this period, a total of 39 patients (18.6%) developed 59 metastases at various sites [Table - 3]. Ten of the 39 patients developed multiple metastases (one in stage II, five in stage III and four in stage IV). The lung was the commonest site of metastases (37%), followed by skeletal system (22%), liver (19%), and brain (8%). Relapse after nephrectomy was associated with pathological stage, with higher stage disease presenting with more relapses ( P < 0.001). However, there was no significant correlation between site of metastases and interval to metastases ( P = 1.0). The temporal profile of metastases is depicted in [Figure - 2]. All patients with recurrences had clear cell carcinoma on reviewing their original histology, except one who had papillary RCC.
One patient developed multiple lung metastases at 14 months and another developed brain metastases at 178 months. Both patients were symptomatic for the same and diagnosis was confirmed on an X-ray chest and CT brain, respectively.
Characteristics of relapses for stage II
Characteristics of relapses for stage III
Characteristics of relapses for stage IV
The current paucity of efficacious treatment modalities for metastatic RCC provides the need for having a stringent follow-up for patients with localized disease after surgical extirpation of the primary, to enable additional therapy/ surgery when the recurrence load is low. A suitable follow up protocol design should be such so as to allow early detection of recurrences, yet at the same time promote patient compliance and not place a significant financial burden upon the patient. Follow up protocols suggested earlier advised intense surveillance guidelines for all patients irrespective of prognosis. It has since become apparent that pathological stage is the best predictor for RCC,, a finding reaffirmed in our study. Newer prognostic indicators include nuclear grading, histological subtype, functional status (Eastern Cooperative Oncology Group [ECOG], and molecular markers (Ki-67, silver staining nucleolar organizing regions, proliferating cell nuclear antigens, and carbonic anhydrase expression). ,, Although several authors have included a combination of such newer prognostic factors to structure outcome prediction models and postoperative nomograms, these have not been currently included in the clinical TNM classification., Although such prognostic models, becoming more inclusive and complex may eventually replace or find place as addition to the existing TNM classification, we chose to utilize the currently existing practical tool to evaluate our data.
Although RCC can metastasize to unusual sites, the commonly favored include the lung, bone, liver, abdomen, and the retroperitoneum in decreasing order of frequency., A similar spectrum of metastatic pattern was noted in our study; however, no correlative pattern was noted to occur with the stage of disease.
Apart from the sites of metastases, the time to metastases is an important consideration in deciding a follow-up protocol. This has been found to correlate with the tumor stage. ,,,, We noted infrequent metastases in stage I disease, whereas in stage II disease nearly 50% metastases had occurred by 24 months. In stage III and IV most metastases had occurred by 12 and nine months, respectively. Overall, 87% (51 metastases) of metastases had occurred before 36 months. Similar overall recurrence has been reported by Sandock et al . Parienty et al. reported that 82% of patients with local recurrences were symptomatic; in contrast others have found lower proportions (31% to 66%) of symptomatic patients. In our study 86% (33 patients) of the patients with metastatic disease were symptomatic for relapse. Similarly, Sandock et al found 73% of their patients with pulmonary metastases and 92% of abdominal metastases were symptomatic at presentation and suggested a protocol that relied on the detection of metastases based on symptoms. However, in a symptom driven protocol, the tumor burden at detection is likely to be large, potentially precluding a surgical resection and have only modest responses to biological therapy.,,
While a CT scan is suggested as the investigation of choice for abdominal imaging in most recently proposed protocols, ,,,, USG of the abdomen has also been suggested as an alternative for abdomen surveillance. Routine screening with CT entails a greater expense than an ultrasonogram of the abdomen, and hence we prefer the latter option. None of the patients in stage I were noted to have recurrent disease in the abdomen, thus a routine abdominal surveillance may not be cost-effective in these patients. This observation has been similarly noted in other recent studies.,,
An X-ray of the chest along with clinical evaluation has been found to detect 100% pulmonary metastases. All recent protocols suggest an X-ray chest as the standard screening modality for pulmonary metastases. ,,,, Our observations and recommendations are similar to those made by others.
We noted that bone and brain metastases were in an advanced stage at presentation. Though metastases at these sites are usually unresectable, they are symptomatic and merit palliation. In view of the same, we suggest that in these instances investigations should be symptom specific instead of routine screening, as suggested earlier also., Based on our findings we have evolved a follow up protocol for operated patients of RCC with localized disease at presentation [Table - 4].
Evaluation of the temporal occurrence of metastases has shown that all metastases except one occurred before 48 months. Therefore, radiological surveillance should be maintained till at least 5 years. Though rigid radiological surveillance beyond five years may not be cost-effective, patients should be kept on a yearly follow-up with clinical evaluation as late metastases can occur. An omission in our study is that of non-inclusion of the histological grade of the clear cell RCC. Our currently existent databank is not reliably or completely able to provide this information. Inspite of this, the suggested follow-up protocol is likely to serve as an adequate, yet cost effective method for postoperative surveillance.
ConclusionsThe occurrence of metastases in the follow up of RCC correlates with the pathological stage of disease at presentation and is representative of the biological behavior of the disease. Predominant majority of metastases develop within 5 years of treatment of the primary. "Tailored" stage-specific follow up protocols are likely to allow early detection of recurrences at a more affordable overall cost, without loss of surveillance benefits/ efficacy. We propose the ultrasonogram of the abdomen as a less expensive option, to CT scan as the investigation of choice for abdominal surveillance, with the CT being used in cases with equivocal or suspicious ultrasonogram findings.
Copyright 2005 - Indian Journal of Cancer
The following images related to this document are available:
Photo images[cn05031t4.jpg] [cn05031f2.jpg] [cn05031t3.jpg] [cn05031t2.jpg] [cn05031t1.jpg] [cn05031f1.jpg]