Sporadic renal cell carcinoma in young and elderly patients: are there different clinicopathological features and disease specific survival rates?

Renal cell carcinoma (RCC) is the most common renal parenchymal malignancy and represents 3% – 6% of all adult malignancies [1]. Patients are generally older than 40 years at diagnosis and the disease occurs predominantly in the seventh decade of life [2]. Only 3 to 7% of all sporadic RCC patients have been reported to be below 40 years [3‐7].

In other tumor entities it has been noted that younger adults often have a less favorable prognosis than elderly patients. Solid organ malignancies including those of breast, colorectal and prostate in young patients have been already investigated [8‐11] and these tumors seem to have a more pejorative survival prognosis. There have been anecdotal reports that renal masses may behave in a more biologically aggressive manner with poorer outcomes in younger patients [3, 6]. To date, the rare studies on early onset RCC have been mostly descriptive with a small number of patients and a relatively short follow-up (table 1) [3, 5, 12‐19]. Some studies lack a group of regular onset patients for comparison at all [12, 13]. Many data are conflicting however and study designs vary considerably as well. In analogy to these conflicting data, major differences in methodology are found in the data published to date. In the present retrospective study we compare clinicopathological features and cancer specific survival of patients with confirmed RCC aged 45 years or younger with patients 75 years or older in univariate and multivariate analysis. By setting a minimum of 30 years between the groups we hoped to increase the discrimination and to shed further light on the clinical behavior of young versus old patients with RCC.

Between 1992 and 2005 a total of 1042 patients underwent partial or radical nephrectomy for renal masses at the Department of Urology of the University of Regensburg. The medical and histopathological data of 70 YP (range: 26 to 45 years, mean: 39 years) and 150 OP (range: 75 to 91 years, mean: 80 years) suffering from RCC were reviewed We did not include middle-aged patients (>45 years and < 75 years) to avoid masking any potential differences in tumor biology between young and old patients. All familial RCC syndromes including von Hippel-Lindau disease were excluded. Pathological evaluation was reviewed and tumor stage was adapted according to the 2002 TNM classification. Patients were evaluated preoperatively by physical examination, laboratory studies, ultrasonography and radiographic staging including chest x-ray and computerized tomography. Staging lymph nodes dissection was performed for all patients. The postoperative outcomes and duration of follow-up were compiled by chart review and patient interview. Complete follow-up information was available for all patients through a department registry. Statistical analysis were completed using SPSS version 12.0 (SPSS, Chicago, IL). Contingency table analysis and two-sided Fisher's exact tests were used to study the statistical association between the two groups. Significant differences concerning tumor progression and cancer specific survival were calculated using the Kaplan-Meier method and log rank test. For cancer related death, patients were surveyed at the time of their last clinical follow-up appointment or at their date of death not related to the tumor. Deaths from causes other than RCC were censored. Uni- and multivariate Cox proportional hazards models were performed to evaluate the relationship between cancer specific survival, tumor progression, TNM pathologic stage, histologic subtypes, tumor grade, incidental or symptomatic disease and, in addition to other variables, age (separated into YP and OP). P values < 0.05 were considered significant.

Just 3.4% to 7.5% of all RCC patients have been reported to be young adults [7]. This is reflected in our series, as 7.2% of 968 RCC patients were aged ≤45 years. It has been suspected that young RCC is characterized by different clinicopathological and biologic character. However previous studies are somewhat contradictory and feature various methodologies. Some previous studies lack a pronounced span of age between young patients and the comparison group. While 20 years were chosen by Gillett [16], Sanchez-Ortiz chose 18 years [18], but even less has been employed [5, 15, 17, 19]. To compare YP and OP we have determined a minimum of 30 years separating both groups. We thus hoped to increase the discrimination between young and old patients, as some overlap might occur blurring the picture. This approach seems to be backed up by epidemiological data suggesting a significantly lower incidence in patients aged up to 45 years [2].

OP were a mean 8 years above the major age of regular onset, which has been reported to be roughly 70 [2]. Thus while it has to be taken into consideration, that OP might somewhat differ from regular onset patients, the present series offers a clear separation between OP and YP. Analysis of our population confirmed the usual prognostic factors for RCC: pTNM tumor stage, tumor grade, tumor diameter, N and M stage and circumstances of renal cell carcinoma discovery. Contrary to expectations, symptoms that may be considered as substitutes for performances status were not retained as independent prognostic factors in our multivariate analysis. Moreover, recent studies considering this parameter did not show any difference according to the age at diagnosis [16, 18]. Controversial data exist on the histopathological character of younger patients. In the present series YP clearly demonstrated lower T stages and lower tumor grade (table 2), while other studies only showed lower T stages in YP [6, 16, 18, 19]. Schiff et al. compared patients younger than 40 years old to patients older than 40 years. Their data differed with the most contemporary series in that more patients in the older group presented with T1 tumors [5].

In contrast to our series a former study of wide span comparison of RCC patients < 40 years and >80 years could not figure out any differences in tumor grade [14]. A variation of distribution of histological subtypes has been reported by Gillett et al. In their to date largest series of 107 younger cases chromophobe RCC were found in 13.1% vs. 3.6% [16]. Also Rodriguez et al. found out differences in histological subtypes, OP significantly more often showed clear cell RCC than YP (91.0% vs. 69.0%, p < 0.001) [6]. In accordance with previous reports [16‐19] in our series however there was an even distribution of subtypes among YP and OP. Previous authors have suggested that younger adults may have more aggressive renal tumors with a poorer prognosis than older patients [3, 6]. YP tended to improved cancer specific survival in the present series. 5-year survival rate was 95.2% in YP and 72.3% in OP (p = 0.009). This finding supports previous reports [5, 15], also in non comparative studies young patients showed similar survival rates [3, 12, 13], and is in analogy to favourable histopathology in YP. The less advanced forms of RCC found in YP may explain the better 5- year cancer specific survival. Several earlier studies [5, 6] had reported comparable survival rates, despite their higher percentages of localised forms in young patients, suggesting that early-onset RCC forms might be more aggressive. However, those studies had short follow-up [17] or had small population size [5], hereby rendering interpretation of their results more difficult. In the present series, young age was an independent predictor of improved cancer specific survival. Siemer et al. found young patients to present with symptoms more often. While in their series young RCC was still related to favourable outcome, symptoms caused by malign renal masses were associated with poor prognosis [19, 21]. In our series YP did not complain of symptoms more often. In synopsis the factors that determine the improved survival outcome in YP with RCC have yet to be clearly defined.

We do not believe in profoundly different tumor characters per se. This notion is supported by a significant number of young patients presenting with advanced disease and tumor related death in this group. However we suspect a reduced momentum of tumor biology in young patients. We hypothesize a better tumor control, which might be due to a superior immunology in this age group slowing the tumors in growth.

To conclude RCC in younger and elder patients, as defined by a minimum of 30 years as discriminatory span, young patients present with favourable clinicopathology and an improved cancer specific 5-year survival rate after adjusting for clinical and pathological variables on multivariate analysis. These obvious differences contribute to the trend toward improve outcome in younger patients for which their young age was an independent prognostic factor. Our data refute past reports of poorer cancer specific survival in younger patients with RCC.