Radical cystectomy in the elderly

Comparison of survival between younger and older patients

Authors

  • Peter E. Clark M.D.,

    Corresponding author
    1. Department of Urology, Wake Forest University Health Sciences, Winston-Salem, North Carolina
    • Department of Urology, Wake Forest University Health Sciences, Medical Center Boulevard, Winston-Salem, NC 27157
    Search for more papers by this author
    • Fax: (336) 716-5711

  • John P. Stein M.D.,

    1. Department of Urology, University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author
  • Susan G. Groshen Ph.D.,

    1. Department of Preventive Medicine, University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author
  • Jie Cai M.S.,

    1. Department of Preventive Medicine, University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author
  • Gus Miranda,

    1. Department of Urology, University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author
  • Gary Lieskovsky M.D.,

    1. Department of Urology, University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author
  • Donald G. Skinner M.D.

    1. Department of Urology, University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, California
    Search for more papers by this author

Abstract

BACKGROUND

The authors reported their experience with radical cystectomy for transitional cell carcinoma (TCC) of the bladder. They compared pathologic features and survival rates between older and younger patients.

METHODS

The authors retrospectively reviewed the records of 1054 patients who underwent radical cystectomy for bladder TCC between 1971 and 1997. Four age groups were compared: < 60 years old at the time of cystectomy (n = 310), 60–69 years old (n = 381), 70–79 years old (n = 313), and ≥ 80 years old (n = 50).

RESULTS

There were no significant differences in pathologic features among the groups regarding frequency of carcinoma in situ, high-grade disease, p53 status, and lymph node positivity. However, in the age groups < 60, 60–69, 70–79, and ≥ 80, 45%, 49%, 51%, and 72% of patients, respectively, had extravesical TCC (P=0.004). Significant differences also were seen in the proportion of patients who received adjuvant chemotherapy (26%, 26%, 15%, and 6%, respectively; P < 0.001). The 5-year overall survival rates for patients < 60, 60–69, 70–79, and ≥ 80 years old were 72%, 58%, 54%, and 33%, respectively (P < 0.001). The 5-year disease recurrence-free survival rates were 75%, 65%, 68%, and 45%, respectively (P = 0.003). The elderly had a lower probability of receiving adjuvant chemotherapy.

CONCLUSIONS

Elderly patients undergoing cystectomy for TCC had similar pathologic features (except for disease stage) as younger patients. In the current series, elderly patients undergoing cystectomy had a higher pathologic stage and were less likely to receive adjuvant chemotherapy. The elderly hade worse disease recurrence-free survival rates. Further work is needed to identify the causes for this and to develop strategies to improve cancer control in elderly patients. Cancer 2005. © 2005 American Cancer Society.

Improvements in technology and health care have contributed to an increase in the average life span of people in the United States, leading to a progressive aging of our population.1 In 1987, 12% (or 29 million people) of the U.S. population was > 65 years old, a figure that is expected to increase to 23% by the year 2050.2 Transitional cell carcinoma (TCC) is predominantly a disease of the elderly with a peak incidence in the seventh decade of life. As a consequence, of the estimated 56,500 new cases of TCC in 2002, 40% will occur in patients > 70 years old.3, 4 Therefore, the management of TCC in the elderly is an important issue that will increase in importance over time.

Overall, 20–40% of patients with TCC will present with or develop muscle-invasive disease. In the United States, a standard of care for muscle-invasive TCC is a radical cystectomy with urinary diversion.5 In the elderly, however, some physicians have advocated the use of other, less invasive and potentially less effective forms of treatment for invasive bladder carcinoma.6 Conversely, a number of centers have shown that radical cystectomy with urinary diversion can be performed safely in carefully selected elderly patients.7–15 Only two of these series included elderly patients who had undergone diversion with a continent cutaneous reservoir,13, 15 whereas a report by Figueroa et al.13 is the only one to include patients undergoing diversion with an orthotopic neobladder. The golas of the current study are to update our experience in a rigorously defined population and to compare the outcomes between younger and older patients. In particular, we explored the influence of age on pathologic characteristics, overall survival rates, and disease recurrence-free survival rates.

MATERIALS AND METHODS

A comprehensive institutional review board-approved database of all patients undergoing radical cystectomy from August 1971 to December 1997 at our institution was reviewed retrospectively. A total of 1471 patients with records available were identified. Patients were excluded from the analysis if they received a cystectomy for a nonbladder primary malignancy (n = 72), if they had non-TCC as the primary bladder histology (n = 107), if they received salvage cystectomy after the failure of definitive radiotherapy (which was defined as ≥ 50 gray [Gy]; n = 126), if the tumor was deemed to be inoperable at the time of cystectomy (n = 23), if they had distant metastases at the time of cystectomy (n = 46), and if there was evidence after resection of macroscopic residual disease at the time of cystectomy (n = 43). The remaining 1054 patients underwent radical cystectomy with curative intent for primary TCC of the bladder. They included 843 men (80%) and 211 women (20%). The median age of the patients at receipt of the surgical procedure was 66 years (range, 22–93 years). The overall clinical outcomes of this strictly defined cohort of patients have been reported previously.16

This group of 1054 patients was categorized into 4 age groups: those < 60 years old at the time of cystectomy (n = 310 [29%]), those 60–70 years old (n = 381 [36%]), those 70–80 years old (n = 313 [30%]), and those ≥ 80 years old (n = 50 [5%]).

All patients considered for radical cystectomy underwent a rigorous preoperative physical examination and screening to ensure adequate cardiac and physical performance status. Screening included routine preoperative studies such as a comprehensive metabolic panel, complete blood count, chest X-ray, electrocardiogram, and abdominal and pelvic computed tomography (CT) scans.

During the last 30 years, the approach to neoadjuvant therapy for invasive TCC has been constantly evolving. For example, from 1971 to 1978, 97 patients from this cohort received short-course/high-dose neoadjuvant radiotherapy (16 Gy) over 4 days immediately before their cystectomy. An analysis of the outcomes of these patients has shown this to be unnecessary and neoadjuvant radiotherapy has not been applied routinely since 1979.17

Radical cystectomy with en bloc bilateral pelvic lymphadenectomy was performed in a standard fashion as previously described.16 The only alteration in technique among some elderly patients was a more limited lymph node dissection (which was left to the discretion of the surgeon performing the procedure) for patients with significant arteriosclerosis of their iliac arteries. All patients underwent urinary diversion in a standard fashion and routine placement of an open gastrostomy tube at the time of cystectomy. A previous report from this institution of 709 patients has shown this procedure to be safe with a low complication rate (< 0.05%) and greatly facilitates patient comfort and pulmonary toilet.18

Patients were followed at 4-month intervals for the first year, at 6-month intervals the second year, and annually thereafter. Routine studies included a thorough history and physical examination and serum studies, including the monitoring of electrolyte and creatinine levels and liver function tests. Chest X-rays and radiographic evaluation of the reservoir and upper tracts via intravenous pyelography, ultrasound, and/or pouch/cystogram as appropriate were done 4 months after surgery and yearly thereafter. Bone scans or CT scans were performed only as clinically indicated. The median follow-up for the entire group was 10.2 years. The maximum follow-up was 27 years and 91% of the patients had ≥ 3 years of follow-up.

The primary end points of the study included disease recurrence-free and overall survival. Three pathologic subgroups were defined based on the 1997 modification of the TNM staging system. Organ-confined disease included patients with a T2 or lower classification of disease (T0, Tis, Ta, T1, or T2) and negative lymph nodes. Extravesical disease was defined as T3 or T4 disease with negative lymph nodes and the final group comprised patients with positive lymph nodes regardless of the T-classification.

Two clinical outcomes, disease recurrence-free survival and overall survival, were analyzed. Time to clinical disease recurrence was measured from the date of cystectomy to the time of the first documented disease recurrence. Patients without evidence of disease recurrence were censored at the time of last follow-up. Survival was measured from the date of cystectomy to the date of death. All deaths were counted as events irrespective of the cause. Surviving patients were censored at the time of the last follow-up.

Kaplan–Meier plots were used to estimate the probabilities of overall survival and disease recurrence-free survival for every year since radical cystectomy.19 The log-rank tests (overall20 and stratified21) were used to compare the differences in survival or disease recurrence in subgroups. The Cox regression method was used to estimate relative risks and 95% confidence intervals (CI) for disease disease recurrence among different age groups.22, 23 In this cohort, stage was the variable most strongly associated with outcomes. After stratifying by stage, only age and adjuvant chemotherapy were associated with the outcomes. Hence, a final model including stage and adjuvant chemotherapy was used to evaluate the association between age and outcomes. The Pearson chi-square test was used to examine the association between important clinical variables. All P values reported in the analyses are two-sided values.

RESULTS

When evaluating pathologic parameters, no significant difference was found among the 4 age groups with respect to the proportion of patients with high-grade disease (≥ Grade 3),24 carcinoma in situ, or positive lymph nodes after the final pathologic evaluation (Table 1). A significant association was noted between age and the proportion of patients with extravesical extension of disease (P = 0.004).

Table 1. Pathologic Parameters Stratified by Age Group
Age group (yrs)Carcinoma in situHigh grade disease (%)Stage pT3/4 (%)Lymph node positive
< 6053834523
60–6961884923
70–7958865123
80 +56927226
P value0.200.120.0040.97

The application of neoadjuvant/adjuvant therapies is summarized in Table 2. Although the rate of extravesical disease after the final pathologic evaluation was significantly higher in the older patients, 26%, 26%, 15%, and 6% of patients in the 4 age groups, respectively (P < 0.001), received adjuvant chemotherapy. This suggests a number of elderly patients (i.e., patients ≥ 80 years old) with extravesical extension of TCC and/or positive lymph nodes did not receive adjuvant chemotherapy.

Table 2. Distribution of Neoadjuvant/Adjuvant Therapy by Age Group
Age group (yrs)Neoadjuvant chemotherapy (%)Neoadjuvant radiation (%)Adjuvant chemotherapy (%)Adjuvant radiation (%)
< 60717263
60–69512262
70–7967153
80 +4660
P value0.69< 0.001< 0.0010.43

Actuarial overall survival is shown in Figure 1. The 5-year actuarial overall survival rates for patients < 60, 60–69, 70–79, and > 80 years old were 72%, 58%, 54%, and 33%, respectively (P < 0.001). The actuarial disease disease recurrence-free survival is shown in Figure 2. The 5-year disease recurrence-free survival rates decrease as age increases (rates of 75%, 65%, 68%, and 55%, respectively; P = 0.003).

Figure 1.

Actuarial overall survival stratified by patient age at the time of radical cystectomy (< 60 years old, 60–69 years old, 70–79 years old, and > 80 years old).

Figure 2.

Actuarial disease recurrence-free survival stratified by patient age at the time of radical cystectomy (< 60 years old, 60–69 years old, 70–79 years old, and > 80 years old).

The relative risk (RR) of disease recurrence comparing the four age groups stratified by pathologic subgroup at cystectomy (organ-confined vs. extravesical disease vs. lymph node-positive disease) is outlined in Table 3. The RR is consistently higher in older patients within all 3 pathologic subgroups, with patients > 80 having the highest risk in 2 of the 3 subgroups (P = 0.003). When the RR of disease recurrence was stratified further by both stage and whether the patients received adjuvant chemotherapy, the pattern of risk increasing with age remained but was no longer as clear-cut (Table 4, P = 0.006). However, the RR for disease recurrence remained highest in patients > 80 years old in 5 of the 6 strata. The RR for disease recurrence was also higher for patients > 80 when controlling for the decade of surgery (the 1970s vs. the 1980s vs. the 1990s) and pathologic subgroup (RR = 2.15, P < 0.001).

Table 3. Relative Risk And 95% Confidence Intervals of Developing Disease Recurrence Based on Age at the Time of Cystectomy Stratified by Pathologic Stagea
Stage< 60 yrs60–69 yrs70–79 yrs80 + yrs
  • a

    n = total number of patients in the group.

  • OC: organ confined disease; EV: extravesical disease, LN positive: lymph node positive disease.

  • P value for association of age with time to disease recurrence, after controlling for stage: P = 0.003.

OC11.95 (1.14–3.33)1.62 (0.90–2.90)1.86 (0.56–6.32)
 (n = 182)(n = 219)(n = 174)(n = 19)
EV11.14 (0.64–2.01)1.19 (0.66–2.14)2.06 (0.91–4.71)
 (n = 55)(n = 75)(n = 66)(n = 18)
LN positive disease11.28 (0.86–1.95)1.11 (0.71–1.72)3.85 (1.85–8.70)
 (n = 72)(n = 87)(n = 73)(n = 13)
Stratified by stage11.41 (1.07–1.87)1.26 (0.93–1.70)2.60 (1.57–4.32)
Table 4. Relative Risk and 95% Confidence Intervals of Recurrence by Age Group Stratified by Pathologic Stage and by Whether the Patient Received Adjuvant Chemotherapya
Stage< 60 yrs60–69 yrs70–79 yrs80 + yrs
Chemotx no
  • a

    n = total number of patients in the group.

  • Chemotx: adjuvant chemotherapy; OC: organ confined disease; EV: extravesical disease, LN positive: lymph node positive disease.

  • P value for association of age with time to recurrence, after controlling for stage and treatment: P = 0.006

OC12.22 (1.22–4.06)1.88 (0.99–3.60)2.40 (0.69–8.34)
 (n = 159)(n = 194)(n = 165)(n = 18)
EV11.09 (0.51–2.34)1.09 (0.52–2.29)1.72 (0.66–4.45)
 (n = 31)(n = 46)(n = 51)(n = 17)
LN +11.34 (0.69–2.63)1.50 (0.78–2.90)2.24 (0.81–6.24)
 (n = 27)(n = 31)(n = 37)(n = 10)
Chemotx yes
OC11.24 (0.38–4.04)0.98 (0.19–5.03)— (—)
 (n = 23)(n = 25)(n = 9)(n = 1)
EV11.17 (0.49–2.77)1.34 (0.47–3.75)8.31 (0.91–75.96)
 (n = 24)(n = 29)(n = 15)(n = 1)
LN +11.27 (0.76–2.13)0.83 (0.45–1.51)12.18 (3.16–46.71)
 (n = 45)(n = 56)(n = 36)(n = 3)
Stratified by stage and treatment11.43 (1.08–1.89)1.26 (0.93–1.70)2.40 (1.43–4.02)

DISCUSSION

The overall survival and disease recurrence-free survival among the older patients in the current series was worse than in the younger patients. Although differences in overall survival among the four age groups are explained readily by higher comorbidity rates in the elderly, the differences in disease recurrence-free survival are not as easily explained. This does not seem to be accounted for by any difference in grade, carcinoma in situ, or lymph node positivity, between younger versus older patients. There were, however, a significantly higher proportion of older patients with extravesical disease at the time of cystectomy. Despite the higher rate of extravesical disease, older patients were significantly less likely to receive adjuvant chemotherapy. Presumably this is due, at least in part, to a perception that elderly patients may not tolerate adjuvant chemotherapy as well as younger patients. It is tempting to explain the worse disease recurrence-free survival of the elderly, then, by higher stage at cystectomy, possibly due to a delay in undergoing surgery, in combination with a reluctance to treat the elderly with chemotherapy. There was a trend toward a higher RR of developing disease recurrence with advancing age when stratified by either stage or by stage and adjuvant chemotherapy (Tables 3 and 4). However, this was not statistically significant. It remains possible that with more numbers, this trend would become significant, suggesting that another factor also may be influencing the results.

We believe that there may be a delay in the time from diagnosis of invasive TCC to the time of referral of the patient to our institution for definitive management. It has been shown in a series by Hautmann et al.25 that offering an orthotopic diversion shortens the time to cystectomy and appears to improve survival. Is it possible that part of the worse disease recurrence-free survival in the elderly seen in the current series is due to a delay in the time to cystectomy? Unfortunately, we do not have data available on the date of diagnosis of invasive TCC. As a result, we were unable to determine whether there was a significant delay in the time from invasive TCC diagnosis to cystectomy in the elderly population.

An association between age and tumor biology is another possible explanation for the differences noted in disease recurrence-free survival. Whether tumors in the elderly are in some way inherently more aggressive than those in younger patients is controversial. Several studies, including ours, have suggested that elderly patients present with more advanced bladder carcinoma than younger patients.3, 26, 27 Autopsy series, however, suggest that bladder carcinoma and other malignancies metastasize at the same rate irrespective of age.28–31 The differences between these studies may be due to a number of factors, including variable access to health care or differences in physician management between age groups.3 Furthermore, we found no difference in the p53 molecular analysis among our patient age groups, which argues against a difference in the biology of tumors stratified by age.

The influence of age on survival after radical cystectomy is also controversial. A report by Thrasher et al.32 studied 531 patients with invasive TCC treated with radical cystectomy and found that age < 65 years old was an independent predictor of better survival. Similarly, a report by Takashi et al.33 found that in a multivariate analysis of 264 patients, age was an independent predictor of survival, but that stage, size of the tumor, and the presence of irritative voiding symptoms were more important.

Narayana et al.34 found that age was an independent predictor of survival, but only in patients without a previous history of bladder carcinoma (P = 0.047). When all patients were considered, the impact of age was not significant (P = 0.083). Other studies have not shown age to be an independent predictor of survival.35, 36

Similar discrepancies exist in the influence of age on outcome after chemotherapy. For example, at least two randomized trials comparing chemotherapy in either an adjuvant setting or for metastatic disease suggested that older patients did better than younger patients,37, 38 whereas at least one study investigating neoadjuvant therapy found no influence of age on outcome.39 All of these studies suffer from the potential confounding influence of the physician's perception in treatment selection of the elderly. Thus, in cystectomy series, there may be a tendency toward delaying surgery for the elderly, whereas, in chemotherapy trials, there may be a trend toward selecting only the healthiest elderly patients to be entered into the trial, thus biasing the results.3

A number of investigators have examined the morbidity of cystectomy in the elderly.7–15 In general, these reports have shown that radical cystectomy can be performed safely in properly selected elderly patients. We are currently analyzing our results from this same cohort of patients and comparing clinical outcomes (including complication rates) between younger and older patients. These results will be published separately. Although the mortality rates across series have ranged from 0% to 18%, the overall mortality rate is 4%.7–15 In the current series, the mortality rate was 0–4% for all 4 age groups (data not shown), which is consistent with other reports in the literature. An in-depth analysis of these mortality rates is part of the clinical outcomes analysis that is currently ongoing and will be published separately.

Whereas the current analysis represents a large cohort of > 1000 patients with long-term follow-up, there are limitations. This was a retrospective review and, as such, is subject to the problem of selection bias. In an attempt to overcome this, we have stratified our results, especially with respect to disease recurrence-free survival, by variables that may have an impact on outcome. Nevertheless, as outlined previously, we cannot exclude the possibility that other factors may have influenced the outcomes we observed. In addition, although this is a large cohort of patients, there are only 50 patients > 80. This represents one of the largest experiences in the literature, yet this is still a relatively small number of patients from a statistical standpoint. This limited our ability to simultaneously account for multiple variables when analyzing outcomes. This was most notable when investigating disease recurrence-free survival across the four age groups stratified by both stage and whether the patient received adjuvant chemotherapy. We believe, therefore, that although most of the differences in disease recurrence-free survival are due to differences in the rate of extravesical disease and the use of adjuvant chemotherapy, we cannot rule out the possibility that other factors may influence the outcomes in the elderly.

Because of the inherent problems with patient selection bias in any retrospective review, including the current study, more prospective work is needed to examine the impact of age on tumor biology. Although in the current analysis, we were unable to detect any differences in disease grade or p53 status between younger and older patients, it remains possible that more sophisticated molecular profiling of bladder TCC may demonstrate differences based on the age of the patient. A careful analysis of other factors not addressed in the current study, such as the time from invasive TCC diagnosis to the time of cystectomy, and the possible impact of these variables on outcomes in the elderly is needed. More work is necessary to improve chemotherapeutic regimens such that more elderly patients, especially those with worse performance status, will be able to undergo adjuvant therapy for higher-risk disease. Finally, studies examining the impact of age and medical comorbidities on overall, disease recurrence-free, and cancer-specific survival are warranted.

To conclude, elderly patients undergoing cystectomy for TCC have similar pathologic features (except for disease stage) as younger patients. Elderly patients undergoing cystectomy have higher pathologic stage and yet are less likely to receive adjuvant chemotherapy. The elderly have worse disease recurrence-free survival. Further work is needed to identify the causes for this and to develop strategies to improve cancer control in the elderly.

Ancillary