Question: Does treatment effect survival of local prostate cancer?
Eric W. Terman
October 7, 1997
Clinical Bottom Lines:
1. Grade III prostate cancer uniformly has a worse outcome, presumably because it is a more aggressive disease.
2. There are no randomized controlled trials comparing treatments.
3. If you are going to treat, prostatectomy looks better at this time. (Please note #1 in comments)
Methods/Evidence:
1. Retrospective study extracted from the SEER cooperative data. SEER is a collaboration that collects all cancer cases in good detail in Connecticut, Hawaii, New Mexico, Iowa, Utah, San Fran, Oakland, Detroit, Atlanta and Seattle.
2. Survival in each subgroup was compared to an age matched cohort.
3. Final analysis was done using an intention to treat model.
| Intention to Treat | Treatment Received | ARR | ||||
n |
%survival |
n |
%Survival |
Prostatectomy vs other | ||
| Grade I | ||||||
| Prostatectomy | 3854 | 94 (91-95) | 3402 | 98 (97-99) | ||
| Radiation Therapy | 4065 | 90 (87-92) | 4188 | 89 (87-92) | 4% prob non sig | |
| Conservative | 9804 | 93 (91-94) | 10133 | 92 (90-93) | 1% prob non sig | |
| Grade II | ||||||
| Prostatectomy | 14287 | 87 (85-89) | 12922 | 91 (89-93) | ||
| Radiation Therapy | 7939 | 76 (72-79) | 8456 | 74 (71-77) | 11% | |
| Conservative | 6198 | 77 (74-80) | 7046 | 76 (73-78) | 10% | |
| Grade III | ||||||
| Prostatectomy | 5133 | 67 (62-71) | 4154 | 76 (71-80) | ||
| Radiation Therapy | 2596 | 53 (47-58) | 2977 | 52 (46-57) | 14% | |
| Conservative | 2236 | 45 (40-51) | 2834 | 43 (38-48) | 22% |
Comments:
Problems:
1. The prostatectomy arm had the advantage of finding nodes that had metastatic disease which could elude standard clinical detection. Thereby the conservative and XRT therapy groups could easily contain more advanced disease.
2. Surgery also allowed more accurate staging in terms of the grade of tumor.
3. The prostatectomy group was on average 5 years younger than the other groups, so survival would be different.
4. The study is retrospective, which can allow multiple biases when churning the data.
5. Patients who died of other causes were dropped out of the data. This is a serious problem.
6. New XRT technology allows increased intensity with little morbidity making past data out of date.
7. Particularly in low grade tumors, quality of life must be balanced by actual amount of life saved.
Good aspects:
1. It’s the only data that we have. This helps in discussing the issues with a patient in terms of survival.
2. There seems to be little bias by the investigators.
Reference:
Lu-Yao, G and Yao, SL Population-based study of long-term survival in patients with clinically localised prostate cancer. Lancet. 1997 349(March 29):906-910.