By Samir P. Kanani, MD
Associate Clinical Professor of Neurosurgery and Radiation Oncology, George Washington University, Radiation Oncology, Inova Fairfax Hospital, Falls Church, VA
Dr. Kanani reports no financial relationships relevant to this field of study.
SYNOPSIS: In this study, the authors evaluated and refined previously published risk stratification for locoregional failure (LF) by applying it to a multicenter patient cohort. The original stratification, which was developed using a single-institution series from the Hospital of the University of Pennsylvania (U Penn), produced three subgroups with significantly different LF risk based on pathologic tumor (pT) classification and the number of lymph nodes identified. This model was then applied to patients in Southwest Oncology Group (SWOG) 8710, a randomized trial of RC with or without chemotherapy. LF was defined as any pelvic failure before or within 3 months of distant failure. The authors found that patients in the U Penn cohort and the SWOG cohort had significantly different baseline characteristics. A revised stratification using pT classification, margin status, and the number of lymph nodes identified produced three subgroups with significantly different LF risk in both cohorts: low risk (≤ pT2), intermediate risk (≥ pT3 with negative margins and ≥ 10 lymph nodes identified), and high risk (≥ pT3 with positive margins or < 10 lymph nodes identified) with 5-year LF rates of 8%, 20%, and 41%, respectively, in the SWOG cohort and 8%, 19%, and 41%, respectively, in the U Penn cohort.
SOURCE: Christodouleas JP, Baumann BC, He J, Hwang WT, et al. Optimizing bladder cancer locoregional failure risk stratification after radical cystectomy using SWOG 8710. Cancer 2014;120:1272–1280.
The role of postoperative radiation therapy in muscle invasive bladder cancer after radical cystectomy (RC) and bilateral pelvic lymphadenopathy is not well established. Several organizations are considering clinical trials to assess the impact of radiation therapy after RC in high-risk patients. Validated criteria to define a high-risk population are lacking. The authors have previously reported an unvalidated LF risk-stratification model based on a database of 442 patients treated at U Penn between 1990 and 2008 with RC with or without chemotherapy. After surgery, patients were evaluated every 4 months for 2 years, every 6 months until year 5, and then annually with routine chest X-rays and bi-annual computed tomography (CT) scans or magnetic resonance imaging (MRI) of the abdomen and pelvis. This model divided patients into three statistically distinct risk groups based on two variables: pathologic tumor classification and total number of benign or malignant lymph nodes identified in the RC pathology specimen. The model indicated that patients with ≤ pT2 tumors were at low risk of LF, those with ≥ pT3 tumors who had ≥ 10 lymph nodes identified were at intermediate risk of LF, and those with ≥ pT3 tumors who had < 10 lymph nodes identified were at high risk of LF.1
The purpose of this study was to validate the previously reported risk stratification using patients enrolled in SWOG 8710, which was a randomized trial that compared RC alone versus three cycles of neoadjuvant methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC) followed by RC. This study was conducted between 1987 and 1998 and included 264 patients operated on by 106 different surgeons at 109 different institutions. The recommended evaluation after surgery was every 3 months for the first year, every 6 months for the second year, and yearly thereafter with chest X-rays. Abdominal or pelvic imaging was not required according to the protocol.2
The primary endpoint of this study was LF, but overall survival and isolated distant metastasis were also recorded. LF was defined as imaging evidence of recurrence in the pelvic soft tissues or lymph nodes below the aortic bifurcation before or within 3 months after distant metastases (DM). The goal of the external validation was to demonstrate that the original LF stratification produced significantly different subgroups when applied to the SWOG cohort. Fine and Gray regression was used to compare the incidences of LF between subgroups and to determine whether the explanatory power of the original risk stratification could be improved with additional covariates.When comparing the cohort of patients from U Penn to that of SWOG, there were differences between the two groups. The SWOG cohort was younger, with less advanced disease. The SWOG cohort was more likely to have positive or unknown margin status and fewer lymph nodes removed. Patients in the U Penn cohort were more likely to have neoadjuvant chemotherapy, while patients in the SWOG cohort all received adjuvant chemotherapy. In the U Penn cohort and the SWOG cohort, the overall 5-year LF estimates were 18% and 15%, respectively, and the 5-year isolated DM estimates were 17% and 20%, respectively.
Applying the original risk stratification to the SWOG cohort, the 5-year LF estimates were 8%, 29%, and 36% for the low-risk, intermediate-risk, and high-risk groups, respectively. The original risk stratification was not fully validated in the SWOG cohort because the risk of LF was not significantly different between the intermediate-risk and high-risk groups. Regression analysis was used to determine whether LF risk stratification in the SWOG cohort could be improved by the addition of one or more patient characteristics within the model. Of these variables, only margin status was associated significantly with LF when controlling for the original risk stratification. Modifications to the original risk-stratification model were made to include margin status. The revised risk-stratification model is as follows: low risk (≤ pT2), intermediate risk (≥ pT3 with negative margins and ≥ 10 lymph nodes identified), and high risk (≥ pT3 with positive margins or < 10 lymph nodes identified). With the revised risk stratification, the 5-year LF estimates were similar in both cohorts: 8%, 20%, and 40% for the low-risk, intermediate-risk, and high-risk groups, respectively. The 5-year OS estimates were 62%, 39%, and 7% for the low-risk, intermediate-risk, and high-risk groups, respectively, in the SWOG cohort and 60%, 31%, and 10% for the low-risk, intermediate-risk, and high-risk groups, respectively, in the U Penn cohort. The risk-stratification model failed to show a significant difference in terms of 5-year isolated DM estimates among the three risk groups. For low risk, the 5-year isolated DM estimate was about 15% for both cohorts, and the 5-year isolated DM estimates were 20-40% for intermediate and high-risk patients.
As systemic therapy continues to improve, local control is becoming increasingly important. There are multiple examples throughout oncology of this concept, and bladder cancer is no exception. Most oncologists have had the experience of caring for patients with pelvic recurrences from a gynecologic cancer, colorectal cancer, prostate cancer, or bladder cancer. Local recurrences can be significantly debilitating for our patients, particularly in the pre-sacral space. In bladder cancer, these recurrences have generally been underreported because of the high rate of distant failure. Local recurrences in the pelvis often precede distant metastasis, as evidenced from a study from MDACC,3 and locoregional recurrence was an independent variable predicting DM.
Christodouleas et al present a now validated method of identifying patients with a higher risk of recurrence within the pelvis who may benefit from additional local therapy in the form of radiation therapy. The authors should be commended for taking a previously identified model from their institution and applying it to a cohort of prospectively followed patients treated at multiple institutions making the results more applicable to the community oncologist. When the authors ran a model that previously stratified patients into three risk groups and found that it did not "work" on the SWOG cohort, they tweaked the model to include margin status, and this resulted in more robust model that could be considered generalizable. This type of research and statistical modeling is essential in the design and analysis of future clinical trials evaluating the potential benefit of adjuvant therapy. To date, the data regarding the benefit of additional local therapy in the form of radiation therapy are limited. One randomized trial from Egypt reported in 1992 demonstrated a benefit to radiation therapy by decreasing LF rates from 50% to 10%.4 This study was done prior to the widespread use of neoadjuvant or adjuvant chemotherapy and in a population where squamous cell histology is more prevalent, thus the local failure rates are higher at 50% compared to 40% in the highest-risk patients from the current study combining the U Penn database and the SWOG database. Nevertheless, the study did demonstrate that radiation therapy can reduce local failure rates to 10%. Another study from Milan comparing 130 patients treated with RC alone to a cohort of 32 patients treated with RC and ≥ 50.4 Gy of radiation therapy demonstrated a benefit to the addition of radiation therapy after RC. This retrospective study demonstrated a benefit in disease-free survival and cause specific survival in patients with pN0-Nx patients treated with adjuvant radiation therapy.
The study from Christodouleas et al published in this month’s issue of Cancer identifies a validated risk-stratification model that oncologist can use to identify patients that may benefit from additional local radiation therapy, forming a basis for additional trials seeking to improve local control. In my opinion, intermediate-risk patients (≥ pT3 with negative margins and ≥ 10 lymph nodes identified) are more likely to benefit from local therapy, as the LF rate is 20% and the 5-year survival is 30-40%. From the above-discussed retrospective studies, I would estimate that post-operative radiation therapy would decrease the recurrence rate to less than 10%. High-risk patients may benefit as well (≥ pT3 with positive margins or < 10 lymph nodes identified); however, the 5-year survival remains poor at ≤ 10% and efforts should be made to decrease the rate of distant metastasis with the use of novel chemotherapy regimens.