Despite regimens that allow for organ preservation in selected patients, surgical resection followed by postoperative RT has become the standard of care for stage III and IV OCSCC [19–21]. Even with this dual therapy, local or regional disease recurrence and low survival remain an concern . Patients with adverse features of ECS, LVI, and PNI are at particular risk of disease progression and higher death rates . Promising results emerged from the use of various combinations of postoperative CRT in randomized studies from 1970–1990. [24, 25] Further advancement in OCSCC treatment didn’t come until 2004 with the RTOG 9501 and EORTC 22931 trials which suggested that improved disease-free survival in addition to local and regional control can be achieved with the addition of adjuvant chemotherapy to postoperative RT. The EORTC trial showed that the addition of adjuvant chemotherapy allowed for significantly increased rates of local control, disease-specific survival, and overall survival, without high incidences of late adverse effects. While improved disease-free survival, local or regional recurrence, and rates of distant metastases were shown in the RTOG trial, overall survival was not found to be superior. These studies, while being landmark trials with level I evidence, were not site-specific for oral cavity tumours and had marked differences in selection criteria . A substantially higher proportion of patients in the RTOG trial had N2-3 disease. Both studies also did not account for confounding variables when calculating survival curves. These factors in combination with a review of the literature in 2005 for postoperative CRT confirmed the need for more evidence for adjuvant CRT therapy . The current study sought to confirm the findings of the RTOG and EORTC studies, while being site specific for OCSCC.
Overall survival for stage III and IV OCSCC was found to be the highest in the S-CRT group (Figure 1). Those treated with S-RT were 2.2 times more likely to die than those treated with S-CRT and 5.3 times more likely to die secondary to disease specific causes than the S-CRT group (p < 0.05). Furthermore, patients in the S-RT were 3.4 more likely to experience disease recurrence than those treated with S-CRT (p < 0.05). This is supported by the notably lower overall loco-regional recurrence rates found in the S-CRT arm. 5-year overall survival rates (58%) for the S-CRT and the S-RT group (40%) were compble to those of the EORTC study. While disease-free survival at 2-years for patients in the S-CRT (76%) group was similar to the findings in the RTOG trial, the S-RT (52%) group did not fare as well. This discrepancy could be attributed to varied sensitivity of different head and neck sites to radiotherapy that was not septed in the RTOG trial .
2- and 5- year metastases-free survival showed that the S-CRT group had the highest rates at 87% and 84% respectively (Figure 1). Patients receiving S-RT were 3.6 times more likely to develop metastasis compared to S-CRT. These findings concur with the RTOG and EORTC trials in that the addition of chemotherapy to adjuvant RT improves distant metastatic control [3, 4]. The 2- and 5- year overall survival for patients with ECS positive pathology was significantly higher (p < 0.05) in patients treated with S-CRT, (both at 87%) than with S-RT, found to be 47 and 32% respectively (Figure 2). PNI and LVI positive patients however did not achieve statistically significant differences in overall or disease-free survival. The S-CRT group disease-free survival at 2- and 5-year intervals was highest in the ECS group at 93 and 93%, (p < 0.05) compared to S-RT which was found to be 41 and 35% respectively. While disease-free survival was not septed by ECS status in the EORTC and RTOG trials, it was shown to significantly improve with postoperative CRT, which is congruent with our findings [3, 4]. Patients with ECS also had improved disease-specific and metastasis-free survival in the S-CRT group (p < 0.05). ECS positive patients therefore seem to benefit more in overall, disease-free, disease-specific, and metastasis-free survival when treated with adjuvant CRT than with adjuvant radiotherapy alone. These results compare favorably with the results of the EORTC and RTOG trials where ECS was the only adverse feature that showed improvement in overall survival with S-CRT in both studies .
Standards of care are ideally based on the highest level of evidence available, which in many cases consists of randomized control trials (RCT). RCTs typically have stringent enrolment criteria and as a result, physicians must be careful in applying results to their practice. The effects of these studies are then important to be examined on the population they are applied to. Population-based studies, which are known for their external validity can then be used to externally validate an RCT . This study used some prospectively collected cancer registry data combined with a retrospective review of charts to determine survival rates. Not all variables were controlled in this study owing to its population-based design. Table 1 shows that treatment groups were not completely balanced. Older patients and those with higher CCI scores tended to receive RT treatments, and those with more advanced disease tended to have received S-CRT or S-RT. These factors were accounted for as possible confounding variables in a vigorous multivariate Cox regression analysis.
Treatment protocols on CRT in addition to adjuvant CRT or RT were not clearly defined at either the Cross Cancer Institute or the Tom Baker Cancer Center. Although this implies a potential selection bias, it also establishes a need for determining optimal therapy protocols for patients requiring CRT as well as postoperative CRT or RT [28, 29].
The RTOG and EROTC trials relied on Kaplan-Meier analyses between treated groups but failed to account for possible confounding variables . It is well known that age, gender, TNM staging, and even performance scores have an effect on survival [30–35]. Because of this, these variables were accounted for in a multivariate Cox-regression analysis. Age adjusted CCI was also included in the multivariate analysis as age and co-morbidities have been shown to affect risk of mortality . Covariates known to influence survival were included, and statistical significance was achieved.
The results of this study support the RTOG and EORTC landmark trials. The addition of postoperative chemotherapy to S-RT for advanced OCSCC appears to improve overall survival by 22% and 18% at 2 and 5 years post-treatment, respectively. Disease free and metastases free survival also benefitted from the chemotherapy addition. This suggestion mirrors previous evidence that adjuvant CRT may prevent distant spread of disease [3, 4, 20, 21]. While ideally an RCT specific for advanced stage OCSCC should be performed, this study serves as the first population based study confirming the survival advantage of triple modality therapy treatment for advanced stage squamous cell carcinoma of the oral cavity.