According to criteria proposed by the World Health Organization or RECIST criteria,the evaluation of tumor response is based on changes in tumor size [22, 23]. However, it is well known to have limitations in the cases with tumors that have obscure margins or scar tissue after treatment and in its utility to assess in-field recurrence. PERCIST—that is, RECIST using 18F-FDG PET༌is considered to overcome such limitations as metabolic changes are closely related to malignant potential of tumors and thus may be the most accurate noninvasive imaging modality for initial staging and response assessment in lung cancer, PERCIST has recently been proposed as a standardized method for evaluation of metabolic tumor response .
In present study, we performed a retrospective review of the 49 patients treated at our institution to compare the PERCIST and RECIST methods for evaluation of therapeutic response to SBRT in patients with early-stage NSCLC. Our data showed a significant difference in the results of response classification using RECIST and PERCIST. PERCIST was found to be the strongest independent predictor of outcomes in patients with early NSCLC receiving SBRT. To our knowledge, this is the first report demonstrating the correlation between therapeutic responses and prognosis in early NSCLC receiving SBRT compared PERCIST and RECIST.
SBRT has emerged as a way to reduce treatment volumes and to facilitate hypofractionation with delivery of large daily tumor doses. Prior to the wide acceptance of SBRT, conventionally fractionated radiation therapy was one of the available options for this patient cohort and provides benefit with median survival times about 1.5 years . Recently, increasing evidence has shown SBRT has become the preferred treatment option with a 70% five-year primary tumor control rate in patients who are medically inoperable or decline surgery [2–6]. In our series of 49 patients treated with SBRT, the 3-year overall survival was 79.6% with a median survival time of 30 months. This is consistent with the results of previous prospective multicenter trials.
Surveillance follow-up after treatment is very important for early detection of locoregional recurrence and metastasis. The Radiation Therapy Oncology Group (RTOG) 0236 trial followed patients with computed tomography (CT) every 3 months during the first 2 years after treatment, then every 6 months for 2 more years . Despite this protocol, many physicians in clinical practice utilize FDG PET-CT for surveillance follow up. Recently, there have been several reports [16–20] on the utility of response evaluation of radiotherapy with 18F-FDG PET in early-stage NSCLC. FDG PET-CT, with 72-94% sensitive and 77-92% specific for evaluating malignancy in early-stage NSCLC, is superior to CT scan for detecting mediastinal metastasis with a sensitivity and specificity of 77% and 86%, respectively, despite the concern for increased FDG activity in areas of inflammation.
In our present study, there were no detected regional recurrences when follow-up evaluation was limited to CT alone. Three malignancies in lung nodules with 20% increase in tumor diameter were evaluated as PD by RECIST criteria. However, these nodules showed no metabolic activity by PET-CT. The addition of PET aided in the detection of 1 regional recurrence and 2 distant metastases in the same 49 patients. This result revealed that PERCIST was more sensitive in detection the CR and progression patients. There was a significant difference in the results of response classification between RECIST and PERCIST. This may be due to metabolic changes being closely related to malignant potential of tumors or that the maximum diameter of the tumor is not detectable with the resolution of CT. It is possible for omission of the primary lesions and the early pathological changes, of course the recurrences included.
We confirmed clinical stage to be univariate risk factors in the present study, whilst location, tumor T stage, node metastasis were not significant risk factors in our univariate analysis [26, 27]. These findings do not agree with previous reports that tumor T stage is risk factors for the outcome in early NSCLC patients  which may be due to the small size of study population.
PERCIST criteria was a univariate predictor of overall survival of early NSCLC, while RECIST was not significant risk factor in our univariate analysis. The relationship between the metabolic changes of tumors and prognosis has been reported in several studies and the data were controversial [12, 29, 30]. Yanagawa’s and Colleagues reported that PERCIST (CMR vs. non-CMR) was the most significant prognostic factor for predicting DFS and OS in the multivariate Cox proportional hazards regression analysis in Esophageal Cancer . Ding and colleagues demonstrated that only PERCIST was a significant factor for predicting DFS in NSCLC . Contrastingly, Lee and colleagues showed that an early metabolic response did not translate into better survival outcome in advanced/metastatic NSCLC .
In agreement with predominant data currently available, our experimental data demonstrate PERCIST as a univariate predictor of overall survival of early NSCLC. Clinical stage was another univariate predictor of overall survival of early NSCLC in our study. However, clinical stage was not significant risk factors in our multivariate analysis. Interestingly, our results show that PERCIST was the only multivariate predictor of overall survival in early NSCLC patients. In fact, SMD, PMD/PMR, CMR in PERCIST criteria was indicative of a 9.900-fold increase in the risk of overall survival in early NSCLC patients [RR 9.900 (95% CI 1.040, 21.591), P=0.001]. Figure 1 demonstrates that SMD/PMD in PERCIST criteria was associated with overall survival in early NSCLC patients treating with SBRT.
Our present study showed PERCIST was considered to be better for evaluation of treatment response in early NSCLC patients treating with SBRT, which was closely related to prognosis. However, the problem of false positive deserves attention, inflammation of the target tissue, scar tissue after treatment, and so on may cause interference to the utility of response evaluation of radiotherapy with 18F-FDG PET.
The limitations of the current study should also be acknowledged. Firstly, this is a retrospective, single-center study with a small study population. Secondly, the follow-up time is relatively short to observe the overall survive. At the same time, PET-CT was performed at an early stage (3-6 months) after SBRT. At this stage, several study reports that inflammation after treatment showed high metabolism activity with high SUV on PET, may implicit false positive of tumor recurrence [31–33]. However, our present study showed PET aided in the detection of 1 regional recurrence and 2 distant metastases which was not in the field of treatment,exclusion of false positive. Finally, we evaluated the longest diameter of just 1 target lesion that was assessed in PERCIST.
Ideally, it might have necessary to evaluate 5 target lesions were evaluated in RECIST 1.1. However, as RECIST 1.1 suggests that just 3 lesions (not 5 lesions) may be used in randomized studies in which tumor progression is the major concern [10. 11], the number of lesions to evaluate when assessing response to therapy may also be important in both RECIST and PERCIST. Further study from this viewpoint will be needed in the future.