Clinical outcomes of CyberKnife stereotactic body radiotherapy for peripheral stage I non-small cell lung cancer

Surgical resection is the primary treatment for early non-small cell lung cancer (NSCLC). However, radiotherapy is the only possible method for treating elderly patients with poor cardiopulmonary function, with poor body condition, or with surgical reluctance. Considering lung cancer is easily affected by respiratory motion, the area needs to be increased to avoid any leakage during tumor irradiation. However, because of numerous restrictions, the radiation dose for tumors is maintained at 60–70 Gy, which leads to a low control rate [1]. Thus, the 2- and 5-year survival rates of early peripheral stage I NSCLC are always maintained at about 39 and 13 % [2], respectively, which are far below the surgical efficacy [3, 4].

As a new means of radiotherapy, CyberKnife stereotactic body radiotherapy (SBRT) avoids the errors caused by respiratory movement through synchronous respiratory tracking technology. By implanting gold fiducials in or around the tumors, tumor movement synchronized with respiratory motion can be truly tracked, more accurately giving the tumor higher doses, while reducing the dose to normal tissues [5‐8].

The planning target volume (PTV) and the volume of normal lung tissues irradiated by high dose are significantly reduced because of the high degree of conformity of the tumor target under CyberKnife SBRT. Thus, the incidence of radiation pneumonitis decreases significantly. The purpose of this study was to report and to analyze the preliminary clinical efficacy of 50 patients with stage I NSCLC.

Radical radiotherapy is the suitable treatment for patients who refused to have operations or those who suffer from stage I NSCLC, but have contraindications for surgery because of objective reasons. Conventional radiotherapy improves the prognosis and increases the quality of life of patients. Although conventional radiotherapy improves the prognosis and quality of life, the dose is difficult to increase because the lung tumor target has low conformity during radiotherapy and it is influenced by respiratory motion. However, the development of CyberKnife provides a good solution to this problem. In this study, CyberKnife was used to treat 50 patients with stage I NSCLC, with an isodose for the tumor target prescription dose reaching 72 to 88 %. This treatment increased the target conformity and allowed the surrounding normal tissues to receive only a small dose of radiation. During the 3–45 months after treatment, few patients developed adverse reactions such as weakness, nausea, emesis, and bone marrow depression. These symptoms were relieved after proper treatments. The local control rate of the pulmonary lesions reached 92 % through regular follow-up and chest CT rechecking. Most of the patients achieved complete remission (the remission rate was 80 %). The quality of life was almost not influenced significantly.

The efficacy of conventional radiotherapy and three-dimensional conformal radiotherapy in stage I NSCLC is difficult to improve, which is mainly because real-time position tracking is impossible during treatment. CyberKnife SBRT with synchrony technology is one of the most mature real-time tracking technologies for clinical applications, and the relevant research results show accuracy up to around 1.5 mm [10]. Therefore, this technology reduces the uncertainty of tumor location through respiration adaptive tracking system and decreases the PTV caused by respiratory movement. This technology reduces the target volume, increases the effective dose, reduces the toxicity to normal tissues, and it significantly improves the efficacy [10‐12]. Of the 50 patients in this study, all received synchronous respiratory tracking technique (synchrony) using gold seed fiducials implanted into the tumors. During the completion process of plans, 80 % of average isodose wrapped the target volume, with the CI reaching 1.17 and the nCI reaching 1.28. By increasing the tumor dose, CyberKnife significantly reduces the dose to normal tissues. The results show that CyberKnife needed only three times sessions in treating NSCLC, and the short-term efficacy was positive, with a total effective rate (CR + PR) of 92 %. The survival rate was 86 % at 1 year and 74 % at 2 years. The disease-specific survival rate was 94 % at 1 year and 86 % at 2 years. As a new exploration in the treatment of early-stage NSCLC, CyberKnife is similar to the surgery in terms of survival rate, and it has an increased local control rate.

Although the CyberKnife has its high accuracy, the border was still the problem during CyberKnife SBRT. A poorly controlled border range PTV could result in slight tumor metastasis, recurrence, or residues. During the planning of CyberKnife, defining the correct PTV range correctly is difficult. Timmerman [13] conducted the first test on increasing the dose of CyberKnife treatment for stage I NSCLC in 2003. The result showed that if the border was controlled within 1 cm, a dose of 60 Gy used three times in 2 weeks was safe. Moreover, Giraud [14] reported that 96 % of cases were had malignant tumor cells ranging from 6 to 8 mm beyond the GTV. Based on this result, Brown [15] extended the primary tumor from 8 to 10 mm and the metastatic tumor from 3 to 5 mm, achieving good results using CyberKnife SBRT. Thus, in the present study, we combined the residual inaccuracy of the synchrony and the micrometastatic characteristics of the tumor. PTV was defined as the border by extending the GTV to 8 mm to cover the border, which covers slight metastasis or unpredictable motion errors. From the results obtained from the 35 months of observation, the 2-year local control rate reached 96 %, which also showed that the boundary should be extended by 8 mm.

The problem of treatment dosage administration of CyberKnife SBRT has confused many experts. If the dose is too high, the tumor control rate is high, but the complications also increase. If the dose is too low, the complications are decreased, but the tumor control rate is low. Timmerman [16] treated 70 patients with stage T1 or T2 (≤7 cm) NSCLC with N0 and M0 pathology. The number of patients was 35 for both T1 and T2, including the peripheral and the central-type patients. The total dose of all patients was 60–66 Gy, divided into three sessions. The treatment was completed within 1–2 weeks. The results showed that the 2-year local control rate was 95 %, the median survival rate was 32.6 months, the 2-year overall survival rate was 54.7 %, and 28 patients died during follow-up. Among those treated for central lung cancer, five cases died of related cancer, six cases died of related treatment, and 17 cases died of related complications. For the patients with peripheral lung cancer, 83 % of patients did not have serious treatment-related toxicity within 2 years, but only 54 % of the central-type patients did not have serious toxicity. The results show that the peripheral lung cancers are well tolerated. Vander Voort vanzyp [17] treated 70 patients with stage I peripheral NSCLC in 2009, who were treated at 45 and 60 Gy, the division of which was completed in three times. Thus, the 2-year local control rate was 96 % at 60 Gy in 2 years, and the 45 Gy group was 78 % ( P = 0.197). The 1-year overall survival rate was 83 %, whereas the 2-year overall survival rate was 62 %. The 60 Gy group did not have treatment-related complications more than grade III. The results from two authoritative studies showed that treating stage I peripheral NSCLC with 60 Gy in three divided doses is safe.

Based on these studies, the patients in our group were all stage I peripheral NSCLC. A dosage from 48 to 60 Gy was given for patients with poor lung function, but the total dose of 60 Gy was given for more than 60 % patients. The 2-year local rate was 96 % in three divided doses, similar to the reported studies [18]. However, the overall survival rate was 74 % in 2 years, slightly higher than that in the reported study [19]. Presumably, patient selection may play a role in the variation of overall survival.

The major symptoms of toxicity of CyberKnife SBRT include short-term weakness, nausea, and mild hematologic toxicity, which improved through active symptomatic treatment. The long-term adverse reactions are mainly local pain and local radiation pneumonia, which improved and returned to normal through positive symptomatic treatment. Long-term radiation therapy-related toxicity greater than grade III was not observed during the 2-year observation. The toxicity reaction increased with the prolongation of follow-up time. However, compared with other clinical studies [20] with median follow-up time of 12 months, we did not find any apparent increase in toxicity. However, further attention to clinical observations and follow-up is necessary to determine whether the long-term adverse reactions will occur.

The samples selected in the present study were stage I peripheral NSCLC, which has its limitations. CyberKnife SBRT is still a new technology, and clinical experience in the dosing and frequency of treatment for Asians is limited, especially in Chinese populations. During the treatment, we mainly referred to the clinical experience in North America and Europe, and the treatment modalities and doses were still in the exploratory stage. Central NSCLC limited the dose and the implantation of markers because of the proximity to the trachea and main bronchi. Thus, selecting treatment cases should be done more carefully.