The number of clinical studies evaluating treatment outcome on the use of IMRT for locally advanced NSCLC is limited. The results reported in our study encourage the use of IMRT in patients with irresectable NSCLC. We found among all patients receiving 66 Gy 1- and 2-year overall survival rates of 71% and 56%, respectively, and a median overall survival of 29.7 months.
Several studies have addressed treatment outcome using 3D-CRT or IMRT for advanced stage NSCLC patients. Nakayama et al. compared the clinical results of high dose 3D-CRT (66 to 84 Gy) with those of conventional two-dimensional radiotherapy (56 to 66 Gy) for patients with Stage III NSCLC . The overall survival rates at 3 years were 9.1% (95% CI, 0.7–18.9%) in the conventional group and 31.0% (95% CI, 18.9–43.1%) in the high-dose group. Wang et al. retrospectively evaluated the outcome of 237 stage III NSCLC patients treated with radiotherapy alone, sequential chemoradiotherapy or concurrent chemoradiotherapy using 3D-CRT . The median overall survival of the entire cohort was 12.6 months, and 2- and 5-year overall survival rates were 22.4% and 10.0%, respectively. Recently, the Memorial Sloan-Kettering Cancer Center retrospectively reviewed treatment outcome of 55 stage I-IIIB inoperable NSCLC patients with large tumor volumes (GTV ≥ 100 cc) treated with IMRT . For patients with stage III disease, 2-year local control and overall survival rates were both 58%, respectively, with a median survival time of 25 months. Liao et al. reviewed the records of 91 patients who received treatment with concurrent chemoradiotherapy using 4DCT/IMRT and compared outcome with patients treated with CT/3D-CRT . The hazard ratio for 4DCT/IMRT was 0.64 (95% CI 0.41-0.98) and statistically significant (p = 0.039) for the overall survival. Based on these reports on 3D-CRT and IMRT [8, 9, 13, 19], and our observations, we consider IMRT as an effective treatment option for patients with locally advanced NSCLC.
Besides favorable outcome data, no severe treatment-related acute pulmonary or esophageal toxicity was observed. In line with previous reports, however, acute esophageal toxicity was enhanced in concurrent schedules compared to sequential chemoradiotherapy [5, 6]. Sura et al. scored acute toxicity on the RTOG scale and reported six patients (11%) to experience grade 3 acute pulmonary toxicity and two (4%) to experience grade 3 acute esophagitis, with no grade ≥4 acute toxicity . Using the CTCAE version 3.0 scoring criteria , Yom et al. reported an incidence of grade ≥3 treatment-related pneumonitis in the IMRT group of 8% (95% CI, 4%–19%) at 6 and 12 months .
Differences in treatment outcome and toxicity may be attributable to different GTV volumes, delivered treatment schedules and radiation dose, tumor stage and the number of patients included in the study. In our study, the patient’s performance status was not explicitly taken into account. Nevertheless, the treatment choice is dependent on the performance score and only patients in good general condition, i.e., Karnofsky index ≥70%, were considered candidates for concurrent chemoradiotherapy.
There are some concerns with respect to the use of IMRT in daily practice. One such concern is the effect of respiratory motion on the accuracy of IMRT delivery . However, several studies have shown that dose variation, introduced by organ motion, almost completely disappears when delivering fractionated radiotherapy [26–28]. Furthermore, in our institute a slow-CT scan is routinely performed for treatment planning purposes incorporating all tumor motion into the definition of the GTV. By doing so, the respiratory motion has largely the same effect on IMRT dose distributions as on conformal radiotherapy techniques. Another concern of IMRT is low-dose radiation exposure of larger volumes of unaffected lung tissue. In the era of 3D-CRT, the mean lung dose (MLD; ) and the volume of unaffected lung tissue receiving a dose of at least 20 Gy (V20) were most often dose-limiting. By applying highly-conformal techniques such as IMRT, however, the MLD alone is more often dose limiting than V20. The reason for this is that by increasing the number of beams, i.e., adding beams from various angles, a larger unaffected lung volume is exposed to a low dose of radiation (below 20 Gy, i.e., V20 decreases). Furthermore, with conformal radiotherapy the PTV is always covered by all beams at all moments while an intrinsic feature of IMRT is that with IMRT, the entire PTV is not always covered by the beam in all its segments. This allows giving a higher dose to the PTV while V20 is unaffected, but with a higher MLD. That is explained by a somewhat higher total dose to the body as a result of increased monitor units delivered by more radiation beams [30, 31]. As a consequence more radiation-induced cancers might be expected in long-term survivors [31, 32].
Despite these concerns, IMRT outcome results are very promising and the toxicity rates acceptable, which supports the consideration of IMRT as an effective and useful radiotherapy technique for the treatment of locally advanced NSCLC.
Notwithstanding the encouraging results of IMRT already reported, more research is needed to further improve outcome for locally advanced NSCLC. To improve local control and outcome with modern radiotherapy techniques in stage III NSCLC patients, a modeling study, investigating the therapeutic gain of individualized dose prescription with dose escalation for various hypofractionation schemes, has recently been conducted at our institution . The encouraging findings have resulted in a clinical phase II trial in which stage III NSCLC patients eligible for (concurrent chemo)radiotherapy are treated with individualized escalated dose.