Skin sparing in intensity-modulated radiation therapy of nasopharyngeal carcinoma

Background and Purpose: Radiation therapy of nasopharyngeal carcinomas (NPCs) involves high doses to the target structures which are superficial to the skin surfaces. As a result, the skin toxicities involved are higher and sometimes worsens to such an extent that radiotherapy needs to be interrupted unplanned. This leads to a break in radiation therapy which overall affects the local control and cure rates. The aim of this study is to decrease the skin dose by contouring skin as an organ at risk (OAR) to include in inverse planning calculation. Materials and Methods: Seventy-three cases of nasopharyngeal cancers were planned for 60 Gy to intermediate-risk planning target volume (PTVIntermediate) and 70 Gy to high risk (PTVHigh), by three different modes of Intensity-modulated radiation therapy (IMRT)- namely conventional sequential intensity-modulated radiation therapy (S-IMRT PH-I and PH-II), Skin Spared sequential intensity-modulated radiation therapy (SS-IMRT PH-I and PH-II), and Skin Spared simultaneously instantaneous boost intensity-modulated radiation therapy (SS-SIB IMRT). The plans were compared by dose volume histograms and dose statistics to the PTV as well as to the OAR's. For PTV, mean dose (Dmean), maximum dose (Dmax), and minimum dose (Dmin) were compared to check the homogeneity index (HI) while sparing the skin. For other OAR's Dmean, Dmax and dose to to 1 cubic cm was used for comparison. The skin doses to various volumes from volume to receive 5 Gy (V5) to volume to receive 70 Gy (V70) were evaluated and compared between the three techniques. Statistical analysis was done using one away ANOVA on the data editor SPSS Version 26.0 (SPSS Inc., Chicago, Illinois, USA) to evaluate the results. Continuous variables were expressed as mean +/- standard deviation, and categorical variables were summarized as frequencies and percentages. Survival analysis was done by Kaplan-Meier Estimator. Results: When the skin was considered as an OAR, the skin volume to receive 5, 10, 15, 20, 30, 40, 50, 60, 70 Gy was reduced by 6.5%, 6.5%, 6%, 11.5%, 7%, 6%, 6%, 5%, 2%, respectively, by SS-IMRT PH-I and II and 2%, 4.05%, 4%, 7%, 5%, 3%, 6%, 5%, 1%, respectively, by SS-SIB IMRT when both the SS techniques were compared with S-IMRT PH-I and II. Volume of skin to receive 20 Gy showed maximum reduction in SS-IMRT PH-I and II. A one-way ANOVA was carried out to find the differences in the skin doses between the three techniques. The skin dose in the two SS techniques, i.e., SS-IMRT PH-I and PH-II and SS-SIB IMRT was found significantly lower than that of IMRT plans without skin as an OAR, i.e., S-IMRT PH-I and PH-II (P = 0.000). The PTV doses were well within the 95%-107% of the prescribed dose (HI) and there were no significant differences in the means of the prescribed dose between the simple and skin spared IMRT techniques. The other OARs doses were also evaluated and there were no significant differences between the means of the doses among the techniques. Conclusions: SS IMRT for NPC has demonstrated reduction in skin dose while using skin as an OAR in the optimization. Moreover, decreased skin dose can decrease the skin related toxicities provided there is no compromise on Target dose coverage and OAR dose. We recommend that skin should be contoured as an OAR for NPC, provided PTV is minimally 3-5 mm beneath skin surface, in order to have a better disease control with lesser toxicities and less unplanned treatment interruptions.