EDITORIAL Complexity in Radiation Therapy: It’s Complicated Efstathios Kamperis, MD, MSc,* Chionia Kodona, MSc, PhD, y Konstantinos Hatziioannou, MSc, PhD, y and Vasileios Giannouzakos, MD* Departments of *Radiation Oncology and y Medical Physics, Papageorgiou General Hospital, Thessaloniki, Greece Received Jun 7, 2019. Accepted for publication Sep 6, 2019. The culmination of radiation therapy’s technical innovation during the past few decades led to the development of volumetric modulated arc therapy (VMAT), 1 which constitutes the evolution of intensity modulated radiation therapy (IMRT). 2 Both of these treatment modalities have greatly increased our ability to conform dose to target volumes and spare surrounding normal tissues. Naturally, they have become the pre- dominant techniques for various treatment sites (eg, head, neck, and prostate). The steep dose gradients achievable with these techniques are materialized via multileaf collimator (MLC), dose rate, and gantry speed modulation at the cost of intrinsic plan complexity. Therefore, they are vulnerable to a wide spectrum of uncertainty sources, such as dose calculation accuracy, leaf positional accuracy, leaf speed accuracy, interleaf leakage and transmission, tongue-and-groove effect, setup errors, changes in anatomy, patient motion, or motion of the inner organs. 3-6 These uncertainty factors have been studied independently of each other, such as the relationship between dose error and gap error for various gap widths in dynamic MLC fields by LoSasso et al. 5 Such errors may compromise tumor coverage and increase the dose to organs at risk. 7,8 In this short communication we raise the following points:  A perfect plan from a dosimetric standpoint, fulfilling the most stringent quality metrics, when delivered may result in a dose distribution significantly different from that planned.  Number of monitor units, beam on time, leakage radia- tion, and integral dose all increase with complexity. In a worst-case scenario, this exposure may double the inci- dence of radiation-induced secondary cancers. 9  The temporal interplay between the delivery of a com- plex plan and intrafraction motion is largely unknown. In the prostate, such effects are too complex and can cause dosimetric variations in both high- and low-dose gradients. 10,11 Remarkably, the use of target margins is not an effective strategy to protect against this uncer- tainty. 11 In head and neck cancer, despite intrafraction motion being small, 12 the interaction of beam modula- tion, organ motion, and dosimetry remains to be explored.  Excessively modulated plans impose a considerable strain on the linear accelerator. This is not an issue per se, provided that a clinically meaningful benefit for the pa- tient is truly conferred.  Complex plans typically consist of many small gaps and irregularly shaped apertures. This increases the uncertainty in dose calculations resulting from poor modeling of small field output factors 13 and in the MLC system (rounded leaf ends, tongue-and-groove, Corresponding author: Efstathios Kamperis, MD, MSc; E-mail: ekamperi@gmail.com Disclosures: We declare no conflict of interest. Int J Radiation Oncol Biol Phys, Vol. 106, No. 1, pp. 182e184, 2020 0360-3016/$ - see front matter Ó 2019 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.ijrobp.2019.09.003 www.redjournal.org