br Unresectable LAPC is well known to undergo
Unresectable LAPC is well-known to undergo distant metas-tases and the role of radiotherapy as local therapy within the con-
Fig. 3. Case presentation. 71-year old man with pancreatic cancer. (a) CT before treatment. Arrows indicate gross tumor volume. (b) Dose distribution image. Total irradiation dose is 67.5 GyE and isodose lines represent 95–10% of the isocenter dose from inside to outside. (c) CT 25 months after CCRT using Dynasore beams. Arrows indicate a reduced tumor.
text of CCRT is not well established . Historically, irradiation field size has been reduced according to chemotherapeutic inten-sity but controversy remains as to how much local lesion control
could affect outcome of those LAPC patients in which distant metastases are frequent and lethal [10,41]. However, OS was clo-sely related with LC in our study and those patients without local
42 PBT for unresectable LAPC
recurrence trended toward longer survival than patients with local recurrences as shown in Supplementary Fig. 2. Moreover, several studies dealing with unresectable LAPC patients report that high dose irradiation improves LC and OS [25,27,31,32]. Taken together, we can therefore consider that excellent control of the primary lesions due to high irradiation dose delivery can control primary tumor activity and prolong survival of unresectable LAPC patients.
In our previous simulation study, the mean primary tumor dose from our 67.5 GyE in 25 fractions protocol (using concomitant boost technique) was 66.2 GyE in GI tract-separated tumors, 62.1 GyE in tumors adjacent to the GI tract, and 64.1GyE as a whole . As shown in the detailed dose classification analysis in Fig. 2, the OS and LC of patients treated with a 67.5 GyE protocol were much higher than the conventional 50 GyE protocol and equivalent or higher than with 54–60 GyE, making it a reasonable and con-vincing result. Furthermore, dose escalation using concomitant boost technique can improve treatment effectiveness. Considering that no patient suffered from severe GI tract toxicity, higher dose delivery using the concomitant boost technique coupled with the physical characteristics of proton beams has the potential to safely and effectively replace the conventional protocol. Further studies into this could clarify the clinical significance of dose escalation using the concomitant boost technique.
In our study, a total of 23 patients were concurrently treated by hyperthermia as machine time allowed. Hyperthermia is known to increase the blood supply to tumors and improve delivery of large doses of cytotoxic oxygen and drugs . It also acts as a radiation-and chemo-sensitizer while inhibiting transcription factors like NF-kB that play a crucial role in pancreatic carcinoma development and progression [43–45]. As pancreatic cancer features chemo-radio resistance and hypoxia, these patients are presumably good candidates for hyperthermic therapy. Although Maebayashi et al.
 previously reported that CCRT combined with hyperthermia prolonged OS, hyperthermia did not function as a significant prog-nostic factor for OS and LC in our study. Future optimization stud-ies are necessary to prove the efficacy of hyperthermic therapy in proton beam CCRT.
There are several limitations of this study. First, because this is a retrospective study, our patient population might have been biased toward favoring the effectiveness of PBT for survival or improve-ment of quality of life. However, the selection of PBT or photon therapy was arbitrarily done by the patients or referring physicians without prompting on our part. Additionally, to further reduce any bias, we performed univariate analyses using several factors. Sec-ond, variations of chemotherapy regimens, use of hyperthermia and our irradiation dose protocol might have introduced bias as an especially wide variety of pre-treatment chemotherapy regi-mens complicated determination of relevance to outcome. Even if we had no say in the pre-treatment chemotherapy and did not base our protocols on patient performance or clinical staging, some bias inevitably remained. Finally, because of the potentially rapid progression of disease, some patients could not be regularly exam-ined at our institute. We made every effort to get as much data as possible through mail to these patients and physicians but patients who were shifted to palliative care or died soon after treatment would complicate any data gathering with regard to exact local treatment effect.
Clinical reports of PBT for unresectable LAPC patients are extre-mely scarce but clinical outcomes are satisfactory among CCRT reports to this point. To the best of our knowledge, this is the first study to prove the effectiveness of CCRT using proton beams by clarifying that high dose administration to tumors leads to improved prognosis for unresectable LAPC patients. We therefore consider that the physical advantages of proton beams can increase the value of radiotherapy in local CCRT treatment strategies for these patients.