Abstract
Background/Aim: We aimed to investigate the dosimetric effects of a spacer placed between the pancreas and surrounding gastrointestinal structures in intensity-modulated radiation therapy (IMRT) planning to provide more effective radiation therapy for locally advanced pancreatic cancer (LAPC). Patients and Methods: Treatment planning was performed for six patients with LAPC based on computed tomography images without spacers and with 5-mm or 10-mm spacers virtually inserted under the supervision of a hepatobiliary pancreatic surgeon. The prescription dose was 63 Gy in 28 fractions. Results: With the exception of one case of pancreatic head cancer, planning target volume receiving ≥95% of the prescribed dose (PTV V95) was achieved by 90% or more by inserting a spacer, and by 95% or more in all 3 cases of pancreatic body and tail cancer by inserting a 10-mm spacer. Conclusion: IMRT with appropriate spacer placement may help provide high-dose treatment for LAPC and improve associated patient outcomes.
The five-year survival rate of pancreatic cancer patients is estimated to be 8%, which is the lowest among all cancer types (1). Despite progress in the treatment of other cancers, resulting in decreasing incidence and mortality rates, pancreatic cancer remains a leading cause of death.
Although margin-negative surgical resection can cure pancreatic cancer, only 10%-15% of patients have resectable pancreatic cancer (2). Therefore, recently, the role of radiation therapy (RT) has been challenged by evidence from a randomised clinical trial (LAP07), which showed no overall survival benefit associated with adding radiation to systemic chemotherapy (3). The difficulty in delivering curative radiation doses to the gross tumour volume without damaging adjacent radiosensitive organs, including the duodenum, stomach, and jejunum, may account for this finding.
New RT techniques, including intensity-modulated RT (IMRT) and particle therapy, which uses either carbon ions (carbon ion radiation therapy; CIRT) or protons (proton beam therapy; PBT), have been introduced for treating various cancer types. These techniques allow higher doses to be delivered to target volumes, thus improving treatment efficacy. Considering its excellent dose distribution, Shinoto et al. showed good CIRT results for treating pancreatic cancer using a 55.2-Gy dose (relative biological effectiveness; RBE) in 12 fractions (4). Meanwhile, Terashima et al. reported that 70 Gy (RBE) of PBT with gemcitabine resulted in high survival and local control rates, leading to better results with high-dose prescriptions than previous reports (5, 6). Nevertheless, the latter study also reported serious adverse events related to the gastrointestinal tract, including gastric ulcers.
To solve this problem, a uniform high-dose prescription to the target may be achieved by creating physical distance with spacers placed between the target and organs at risk (OARs). This approach has shown promising results. Previous studies have reported regarding RT using surgical and biologic mesh spacers for liver cancer (7-10). Fukumoto and Sasaki et al. developed a bioabsorbable polyglycolic acid (PGA) spacer and showed that particle therapy using this spacer could be successfully and safely performed on tumours located in the abdominal and pelvic regions (11, 12). The authors have proposed the concept of space-making particle therapy. In Japan, the use of spacers in particle therapy to the abdominal area is covered by public health insurance, making it part of the standard treatment. Recently, the first case of a pediatric malignant tumour who was treated with PGA spacer and proton therapy was reported with surgical (13) and radiotherapy aspects (14). However, because the number of particle therapy centres is small, limited access restricts its widespread use. IMRT may reduce the dose to OARs and increase the target dose in X-ray therapy; nevertheless, the dose to areas near OARs must be reduced, restricting the likelihood of achieving uniform coverage of the prescribed dose.
RTOG1201 (15), a phase II randomised controlled trial, incorporated chemoradiation therapy with high-dose IMRT (63 Gy in 28 fractions). However, the trial’s protocol limited the dose to the gastrointestinal tract, and the study was terminated early owing to the slow recruitment of cases. No other clinical trials on X-ray therapy with high-dose prescriptions have progressed to date, probably due to concerns about radiation-related adverse events such as gastrointestinal ulcers and bleeding.
It may be possible that the use of the PGA spacer can maximise the potential efficaciousness of RT for pancreatic cancer, even in X-ray therapy. Thus, this study aimed to evaluate the efficacy of IMRT with spacer placement for LAPC in a planning study.
Patients and Methods
We retrospectively and randomly selected six LAPC cases for whom computed tomography (CT) data for treatment planning was obtained from August 2014 to January 2018. The diagnosis of unresectability was judged by imaging modalities such as CT, and informed consent was obtained from all patients. All cases had a set of 2-mm-thick non-contrast-enhanced CT images acquired in the supine position under respiratory gating for treatment planning purposes. CT scanning was designed for respiratory movement, and the expiratory phase was used for treatment planning.
We have the clinical experience of using a bioabsorbable PGA spacer (Neskeep®; Alfresa Pharma, Osaka, Japan) for LAPC patients treated with PBT. Figure 1 shows a representative case. We mimicked the process of using this product using a “virtual spacer” in treatment planning. Virtual spacers were contoured by radiation oncologists based on recommendations of hepato-biliary-pancreatic surgeons with experience in spacer placement. Spacers with 5-mm and 10-mm thickness were adopted as virtual spacers based on Neskeep® lineup. OARs overlapping the virtual spacer were shrunk accordingly. An example of virtual spacer placement is shown in Figure 2. Velocity version 4.0 and Eclipse version 15.6.8 (Varian Medical Systems, Palo Alto, CA, USA) were used for contour and treatment planning.
Placement of a bioabsorbable polyglycolic acid spacer (arrows) on the anterior surface of the pancreas.
Placement of a virtual spacer in a patient: a) reference image (without spacer), b) 5-mm spacer, c) 10-mm spacer. Red, gross tumour volume; light blue, clinical target volume; yellow, planning target volume; pink, stomach; brown, small bowel; light purple, duodenum; green, spacer.
Tumours defined by CT and other modalities, including magnetic resonance imaging and positron emission tomography, were delineated as gross tumour volume (GTV). The clinical target volume (CTV) was defined as GTV plus a 5-mm basic margin. The planning target volume (PTV) was defined as CTV plus a setup margin (5 mm). The prescribed dose was 63 Gy in 28 fractions, which corresponded to the value of the high-dose prescription arm of the RTOG1201 trial (15). The V95 (volume receiving ≥95% of the prescribed dose) of the PTV was expected to exceed 90%. However, if dose constraints of OARs could not be met, OARs were prioritised. The clinical goals for PTV and OARs were based on those set in RTOG1201 (Table I).
Clinical goals in treatment planning (based on RTOG 1201 regimen).
Results
Table II presents case characteristics. Among clinical cases, three were of pancreatic head cancer, and three were of pancreatic body and tail cancer. One case with pancreatic head cancer (No. 3) met the clinical goal of PTV (V95 ≥95%) even without a spacer. All cases with pancreatic body and tail cancer (No. 4-6) met the PTV V95 of ≥95% with a spacer, whereas only a single case with pancreatic head cancer did so (No. 3). In pancreatic body and tail cancer cases (No. 4-6), 10-mm spacers achieved the PTV V95 of ≥95% in all cases, while 5-mm spacers did so in a single case (No. 6).
Case characteristics and changes in PTV V95 due to placing PGA spacers.
Figure 3 shows IMRT plans without a spacer and with 5-mm and 10-mm spacers in the case of the pancreatic body and tail cancer (No. 4). An improvement in dose-volume histogram (DVH) parameters associated with spacers was demonstrated. Even 5-mm spacers improved PTV coverage compared with no spacers; however, 10-mm spacers were needed to satisfy the goal of PTV V95 of ≥95%.
Intensity-modulated radiation therapy treatment plans for a patient with pancreatic body and tail cancer a) without spacer, b) with 5-mm spacer, c) with 10-mm spacer, and d) a dose-volume histogram (DVH). Yellow, planning target volume; pink, stomach; brown, small bowel; light purple, duodenum; green, spacer. Dotted line, no spacer; broken line, 5-mm spacer; solid line, 10-mm spacer in DVH.
Figure 4 shows dose distributions of IMRT without a spacer and with 5-mm and 10-mm spacers in a case of pancreatic head cancer (No. 1). Because of a technical issue, placing a spacer between a pancreatic head tumour and duodenum was difficult. No improvement to DVH parameters associated with spacers was observed.
Intensity-modulated radiation therapy plans for a patient with pancreatic head cancer a) without spacer, b) with 5-mm spacer, c) with 10-mm spacer, and d) a dose-volume histogram (DVH). Yellow, planning target volume; pink, stomach; brown, small bowel; light brown, large bowel; light purple, duodenum; green, spacer. Dotted line, no spacer; broken line, 5-mm spacer; solid line, 10-mm spacer in DVH.
Discussion
This is the first study to demonstrate the efficacy of spacer placement in IMRT for LAPC in a planning study. Dose reduction to the stomach and intestines can be achieved in IMRT for pancreatic body and tail cancer using a 10-mm spacer; this approach is associated with good PTV coverage, while completely satisfying the dose constraints of OARs.
CIRT and PBT enable dose reduction to the OARs near the PTV by taking advantage of a physical characteristic called the Bragg peak. IMRT using X-rays lacks this characteristic. However, we showed that using a spacer in IMRT may help achieve high-dose coverage on par with that of particle therapy for LAPC.
Previous reports have revealed the benefits of high-dose RT in pancreatic cancer patients. Krishnan et al. (16) and Chung et al. (17) reported that RT of approximately 70 Gy (biological equivalent dose) improved overall survival. Of note, the incidence rate of gastrointestinal bleeding was lower in Krishnan et al.’s study than in Chung et al.’s study (16, 17). This finding might be because the target in this study was limited to cases in which the tumour was ≥1 cm of the nearest element of the gastrointestinal tract. This finding suggests that the criterion of 10-mm distance is clinically valid for reducing the risk of gastrointestinal toxicities, further supporting the present study findings.
This study had some limitations. First, this was a retrospective treatment planning study. However, we followed a procedure that was previously performed in clinical practice; we expect that future prospective clinical trials will validate our findings. Second, this study was included in a small number of cases. The characteristics of anatomical structures tend to vary among patients. Future studies are required to examine the advantages and disadvantages of spacer placement in larger sample sizes. Finally, in this study, it was not possible to achieve dose reduction to the duodenum by inserting spacers; owing to the anatomical characteristics of this area, it is not possible to insert PGA spacers between the duodenum and pancreas. However, Rao et al. used hydrogels to create spaces between the head of the pancreas and duodenum (18, 19). Future studies may examine the role of PGA and hydrogel spacers in overcoming this problem.
In conclusion, this planning study demonstrated that IMRT with appropriate spacer placement might help achieve safe and effective RT for LAPC. Future clinical trials are required to verify these findings.
Acknowledgements
The Authors wish to thank Editage (www.editage.jp) for English language editing.
Footnotes
Authors’ Contributions
Conception and design: R Sasaki, H Kawaguchi, Y Demizu, T Ishihara and D Miyawaki. Treatment planning: R Sasaki, H Kawaguchi, N Mukumoto, S Komatsu, H Akasaka, M Shinoto, Y Shioyama, K Nakamura and T Fukumoto. Writing and revision of the manuscript: H Kawaguchi and Y Demizu.
Conflicts of Interest
The Authors state that they have no conflicts of interest relative to this study.
- Received December 7, 2020.
- Revision received December 16, 2020.
- Accepted December 17, 2020.
- Copyright© 2021, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
