Proton beam therapy for tumors of the upper abdomen
Abstract
Proton radiotherapy has clear dosimetric advantages over photon radiotherapy. In contrast to photons, which are absorbed exponentially, protons have a finite range dependent on the initial proton energy. Protons therefore do not deposit dose beyond the tumor, resulting in great conformality, and offers the promise of dose escalation to increase tumor control while minimizing toxicity. In this review, we discuss the rationale for using proton radiotherapy in the treatment of upper abdominal tumors—hepatocellular carcinomas, cholangiocarcinomas and pancreatic cancers. We also review the clinical outcomes and technical challenges of using proton radiotherapy for the treatment of these malignancies. Finally, we discuss the ongoing clinical trials implementing proton radiotherapy for the treatment of primary liver and pancreatic tumors.
Introduction
The use of external-beam radiotherapy for upper abdominal tumors has historically been limited by toxicity to the normal surrounding tissues. Although recent advances in photon radiotherapy have enhanced dose conformality to the target, substantial volumes of surrounding critical structures still receive radiation doses that prevent further dose escalation. Minimizing radiation toxicity while maintaining high local tumor control rates forms the basis for using proton radiotherapy to improve the therapeutic index.
Proton radiotherapy has clear dosimetric advantages over photon radiotherapy. In contrast to photons, which are absorbed exponentially, protons have a finite range dependent on the initial proton energy. Protons therefore do not deposit dose beyond the tumor, resulting in great conformality, and offers the promise of dose escalation to increase tumor control while minimizing toxicity. In this review, we discuss the rationale for using proton radiotherapy in the treatment of upper abdominal tumors—hepatocellular carcinomas, cholangiocarcinomas and pancreatic cancers. We also review the clinical outcomes of using proton radiotherapy for the treatment of these malignancies (Table 1). Finally, we discuss the ongoing clinical trials implementing proton radiotherapy for the treatment of primary liver and pancreatic tumors.
| Inclusion criteria | Age range | Adult (>18 years old) |
|---|---|---|
| Language | English only | |
| Species | Humans | |
| Radiation modality | Protons | |
| Publication types |
| |
| Timeframe | 2005–2019 | |
| Exclusion criteria | Publication types |
|
| Other | Otherwise not relevant or out of scope |
Hepatobiliary tumors
Liver transplantation and/or surgery are the only curative options for primary hepatobiliary cancers. Unfortunately, many patients with these malignancies are not surgical candidates due to advanced local extent of disease, insufficient hepatic function, or medical inoperability. Historically, treatment of hepatobiliary cancers with conventional radiotherapy was limited by radiation-induced liver disease. However, with proton radiation therapy, a superior dose distribution can be achieved that spares the surrounding liver. Via the use of proton therapy, tumoricidal doses may be delivered (Figure 1). Table 2 summarizes the radiotherapy outcomes with protons for hepatobiliary tumors.
Proton radiotherapy plan for a primary liver tumor (45 CGE in five fractions). CGE,cobalt gray equivalent.
| Study | Histology | Year | Design | Number of proton patients | CP class | Tumor size (range) | TVT (%) | Median dose (CGE) | Fractions | OS | LC | Late toxicity (non-hematologic) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Chiba et al. University of Tsukuba | HCC | 2005 | Retrospective | 162 | A,B,C | 3.8 cm (1.5–14.5 cm) | 0.154 | 72 | 16 | 5 year: 23.5 | 5 year: 86.9% | Grade ≥ 2: 3.1% |
| Nakayama et al. University of Tsukuba | HCC | 2009 | Retrospective | 318 | A,B,C | NR | 0.138 | 72.6 | 22 | 1-:year 89.5% 5-:year 44.6% | NR | Grade ≥ 3: 1.57% |
| Makita et al. Southern Tohoku Proton Therapy Center | CAC | 2014 | Retrospective | 28 | NR | 5.2 cm (2.0–17.5 cm) | NR | 68.2 | 31 | 1 year: 49.0% | 1 year: 67.7% | Grade ≥ 2: 25.0% |
| Sanford et al. Massachusetts General Hospital | HCC | 2019 | Retrospective | 49 | A,B,C | NR | 67 | 15 | Median: 31 months | 2 year: 93% | NR | |
| Kawashima et al. National Cancer Center Hospital East, Chiba, Japan | HCC | 2005 | Phase II | 30 | A,B | 4.5 cm (2.5–8.2 cm) | 0.4 | 76 | 20 | 2 year: 66% | 2 year: 96% | Grade ≥ 2 GI or Pulm: 0% |
| Fukumitsu et al University of Tsukuba | HCC | 2009 | Prospective | 51 | A,B | 2.8 cm (0.8–9.3 cm) | NR | 66 | 10 | 5 year: 38.7% | 5 year: 87.8% | Grade ≥ 2: 5.8% |
| Bush et al. Loma Linda University | HCC | 2011 | Phase II | 76 | A,B,C | NR | NR | 63 | 15 | NR | 5 year: 80% | Grade ≥ 3: 0% |
| Hong et al. Massachusetts General Hospital | HCC, CAC, Metastases | 2014 | Phase I | 15 | A,B | ≥6 cm | 0 | 60 | 15 | Median: 12 months | 1 year: 53% | Grade ≥ 3: 26.6% |
| Hong et al. Multi institutional | HCC, CAC | 2016 | Phase II | 83 | A,B | 5 cm (1.9–12.0 cm) for HCC 6.0 cm (2.2–10.9 cm) for CAC | NR NR | 58.5 | 16 | 2 year: 63.2% for HCC 2 year: 46.5% for CCA | 2 year: 94.8% for HCC 2 year: 94.1% for CAC | Grade ≥ 3: 3.6% |
| Bush et al. Loma Linda University | HCC | 2016 | Interim Phase III TACE vs Proton | 33 | 2.8 cm (mean) | 0 | 70.2 | 15 | 2 year: 59% (Proton and TACE patients) | 2 year: 88% | NR |
Retrospective evidence evaluating clinical outcomes for Proton radiotherapy in primary hepatobiliary tumors
Hepatocellular carcinoma
One of the first studies on the use of proton radiotherapy for liver tumors was from the University of Tsukuba.1 In this study, Chiba et al, reviewed the outcomes of 162 patients with 192 hepatocellular carcinomas treated from November 1985 to July 1998. The patients in the series were considered unresectable for various reasons, including hepatic dysfunction (72 patients had Child-Pugh class B or C cirrhosis,) multiple tumors, recurrence after resection, and comorbidities precluding surgery. The median age was 62.5 years and the median tumor size was 3.8 cm (range, 1.5–14.5 cm). Patients were treated with four different hypofractionation regimens but the median dose was 72 cobalt gray equivalent (CGE) in 16 fractions over 29 days. At a median follow-up of 31.7 months (range, 3.1–133.2 months), the 5 year local control and overall survival rates were 86.9% and 23.5%, respectively. Patients who had solitary tumors and Child class A cirrhosis (n = 50) had better outcomes, with a 5 year overall survival rate of 53.5%. Few patients (9.7%) experienced elevated plasma aspartate transaminase and alanine transaminase levels by a factor of 3 or less, but these acute reactions subsided within 2 weeks of treatment completion. Five patients (3%) experienced Grade II or higher late toxicity: fibrotic stenosis of the common bile duct, biloma with infection, and ulcers or the stomach and colon. Of note, all patients who experienced Grade II or higher toxicity were treated before 1995. The authors concluded that proton radiotherapy for patients with hepatocellular carcinoma is safe, effective, and repeatable.
The largest series (318 patients) with hepatocellular carcinoma treated with proton radiotherapy was published in 2009 by Nakayama and colleagues.2 These patients were treated from November 2001 to December 2007 with different dose fractionation schemes based on tumor location as follows: tumors within 2 cm of the gastrointestinal tract were treated with 77.0 CGE in 35 fractions, tumors within 2 cm of the porta hepatis with 72.6 CGE in 22 fractions, and tumors >2 cm away from the gastrointestinal tract and porta hepatis with 66 CGE in 10 fractions. 63 patients received two or more courses of proton radiotherapy. The 5 year overall survival was 44.6%. The 5 year overall survival for patients with Child-Pugh A and B disease was 51.9% and 44.5%, respectively. Cox regression analysis showed that survival was statistically significantly associated with Child-Pugh classification, performance status, T stage, and planning target volume (p < 0.05). Treatment was well tolerated, with only five patients developing non-hematological Grade III toxicities. Four of these patients developed skin toxicity and one patient experienced hemorrhage of the colon, which was surgically treated. The authors concluded that proton radiotherapy for hepatocellular carcinoma was safe and effective, and advised consideration of the therapy for patients with limited treatment options.
Recently, Sanford and colleagues published a retrospective study comparing the outcomes of patients with non-metastatic unresectable hepatocellular carcinoma treated with photon (n = 84) and proton (n = 49) radiotherapy.3 The study reviewed patients treated between 2008 and 2017 at the Massachusetts General Hospital who had not previously been treated with liver-directed radiotherapy and did not receive further liver-directed radiotherapy within 12 months after completion of index treatment. With a median follow-up of 14 months, proton radiotherapy was associated with improved overall survival (adjusted hazard ratio = 0.47, p = 0.008). The median overall survival for proton and photon patients was 31 and 14 months, respectively. Proton radiotherapy was also associated with a decreased risk of non-classic radiation-induced liver disease (odds ratio = 0.26, p = 0.03). There was no difference in locoregional recurrence, including local failure, between protons and photons. The authors concluded that proton radiotherapy was associated with improved survival, which may be driven by decreased incidence of post-treatment liver decompensation.
While these finding suggest a possible cause for the difference in outcomes with protons vs photons, these results should be interpreted with a great deal of caution given the retrospective nature. In particular, any retrospective analysis is subject to patient selection, with modality selection influenced by many issues including but not limited to tumor location (dome or not), and proximity to mucosal surfaces.
Cholangiocarcinoma
In 2014, Makita and colleagues published on the clinical outcomes and toxicity of proton beam therapy for advanced cholangiocarcinoma.4 The study included 28 patients with tumors in the following locations: intrahepatic/peripheral cholangiocarcinoma (n = 6), hilar cholangiocarcinoma/Klatskin tumor (n = 6), distal extrahepatic cholangiocarcinoma (n = 3), gallbladder cancer (n = 3), and local or lymph node recurrences (n = 10). The median age was 78 years and the median tumor size was 5.2 cm (range, 2.0–17.5 cm). Various fractionation regimens were used for treatment, with an estimated median BED10 of 75.8 CGE (range 61.7 Gy–105.6 CGE). With a median follow-up of 12 months (range, 3–29 months), overall survival, progression-free survival, and local control rates at 1 year were 49.0%, 29.5%, and 67.7%, respectively. At 1 year, there was higher local control in those patients who had received biologically equivalent doses of (BED10) greater than 70 CGE as compared to those had received less than 70 CGE (83.1 vs 22.2%). Seven (7) patients experienced late Grade III gastrointestinal toxicities, including cholangitis (n = 2), common bile duct stenosis (n = 1), duodenal ulcer (n = 1), duodenal hemorrhage (n = 2) and duodenal stenosis (n = 1). The authors concluded that proton radiotherapy was effective, as evidenced by the high local control rates for patients who had received BED10 of greater than 70 CGE.
Prospective evidence evaluating clinical outcomes for Proton radiotherapy in primary hepatobiliary tumors
In 2005, a single institutional, single-arm prospective study of proton radiotherapy for hepatocellular carcinoma was published by Kawashima and colleagues from the University of Tsukuba.5 Patients were required to have solitary non-metastatic HCC up to 10 cm in size with no good performance status and no history of abdominal radiation. Patients had either Child Pugh class A or B cirrhosis and were treated with 76 CGE in 20 fractions from 1999 to 2003. The median tumor size was 4.5 cm (range, 2.5–8.2 cm), and the median age was 70 years. 12 patients had vascular invasion. After a median follow-up of 31 months, only 1 patient experienced a local failure. 2-year overall survival and local progression-free rates were 66% and 96%, respectively. On multivariate analysis, the only factor that was associated with overall survival was functional hepatic reserve (p = 0.026). Hepatic insufficiency developed in eight patients within 6 months of treatment completion. One patient developed transient skin erosis at 4 months that resolved spontaneously and another patient developed painful subcutaneous at 6 months that did not require opioid analgesics. None of the patients developed Grade II or higher gastrointestinal toxicity.
A second prospective study of hypofractionated proton radiotherapy was published by Fukumitsu et al, in 2009.6 This study included 51 patients with hepatocellular carcinomas treated from 2001 to 2004. Patients with tumors ≤10 cm in size and located more than 2 cm from the porta hepatis and gastrointestinal tract were treated with proton radiotherapy to 66 CGE in 10 fractions. 31 (60.8%) of patients had solitary tumors and 20 (39.2%) patients had multiple tumors. The median tumor diameter was 2.8 cm (range, 0.8–9.3 cm) and all patients had either Child-Pugh Class A or B liver disease. 5-year survival and local control rates were 38.7% and 87.8%, respectively. For patients with Child-Pugh Class A cirrhosis, 5 year survival was 42.1%. The treatment was well tolerated, with three patient developing rib fracture and one patient developing Grade III pneumonitis. 40 patients did not change Child-Pugh class, three patients improved from Child-Pugh class B to A, and 8 patients declined from Child Pugh class A to B. The authors concluded that hypofractionation with protons for tumors located at least 2 cm from the porta hepatis and gastrointestinal tract is safe.
In 2011, Bush and colleagues at Loma Linda University Medical Center published a Phase II trial of 76 patients with hepatocellular carcinoma treated with proton radiotherapy to 63 CGE in 15 fractions.7 Patients were included regardless of tumor size, transplant candidacy, or alpha-fetoprotein (AFP) level. Only patients with cirrhosis were included: 24% were Child-Pugh class C and 54% were outside of Milan criteria. The mean tumor size was 5 cm and 11 patients had multiple tumors. Despite these unfavorable characteristics, median progression-free survival was 36 months, local control was 80%, and median time to failure was 18 months. For patients within Milan criteria, 3 year progression-free survival was 60%. Of the 18 patients who went onto liver transplantation, 6 had pathological complete response and 7 had microscopic residual disease. Treatment was well tolerated with no grade ≥3 toxicities. However, five patients who were treated early on experienced Grade II gastrointestinal toxicity, with bleeding, inflammation, or ulceration. Subsequently, greater care was taken to reduce field margins when tumors were located adjacent to the bowel.
In 2014, Hong and colleagues at the Massachusetts General Hospital published a feasibility study of respiratory-gated proton radiotherapy for liver tumors.8 Eligibility criteria included Childs Pugh class A or B cirrhosis, 1–3 lesions, tumors size ≤6 cm, and a diagnosis of unresectable hepatocellular carcinoma, intrahepatic cholangiocarcinoma, or liver metastases. A total of 15 patients were included and treated with a dose of 45 to 75 CGE in 15 fractions. Dose fractionation was based on normal tissue constraints. With a median follow up of 69 months, the 1 year progression-free survival and overall survival rates were 40, and 53%, respectively. One patient developed a marginal recurrence, three had hepatic recurrences elsewhere in the liver, and two had extrahepatic progression. Grade III toxicities included hyperbilirubinemia (n = 2) and gastrointestinal bleeding (n = 1). One patient with a congenital single ventricle and long-standing portal hypertension died after developing a stomach perforation adjacent to his liver lesion. A maximum tolerated dose to the liver was not established as dose escalation was limited by organs at risk, which differed based on tumor location.
Given the feasibility of respiratory-gating proton radiotherapy for the treatment of liver tumors, Hong and colleagues pursued a Phase II study, which was published in 2016.9 This multi-institutional trial included patients with localized unresectable hepatocellular carcinoma and intrahepatic cholangiocarcinoma. 83 patients with Childs Pugh class A or B, ECOG performance status of 0–2, and no prior radiation received a median dose of 58 CGE in 15 fractions (maximum of 67.5 CGE). The median tumor size was 5 cm (range 1.9–12 cm) for patients with hepatocellular carcinoma (n = 44) and 6.0 cm (range 2.2–10.9 cm) for intrahepatic cholangiocarcinoma (n = 37). With a median follow-up of 19.5 months among survivors, the 2 year local control rates were 94.8% and 94.1% for patients with HCC and intrahepatic cholangiocarcinomas, respectively. The 2 year overall survival rates were 63.2% for patients with HCC and 46.5% for patients with intrahepatic cholangiocarcinomas. Three of the intrahepatic cholangiocarcinoma patients developed Grade III toxicity: liver failure and ascites, gastric ulcer, and elevated bilirubin. Three patients developed worsening liver function and changed Child Pugh class from A to B. Overall, this trial demonstrated that high-dose hypofractionated proton therapy is safe and associated with high local control rates.
In 2016, Bush and colleagues at Loma Linda University medical Center reported an interim analysis of a randomized trial comparing TACE vs proton radiotherapy.10 Patients were eligible for the study if they had a diagnosis of hepatocellular carcinoma and met either Milan or San Francisco criteria for transplant. Patients who were randomized to TACE underwent at least one procedure, with additional TACE for persistent disease. Those who were randomized to the proton arm received 70.2 CGE in 15 fractions. The primary end point was progression-free survival. Of the 69 patients included in the study, 36 were randomized to TACE and 33 to proton therapy. Overall, the median survival was 30 months and the 2 year survival was 59%. The total hospitalization days within 30 days of the treatments were 166 vs 24 for TACE and protons, respectively (p < .001). 12 patients underwent liver transplantation—10 who had undergone TACE and 12 who received proton radiotherapy. In these patients, the pathologic complete response rate was 25% for TACE and 10% for proton radiotherapy. Notably, 2 year local control (88% vs 45%, p = 0.06) and progression-free survival (48% vs 31%, p = 0.06) favored the proton arm.
As noted above, the lack of randomized data comparing protons vs photons in hepatobiliary cancers does not answer the fundamental question as to whether protons are actually required to produce the above outcomes. Specifically, the Phase II study by Hong and colleagues in cholangiocarcinoma represents the largest prospective evaluation of high dose radiation in unresectable intrahepatic cholangiocarcinoma. However, it remains unknown if these results are feasible with photons. At Massachusetts General Hospital, off-study these patients are uniformly treated with photons, rather than protons.
Pancreas tumors
Although chemoradiotherapy for adenocarcinoma of the pancreas has many potential benefits, the standard regimen consists of daily radiation for 6 weeks. This puts a significant drain on sick patients with limited life expectancies and takes time away from full-dose chemotherapy in a disease that has a proclivity for systematic spread. In addition, standard doses, which protect the gastrointestinal tract, have had minimal impact on outcomes. An exciting opportunity exists for protons to improve the therapeutic index, with the goal of decreasing toxicity (Figure 2). Table 3 summarizes the radiotherapy outcomes with protons for pancreatic tumors.
Proton radiotherapy plan for pancreatic cancer (25 CGE in five fractions). CGE,cobalt gray equivalent.
| Study | Resectability | Year | Design | Number of proton patients | Dose | OS | Toxicity |
|---|---|---|---|---|---|---|---|
| Hong et al. Massachusetts General Hospital | Resectable | 2011 | Phase I | 15 | 30 CGE in 10 fractions over 2 weeks to 25 CGE in five fractions over 1 week with capecitabine | NR | Grade ≥ 3: 26.7% |
| Hong et al. Massachusetts General Hospital | Resectable | 2014 | Phase I/II | 50 | 25 CGE in five fractions over 1 week with capecitabine | Median: 17 months | Grade ≥ 3: 4.1% |
| Hong et al. Massachusetts General Hospital | Resectable | 2017 | Phase II preliminary | 50 | 25 CGE in five fractions over 1 week with capecitabine and hydroxychloroquine | Median: 23.3 months | NR |
| Murphy et al. Massachusetts General Hospital | Borderline | 2018 | Phase II | 48 | 25 CGE in five fractions over 1 week with capecitabine | NR | NR |
| Terashima et al. Hyogo Ion Beam Medical Center | Locally advanced | 2012 | Phase I/II | 50 | 67.5 CGE in 25 fractions with concurrent gemcitabine | 1 year: 76.8% | Grade ≥ 3: 10% |
| Murphy et al. Massachusetts General Hospital | Locally advanced | 2018 | Phase II preliminary | 7 | 25 CGE in five fractions over 1 week with capecitabine for patients who became resectable after chemotherapy | NR for proton | NR for proton |
Prospective evidence evaluating clinical outcomes for Proton radiotherapy in pancreatic tumors
Resectable pancreatic cancer
The literature examining the utility of proton radiotherapy for pancreatic cancer is sparse. In 2011, Hong and colleagues at the Massachussetts General Hospital published on the feasibility of pre-operative short-course chemoradiation with proton radiotherapy and capecitabine for resectable pancreatic cancer. In this Phase I study, 15 patients with pancreatic head cancer and no prior therapy were treated with four escalating dose levels ranging from 30 CGE in 10 fractions over 2 weeks to 25 CGE in five fractions over 1 week. Surgical resection was performed 4–6 weeks after the lower dose levels, and 1–3 weeks after the highest dose level. Of the 15 patients, 11 proceeded to surgical resection. The patients who did not undergo surgery had either metastatic disease (n = 4) or unresectable cancer (n = 1). The treatment was well tolerated with no unexpected 30 day post-operative complications. Grade III toxicity were as follows: biliary obstruction (n = 2), biliary obstruction with infection (n = 1), and positional shoulder pain (n = 1). The authors concluded that preoperative chemoradiation with 1 week of proton radiotherapy and capecitabine followed by early surgery was feasible.
Given these promising results, Hong and colleagues published a follow-up Phase I/II study in 2014.11 The study included 50 patients with resectable pancreatic cancer, of 15 were treated in the Phase I portion as described above. The remaining 35 patients were treated in the Phase II portion with 25 CGE in five fractions over 1 week. The clinical target volume was defined as gross tumor volume with a 1 cm margin, respecting anatomical boundaries, as well as elective nodal coverage of the celiac, portahepatic, superior mesenteric artery and para-aortic (through the level of the third portion of the duodenum) areas. The planning target expansion was determined based on respiratory motion and set-up variation. The primary endpoint was to demonstrate a rate of Grade III or higher toxicity of less than 20%. Two of the Phase II patients (4.1%) experienced Grade III toxicity (one patient with colitis and one patient with chest wall pain). Of the 48 patients eligible for analysis, 37 underwent surgery, with pathology revealing positive lymph nodes in 30 patients. With median follow-up of 38 months, the median progression-free survival and the overall survival were 10 and 17 months, respectively. Locoregional failure occurred in 16.2% of the patients who underwent resection and 72.9% of all patients failed distantly. In exploratory analyses, KRAS status, high CXCR7 expression, and high levels of CEA, CA 19–9 and HGF levels were associated with worse outcomes. The authors concluded that short-course chemoradiation with proton radiotherapy is well tolerated and associated with high local control rates in patients with resectable pancreatic cancer.
In 2017, Hong and colleagues presented the preliminary results of a Phase II study of hydroxychloroquine and pre-operative short-course chemoradiation for resectable pancreatic cancer at the American Society of Clinical Oncology annual meeting (NCT01494155). In this study, 50 patients received hydroxychloroquine, a potent inhibitor of autophagy, in addition to pre-operative short-course chemoradiotherapy with protons (25 CGE in 5 fractions) or photons (30 Gy in 10 fractions). Surgery was completed 1–3 weeks after completion of chemoradiation and patients were recommended to resume hydroxychloroquine and continue until progression in addition to having 6 months of adjuvant gemcitabine. The primary objective of the study was to determine the progression-free survival of the addition of hydroxychloroquine to the regimen. A total of 38 patient had R0 resection and 8 had R1 resection. The median follow-up in 26 surviving patients was 18.3 months. The median progression-free and overall survival was 11.7 months and 23.3 months, respectively. The authors concluded that hydroxychloroquine with pre-operative short-course chemoradiotherapy was well tolerated but did not significantly impact disease-free survival.
Borderline resectable pancreatic cancer
In 2018, Murphy and colleagues at the Massachussetts General Hospital reported the results of a Phase II trial on total neoadjuvant therapy with FOLFIRINOX followed by individualized chemoradiotherapy for borderline resectable pancreatic cancer.12 The primary end point was R0 resection rates. The study included 48 patients with good performance status who received 8 cycles of FOLFIRINOX and were then re-evaluated with imaging. Patients with resolution of vascular involvement received chemoradiation with proton radiotherapy (25 CGE in 5 fractions) or photon radiotherapy (30 Gy in 10 fractions). Those who had continued vascular involvement received long-course chemoradiation with photons to 50.4 Gy with a vascular boost to 58.8 Gy. 27 patients had short-course chemoradiation and 17 patients had long-course chemoradiation. A total of 31 patients underwent R0 resection; among those who underwent resection, the R0 resection rate was 97%. Median progression-free and overall survival of the entire cohort was 14.7 months and 37.7 months, respectively. Among those who underwent surgery, median progression-free survival was 48.6 months and the median overall survival has not yet been reached. The authors concluded that preoperative FOLFIRINOX followed by individualized chemoradiotherapy in borderline resectable pancreatic cancer patients shows high rates of R0 resection and favorable survival outcomes.
Locally advanced pancreatic cancer
In 2012, Terashima and colleagues published a Phase I/II trial on gemcitabine-concurrent proton radiotherapy for locally advanced pancreatic cancer.13 A total of 50 patients were enrolled in the study: 5 patients with GI-adjacent cancer received 50 CGE in 25 fractions, 5 patients with non-GI-adjacent cancer received 70.2 CGE in 26 fractions, and 40 patients received 67.5 CGE in 25 fractions regardless of cancer location. Gemcitabine was administered concurrently for a total of 3 weeks. With a median follow-up on 12.5 months, the 1-year freedom from local-progression, progression-free, and overall survival rates were 81.7%, 64.3%, and 76.8%, respectively. The scheduled therapy was feasible in all but six patients secondary to acute hematologic of GI toxicities. Four patients experienced Grade III toxicity (gastric ulcer and hemorrhage). One patient expired from a gastric hemorrhage; this patient was treated with the highest radiation dose schedules and received a maximum dose of 52 CGE to the stomach. With limited follow-up, the authors concluded that the results were encouraging.
In 2014, Takatori and colleagues published a single-center observational study on the upper gastrointestinal complications associated with gemcitabine-concurrent proton radiotherapy for locally advanced pancreatic cancer.14 In this prospective study, 91 patients had endoscopic examinations before and after proton radiotherapy to determine the incidence rates of ulcers, hemorrhage, and perforation associated with this treatment. Patients received gemcitabine concurrently with radiation therapy for a total of 3 weeks; radiation dose was 67.5 CGE in 25 fractions. Almost half (49.4%) of the patients had radiation-induced ulcers in the stomach and duodenum, with the most common locations being the lower stomach (59%) and the horizontal part of the duodenum (39%). Endoscopy did not reveal gastrointestinal hemorrhage or perforation. Given that approximately half of patients had ulcers, the authors concluded that further improvements with gemcitabine-concurrent proton radiotherapy are warranted.
In 2018, Murphy and colleagues presented preliminary results of a Phase II trial on the combination of TGF-b1 inhibitor losartan and FOLFIRINOX for locally advanced pancreatic cancer at the American Society of Clinical Oncology annual meeting (NCT01821729). In this study, patients received 8 cycles of FOLFIRINOX and losartan. If the tumor was deemed resectable after chemotherapy, patients underwent chemoradiation with protons (25 CGE in 5 fractions) and capecitabine. If the tumor was still unresectable, patients received chemoradiation with photons to 50.4 Gy with a vascular boost to 58.8 Gy. The primary endpoint was R0 resection rate. A total of 49 patients were evaluable. 10 patients received fewer than 8 cycles of systemic therapy due to progression (n = 4), losartan intolerance (n = 3), and toxicity (n = 3). Approximately half (51%) of patients experienced Grade III or greater toxicity, with diarrhea, thrombocytopenia, nausea, and neutropenia/febrile neutropenia. A total of 46 patients underwent chemoradiation: 7 patients (14%) had proton radiotherapy and 39 pts (80%) had long-course chemoradiation with photons. Of the 39 patients where surgery was attempted, 34 patients were resected with a R0 resection in 30 patients and R1 resection in 4 patients. The median progression-free and overall survival were 17.5 and 31.4 months, respectively, for the entire cohort. The median progression-free and overall survival among resected patients were 21.3 and 33.0 months, respectively. Biomarker analysis showed superior outcomes for patients with lower plasma levels of HGF. Given the high R0 resection rates, a multicenter Phase II trial is planned.
Technical challenges of proton therapy for tumors of the upper abdomen
Although a detailed discussion regarding the unique challenges of treatment planning of proton radiotherapy is beyond the scope of this review, a brief discussion is warranted. Article search methodology for this section used PubMed and Google Scholarengines. PubMed searches were performed with the search criteria for “proton” AND “therapy” AND “motion,” as well as “proton” AND “radiotherapy” AND “motion” in the title and abstract. The search was limited to articles appearing since 2009 in order to focus on current topics. A total of 305 articles were returned by this search. From these, articles were excluded according to the following criteria: if they related to imaging only, and mentioned proton therapy only as a possible application; review articles; clinical studies not including critical discussion of technical factors; case reports; conference proceedings; consensus guidelines; and articles relating to interfraction but not intrafraction motion. A similar search was performed using Google Scholar to find related articles not identified by the PubMed search. Subsequently, Google Scholar searches were performed for specific terms identified by the initial Google Scholar search, including: “proton therapy abdominal compression,” “proton therapy respiratory gating,” “proton therapy robust planning,” “proton therapy bowel gas,” “proton therapy interplay effect,” and “proton therapy rescanning.” The purpose of this section is to catalog technical challenges relevant to upper-abdominal proton therapy. Summarizing the entire body of literature related to these technical challenges isoutside the scope of this review.
While proton radiotherapy offers clear dosimetric advantages over photon radiotherapy by limiting radiation dose to normal surrounding tissues, careful attention must be paid to uncertainties regarding range, setup, and motion. Protons are very dependent on the electron density of the tissues they traverse, which can lead to inaccurate plan delivery when there is motion. Motion management devices such as abdominal compression belts, daily image guidance with fiducial placement, and respiratory gating approaches, can help minimize these uncertainties.15,16 In addition, differences in bowel gas patterns can be challenging.17 To address this, patients can modify their diet and planners can chose beam arrangements that avoid passing through gas. In addition, changes in body contour and weight can result in overdosing of organs at risk and underdosing of the target. Therefore, special attention has to be paid to weight loss or accumulation of ascitic fluid. Adaptive planning may be necessary in select cases. Larger planning margins, compensator smearing and the utilization of several beams, may also assist in preventing errors.18
Current motion management strategies for proton therapy include four-dimensional (4D) dose calculations, rescanning, tumor tracking or gating, and motion-robust planning. 4D dose calculations seek to explicitly model the anticipated body motion based on a planning 4D-CT or 4D-MRI scan.19 Calculation of each beam on representative breathing phase images changes target coverage estimations considerably, unlike photon therapy where the internal target volume technique effectively guarantees coverage.20–22 However, it is widely reported that 4D-CT is an imperfect estimation of patient breathing motion due to breathing irregularities and baseline drifts.23 A recent innovation is the use of phase-resolved 4D cone beam CT reconstruction to accurately estimate fraction dose accounting for the patient’s breathing pattern at the time of treatment.24
The interplay effect describes an accidental synchronization of intensity modulated beamlet delivery with anatomic motion leading to an interference pattern.25 The interference can be positive, leading to hot spots, or negative leading to cold spots. In photon therapy this potentially arises due to multileaf collimator motion but is generally believed to be negligible.26 For scanned proton beams, the magnitude of the interplay effect is highly dependent on patient and beam characteristics and may be significant in some cases27,28 Rescanning refers to delivering the scanned pencil beam in multiple passes to average out any hot or cold spots resulting from the interplay effect, however rescanning alone does not allow for reduction in target volume expansions for motion uncertainties, and is a complement rather than a replacement to other motion management strategies.19,29,30 The interplay effect does not apply to passively scattered proton beams.
The use of respiratory gating is well accepted in proton therapy.8,31–33 Gating-based external surrogates or fiducial markers may be regarded as a standard technique. Gated marker-less tracking is a recent innovation in both photon and proton therapy that has been clinically implemented for lung and liver by at least one proton center.34
Robust optimization in radiation therapy refers to the practice of reducing the sensitivity of treatment plans to known uncertainties by incorporating them into the optimization process. The internal target volume technique could be considered a crude motion-robust optimization method, which is effective for photon therapy but not for proton therapy.35 In the worst-case optimization technique, a dose distribution representing the lowest possible target dose and highest organ at risk doses is composed from multiple dose distributions computed using various uncertainty perturbations.36 This dose is combined in a weighted fashion with the nominal dose distribution in the objective function. Recent work has focused on combining robust optimization with 4D dose calculations37–42 [R8].
Recent clinical results for upper-abdominal malignancies indicate that proton therapy may live up to the promise of the physics-based advantages it holds over photon therapy. However it must be acknowledged that the techniques of photon therapy continue to evolve. Particularly relevant for upper-abdominal tumors are optimized non-coplanar radiotherapy and MRI-guided radiotherapy. A number of groups have demonstrated that treatment planning techniques using computer-optimized non-coplanar beam and arc selection lead to improved critical organ dosimetry in retrospective liver radiotherapy planning studies.43–46 Non-coplanar planning for proton therapy is complicated by physical constraints of the delivery nozzle as well as considerations of expanded range uncertainty due to increased target depth in tissue.47 MRI-guided online adaptive photon radiotherapy for abdominal tumors is under investigation by a number of groups and has shown promising clinical results in a multi-institutional retrospective study of pancreatic cancer treatments.48–52 Principle advantages arising from MRI guidance for photon radiotherapy include gating and/or tracking based on real-time soft-tissue images, and the availability of simulation-quality images for on-table adaptation to avoid exceeding stomach and bowel tolerance doses.53,54 In addition to these expected benefits, MRI-guided proton therapy could potentially ameliorate proton range uncertainties by allowing a per-fraction measurement of proton path lengths in tissue. Incorporating MRI into proton therapy has been deemed feasible.55 Significant implementation challenges include deviation of proton trajectories in the fringe magnetic field, and interaction of the magnetic field of the MRI with the beam steering magnetics and beam monitoring instrumentation.56,57
Future directions
Proton radiotherapy is emerging as a safe and efficacious treatment for primary hepatobiliary and pancreatic tumors. However, questions remain regarding the utility of protons as compared to photons.
The NRG Oncology recently opened a Phase III trial (NCT03186898) evaluating radiation therapy with protons or photons in treating patients with unresectable or locally recurrent hepatocellular carcinoma. The study is planning to enroll 186 participants with three or fewer single or multinodular tumors. For patients with a single lesion, the maximum tumor size is 15 cm; for patients with two lesions, the maximum size is 10 cm; and for patients with three lesions, the maximum size is 6 cm. Tumors may have portal vein involvement or thrombosis only if there is a single lesion. Patients can have no worse liver function than Childs B7 cirrhosis. Patients will be stratified by the planned number of fractions (5 or 15 as determined by the treating physician) and tumor vascular thrombus and randomized to either proton or photon radiotherapy. Patients will be planned to a goal of 50 CGE in 5 fractions or 67.5 CGE in 15 fractions, but prescription dose may be lowered based on liver metrics. The primary end point is overall survival, and progression-free survival, local control, and toxicity are secondary end points.
There is a strong rationale for using proton radiotherapy for pancreatic cancer. Proton radiation techniques may allow for delivery of pre-operative radiation in a shortened schedule, thus eliminating the delay to systemic therapy. However, ablative doses (above 25 CGE in five fractions) have not yet been delivered using proton radiotherapy. In addition, the techniques of proton therapy have not evolved as quickly as those using photons. In particular, MRI-guided online adaptive photon radiotherapy for abdominal tumors has shown promising clinical results. Additionally, even new forms of proton delivery, such as intensity-modulated proton therapy, are limited to a few beam angles, rather than the 360 degrees that are available to photon treatment plans. This limitation often limits the conformality of the high dose regions and negates the benefit of the Bragg peak.
We await the final results of two studies: the Phase II study on the addition of hydroxychloroquine preoperative concurrent short-course radiation therapy for resectable pancreatic cancer, and the Phase II study on FOLFIRINOX combined with losartan followed by individualized chemoradiotherapy for locally advanced pancreatic cancer.
REFERENCES
1. . Proton beam therapy for hepatocellular carcinoma: a retrospective review of 162 patients. Clinical Cancer Research 2005; 11: 3799–805. doi: https://doi.org/10.1158/1078-0432.CCR-04-1350
2. . Proton beam therapy for hepatocellular carcinoma: the University of Tsukuba experience. Cancer 2009; 115: 5499–506.
3. . Protons vs photons for unresectable hepatocellular carcinoma: liver decompensation and overall survival proton radiotherapy for hepatocellular carcinoma. Int. J. Radiat. Oncol. Biol. Phys 2019;.
4. . Clinical outcomes and toxicity of proton beam therapy for advanced cholangiocarcinoma. Radiation Oncology 2014; 9: 26. doi: https://doi.org/10.1186/1748-717X-9-26
5. . Phase II study of radiotherapy employing proton beam for hepatocellular carcinoma. Journal of Clinical Oncology 2005; 23: 1839–46. doi: https://doi.org/10.1200/JCO.2005.00.620
6. . A prospective study of hypofractionated proton beam therapy for patients with hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2009; 74: 831–6. doi: https://doi.org/10.1016/j.ijrobp.2008.10.073
7. . The safety and efficacy of high-dose proton beam radiotherapy for hepatocellular carcinoma: a phase 2 prospective trial. Cancer 2011; 117: 3053–9. doi: https://doi.org/10.1002/cncr.25809
8. . A prospective feasibility study of respiratory-gated proton beam therapy for liver tumors. Pract Radiat Oncol 2014; 4: 316–22. doi: https://doi.org/10.1016/j.prro.2013.10.002
9. . Multi-Institutional phase II study of high-dose Hypofractionated proton beam therapy in patients with localized, unresectable hepatocellular carcinoma and intrahepatic cholangiocarcinoma. JCO 2016; 34: 460–8. doi: https://doi.org/10.1200/JCO.2015.64.2710
10. . Randomized clinical trial comparing proton beam radiation therapy with transarterial chemoembolization for hepatocellular carcinoma: results of an interim analysis. Int J Radiat Oncol Biol Phys 2016; 95: 477–82. doi: https://doi.org/10.1016/j.ijrobp.2016.02.027
11. . A phase 1/2 and biomarker study of preoperative short course chemoradiation with proton beam therapy and capecitabine followed by early surgery for resectable pancreatic ductal adenocarcinoma. Int J Radiat Oncol Biol Phys 2014; 89: 830–8. doi: https://doi.org/10.1016/j.ijrobp.2014.03.034
12. . Total neoadjuvant therapy with Folfirinox followed by individualized chemoradiotherapy for borderline resectable pancreatic adenocarcinoma. JAMA Oncol 2018; 4: 963–9. doi: https://doi.org/10.1001/jamaoncol.2018.0329
13. . A phase I/II study of gemcitabine-concurrent proton radiotherapy for locally advanced pancreatic cancer without distant metastasis. Radiotherapy and Oncology 2012; 103: 25–31. doi: https://doi.org/10.1016/j.radonc.2011.12.029
14. . Upper gastrointestinal complications associated with gemcitabine-concurrent proton radiotherapy for inoperable pancreatic cancer. J Gastroenterol 2014; 49: 1074–80. doi: https://doi.org/10.1007/s00535-013-0857-3
15. . Evaluation of motion mitigation using abdominal compression in the clinical implementation of pencil beam scanning proton therapy of liver tumors. Med Phys 2017; 44: 703–12. doi: https://doi.org/10.1002/mp.12040
16. . Comparison of pancreatic respiratory motion management with three abdominal corsets for particle radiation therapy: case study. J. Appl. Clin. Med. Phys 2019; 20: 111–9.
17. . Comparison of gastric-cancer radiotherapy performed with volumetric modulated Arc therapy or single-field uniform-dose proton therapy. Acta Oncol 2017; 56: 832–8. doi: https://doi.org/10.1080/0284186X.2017.1297536
18. . Spot scanning proton beam therapy for prostate cancer: treatment planning technique and analysis of consequences of rotational and translational alignment errors. Int J Radiat Oncol Biol Phys 2010; 78: 428–34. doi: https://doi.org/10.1016/j.ijrobp.2009.07.1696
19. . Lomax, a. J. A statistical comparison of motion mitigation performances and robustness of various pencil beam scanned proton systems for liver tumour treatments. Radiother. Oncol. J. Eur. Soc. Ther. Radiol. Oncol 2018; 128: 182–8.
20. . Four-Dimensional proton treatment planning for lung tumors. Int J Radiat Oncol Biol Phys 2006; 64: 1589–95. doi: https://doi.org/10.1016/j.ijrobp.2005.12.026
21. . The relative accuracy of 4D dose accumulation for lung radiotherapy using rigid dose projection versus dose recalculation on every breathing phase. Med Phys 2017; 44: 1120–7. doi: https://doi.org/10.1002/mp.12069
22. . Adequate margin definition for scanned particle therapy in the incidence of intrafractional motion. Phys Med Biol 2013; 58: 6079–94. doi: https://doi.org/10.1088/0031-9155/58/17/6079
23. . Initial clinical observations of intra- and interfractional motion variation in MR-guided lung SBRT. Br J Radiol 2018; 43:
20170522 . doi: https://doi.org/10.1259/bjr.2017052224. . Feasibility of 4DCBCT-based proton dose calculation: an ex vivo porcine lung phantom study. Zeitschrift für Medizinische Physik 2018. doi: https://doi.org/10.1016/j.zemedi.2018.10.005
25. . Effects of motion on the total dose distribution. Semin Radiat Oncol 2004; 14: 41–51. doi: https://doi.org/10.1053/j.semradonc.2003.10.011
26. . Dosimetric impact of breathing motion in lung stereotactic body radiotherapy treatment using Image-Modulated radiotherapy and volumetric modulated Arc therapy. Int J Radiat Oncol Biol Phys 2012; 83: e251–6. doi: https://doi.org/10.1016/j.ijrobp.2011.12.001
27. . Motion interplay as a function of patient parameters and spot size in spot scanning proton therapy for lung cancer. Int J Radiat Oncol Biol Phys 2013; 86: 380–6. doi: https://doi.org/10.1016/j.ijrobp.2013.01.024
28. . Respiratory liver motion estimation and its effect on scanned proton beam therapy. Phys Med Biol 2012; 57: 1779–95. doi: https://doi.org/10.1088/0031-9155/57/7/1779
29. . Comparative study of layered and volumetric rescanning for different scanning speeds of proton beam in liver patients. Phys Med Biol 2013; 58: 7905–20. doi: https://doi.org/10.1088/0031-9155/58/22/7905
30. . Surface as a motion surrogate for gated re-scanned pencil beam proton therapy. Phys Med Biol 2017; 62: 4046–61. doi: https://doi.org/10.1088/1361-6560/aa66c5
31. . The M. D. Anderson proton therapy system. Med Phys 2009; 36(9Part1): 4068–83. doi: https://doi.org/10.1118/1.3187229
32. . A proton beam therapy system dedicated to spot-scanning increases accuracy with moving tumors by real-time imaging and gating and reduces equipment size. PLoS One 2014; 9:
e94971 . doi: https://doi.org/10.1371/journal.pone.009497133. . Physics controversies in proton therapy. Semin Radiat Oncol 2013; 23: 88–96. doi: https://doi.org/10.1016/j.semradonc.2012.11.003
34. . Carbon-Ion pencil beam scanning treatment with gated Markerless tumor tracking: an analysis of positional accuracy. Int J Radiat Oncol Biol Phys 2016; 95: 258–66. doi: https://doi.org/10.1016/j.ijrobp.2016.01.014
35. . Robust optimization of intensity modulated proton therapy. Med Phys 2012; 39: 1079–91. doi: https://doi.org/10.1118/1.3679340
36. . Worst case optimization: a method to account for uncertainties in the optimization of intensity modulated proton therapy. Phys Med Biol 2008; 53: 1689–700. doi: https://doi.org/10.1088/0031-9155/53/6/013
37. . Exploratory study of 4D versus 3D robust optimization in intensity modulated proton therapy for lung cancer. Int J Radiat Oncol Biol Phys 2016; 95: 523–33. doi: https://doi.org/10.1016/j.ijrobp.2015.11.002
38. . Motion-robust intensity-modulated proton therapy for distal esophageal cancer. Med Phys 2016; 43: 1111–8. doi: https://doi.org/10.1118/1.4940789
39. . Motion mitigation in scanned ion beam therapy through 4D-optimization. Physica Medica 2014; 30: 570–7. doi: https://doi.org/10.1016/j.ejmp.2014.03.011
40. . Robustness of 4D-optimized scanned carbon ion beam therapy against interfractional changes in lung cancer. Radiotherapy and Oncology 2017; 122: 387–92. doi: https://doi.org/10.1016/j.radonc.2016.12.017
41. . 4D robust optimization including uncertainties in time structures can reduce the interplay effect in proton pencil beam scanning radiation therapy. Med. Phys 2018;.
42. . Potential for improvements in robustness and optimality of intensity-modulated proton therapy for lung cancer with 4-Dimensional robust optimization. Cancers 2019; 11:
35 . doi: https://doi.org/10.3390/cancers1101003543. . iCycle: integrated, multicriterial beam angle, and profile optimization for generation of coplanar and noncoplanar IMRT plans. Med Phys 2012; 39: 951–63. doi: https://doi.org/10.1118/1.3676689
44. . 4π non-coplanar liver SBRT: a novel delivery technique. Int J Radiat Oncol Biol Phys 2013; 85: 1360–6. doi: https://doi.org/10.1016/j.ijrobp.2012.09.028
45. . Vmat plus a few computer-optimized non-coplanar IMRT beams (VMAT+) tested for liver SBRT. Radiotherapy and Oncology 2017; 123: 49–56. doi: https://doi.org/10.1016/j.radonc.2017.02.018
46. . Optimizing highly noncoplanar VMAT trajectories: the novo method. Phys. Med. Biol. 2018; 63: 025023. doi: https://doi.org/10.1088/1361-6560/aaa36d
47. . Comparison of different treatment planning approaches for intensity-modulated proton therapy with simultaneous integrated boost for pancreatic cancer. Radiation Oncology 2018; 13: 228. doi: https://doi.org/10.1186/s13014-018-1165-0
48. . Phase I trial of stereotactic MR-guided online adaptive radiation therapy (smart) for the treatment of oligometastatic or unresectable primary malignancies of the abdomen. Radiotherapy and Oncology 2018; 126: 519–26. doi: https://doi.org/10.1016/j.radonc.2017.11.032
49. . Retrospective evaluation of decision-making for pancreatic stereotactic MR-guided adaptive radiotherapy. Radiotherapy and Oncology 2018; 129: 319–25. doi: https://doi.org/10.1016/j.radonc.2018.08.009
50. . Dosimetric benefits and practical pitfalls of daily online adaptive MRI-guided stereotactic radiation therapy for pancreatic cancer. Pract Radiat Oncol 2019; 9: e46–54. doi: https://doi.org/10.1016/j.prro.2018.08.010
51. . A new era of image guidance with magnetic resonance-guided radiation therapy for abdominal and thoracic malignancies. Cureus 2018; 10: e2422:
e2422 . doi: https://doi.org/10.7759/cureus.242252. . Using adaptive magnetic resonance image‐guided radiation therapy for treatment of inoperable pancreatic cancer. Cancer Med 2019; 8: 2123–32. doi: https://doi.org/10.1002/cam4.2100
53. . Gadoxetate for direct tumor therapy and tracking with real-time MRI-guided stereotactic body radiation therapy of the liver. Radiotherapy and Oncology 2016; 118: 416–8. doi: https://doi.org/10.1016/j.radonc.2015.10.024
54. . Magnetic resonance imaging for target delineation and daily treatment modification. Semin Radiat Oncol 2018; 28: 178–84. doi: https://doi.org/10.1016/j.semradonc.2018.02.002
55. . Future of medical physics: Real‐time MRI‐guided proton therapy. Med Phys 2017; 44: e77–90. doi: https://doi.org/10.1002/mp.12371
56. . Proton beam deflection in MRI fields: implications for MRI-guided proton therapy. Med Phys 2015; 42: 2113–24. doi: https://doi.org/10.1118/1.4916661
57. . Emerging technologies in proton therapy. Acta Oncol 2011; 50: 838–50. doi: https://doi.org/10.3109/0284186X.2011.582513
