Phase I study evaluating the treatment of patients with locally advanced pancreatic cancer with carbon ion radiotherapy: the PHOENIX-01 trial

Cancer of the pancreas is the fourth most common cause of cancer-related death in men and women. Ductal adenocarcinoma is the most common variant in over 90% of all pancreatic malignancies. Patients commonly present with syptoms such as weight loss, jaundice, pain, dyspepsia or stool irregularities.

Surgical resection is considered the treatment of choice; in over 80% of all patients surgery alone does not convert into long-term tumor control. In highly specialized centers, surgery associated mortality has been reduced to < 5%, but even in these centers overall survival after surgery alone remains to be around 20 months, with a five year survival rate of about 20% [1, 2]. Several studies have shown that strong prognostic factors are resection status, presence of lymph-node metastases, tumor size as well as tumor DNA content [3‐6]; with respect to resection status, only R0-resection seems for provide a major benefit as compared to R1 or R2 resections.

Pancreatic cancer is commonly classified as resectable, borderline resectable or unresectable. Interdisciplinary discussion on a case-by-case basis is recommended for all patients with respect to optimal treatment of patients; individual decisions for local treatment may vary depending on tumor size, symptoms or other.

Special focus has been set on the group of patients with non-metastasized, but locally advanced and thus non-resectable or borderline resectable patients. In these cases, alternatives to radical surgery must be evaluated. In the past, chemotherapy alone was administered, usually with gemcitabine or 5-FU.

Several concepts of chemoradiation have been established, and clincial data has shown beneficial results with acceptable toxicity. A GITSG study initially defined the role for radiation and chemotherapy with bolus 5-FU, comparing split-course radiotherapy with 40 Gy with chemotherapy to radiation with 60 Gy in combination with chemotherapy or radiation alone: Median survival could be increased from 22.9 weeks to 42.2 weeks [7]. Subsequent studies aimed at the optimization of 5FU application, and most contemporary studies have left the regimen of split-course radiotherapy. Gemcitabine has also been used as a radiation sensitizer in pancreatic cancer [8‐12], and several studies have shown that gemcitabine and radiation seem to be equieffective to 5-FU and radiation [10, 13].

To date, no specific randomized trials evaluating the effect of chemoradiation have been performed. Results from a prospective study evaluated clinical outcome after radiation, gemcitabine and cisplatinum in patients with locally advanced pancreatic adenocarcinoma. The regimen showed acceptable toxicity, but no benefit compared to other single-agent chemotherapeutic treatments was shown [14]. A retrospective analysis performed by the MD Anderson Cancer Center revealed that in resectable patients preoperative radiation is not disadvantageous, and since postoperative recovery often extends the time until (about 25% of all patients) adjuvant treatment can be initiated [15]. Several other arguments can be brought up supporting radiochemotherapy before any surgical resection: Likelihood of R0 resections due to tumor shrinkage, selection of patients with more stable disease for surgery, treatment of micrometastases at an earlier stage and treatment of tissue not modified by surgery and potentially more sensitive to chemotherapy and radiation [9, 16, 17]. The MD Anderson group showed that median survival was 21 months, 31% of all patients demonstrated no evidence of disease, in 132 patients treated with preoperative chemoradiation and subsequent pancreaticoduodenectomy [15]. In our institution we have shown that 26% of patients with locally advanced pancreatic cancer can undergo secondary resection after gemcitabine-based chemoradiation, and that patients with a complete resection are associated with a relatively good long-term prognosis [18].

The idea to bring locally advanced tumors into a resectable status has been addressed explicitly in several studies [17, 19‐24]. There is evidence that the rate of margin-free resection can be increased by preoperative chemoradiation [25]; however, no randomized trials have confirmed this hypothesis. Palmer and colleagues could show that gemcitabine-based chemotherapy given as neoadjuvant therapy for potentially resectable pancreatic cancer lead to an increase in resectability as compared to gemcitabine alone [21]. On the other hand, some authors have shown that even after combined radiation and chemotherapy the rate of potentially resectable patients did not increase, and a substantial amount of patients presented with advanced disease 4–6 weeks after completion of radiochemotherapy, when restaging was performed [8, 26]. At the Heidelberg Center, a novel approach of using modern intensity modulated radiotherapy (IMRT) in combination with gemcitabine and the EGFR-antibody cetuximab has been performed with a comparable intent [27]. The final study results are will be reported soon, the study has finished recruitment.

To date, within treatment recommendations for locally advanced pancreatic cancer without distant metastases, radiation or radiation alone can be found as possible approaches. Besides modifying systemic treatment and keeping photon radiation as a constant factor, new radiation modalities such as particle therapy offer distinct physical and biological characteristics and are a promising treatment alternative for patients with pancreatic cancer potentially increasing local responses, leading to higher local tumor control rates, higher rates of resectability, as well as an increase in survival.

Proton as well as carbon ion radiotherapy are characterized by an inverted depth-dose-curve, with low dose deposition within the entry channel, and a defined high local dose deposition in the so called Bragg Peak. Therefore, sparing of normal tissue in front of and behind the defined target volume is possible. More conformal dose delivery with increased sparing of healthy surrounding normal tissue is possible, enabling the application of higher local doses to the target volumes. Carbon ions additionally offer a higher biological effectiveness due to the characteristic and severe radiation damages produced in target tissues. For pancreatic cancer cell lines, RBE for carbon ion beams values between 1.16 and 2.46 have been reported depending on cell line and endpoint [28]. Is has been shown that carbon ion radiotherapy leads to an increase in local control especially in radiation resistant tumors, such as chordomas and chondrosarcomas, adenoidcystic carcinomas, melanomas as well as HCC, as well as pancreatic cancer [29].

At Japanese institutions clinical trials have been conducted to evaluate carbon ion radiotherapy for pancreatic cancer. Between 2000 and 2003, 22 patients with localized, resectable adenocarcinoma of the pancreas were treated with preoperative carbon-ion radiotherapy. Doses between 44.8 Gy E and 48 Gy E in single doses of 2.8 Gy E and 3.0 Gy E were applied. Local control rate was 100%, and overall survival was 59% at 1 year. A subgroup of patients did not receive post-radiotherapeutic resection, which showed a significantly lower outcome (1 year overall survival of 3%) as compared to patients receiving surgery (86% overall survival at 1 year [30]). This was followed by a more hypofractionated regimen in a study with comparable inclusion criteria, increasing applied dose from 30 Gy E to 35.2 Gy E in 8 fractions. Still, no local tumor recurrences were observed, and overall survival at 1 and 5 years in patients treated with surgical resection after preoperative carbon ion radiotherapy was 89% and 51% [30]. Patients with locally advanced and in the first step inoperable pancreatic cancer were included into a Phase I/II trial. Inclusion criteria included patients with histologically confirmed ductal carcinoma of the pancreas, volume of 14 cm or less in diameter, inoperable lesions; patients were treated with increasing doses from 38.4 Gy E to 52.8 Gy E in 12 fractions. Overall survival at 1 year was 60% with a local control rate of 81%, and patients receiving higher doses showed a significant benefit in local control and overall survival [30]. In analogy to treatment schedules with photons, also the combination of chemotherapy with gemcitabine and carbon ions was transferred into a trial protocol: In the first part, a carbon ion dose of 43.2 Gy E in 12 fractions was set as constant, weekly gemcitabine was increased from 400 mg/m² to 1000 m². Acute hematological toxicity as well as non-hematological side effects were low, no grade 4 and 5 toxicities were observed. Only in the 700 mg/m² and 1000 mg/m² gemcitabine arm, 3/6 (50%) and 8/12 (75%) developed grade III hematological toxicity. In all three treatment arms local control was comparable during follow up, but survival was higher with increasing doses of chemotherapy. At present a constant dose of gemcitabine at 1000 mg/m² is applied with a carbon ion dose escalated to 50.4 Gy E [30].

Therefore, in the present PHOENIX-01 trial, carbon ion radiotherapy using the active rasterscanning technique will be evaluated in patients with advanced pancreatic cancer in combination with weekly gemcitabine and adjuvant gemcitabine.

The purpose of the trial is to evaluate carbon ion radiotherapy in patients with locally advanced pancreatic cancer. With respect to toxicity, the optimal dose of carbon ion radiotherapy will be determined. Therefore, the primary endpoint is toxicity, secondary endpoints are evaluation of progression-free survival, response, and overall survival after carbon ion radiotherapy.

Focus of the analysis is to evaluate safety and efficacy of carbon ion radiotherapy in patients with locally advanced pancreatic cancer. Therefore, the aim of the trial is to observe low rates of toxicity with high local doses due to effect of the altered biology of carbon ions on pancreatic cancer cells as well as the superior physical characteristics.

The primary objective is toxicity of carbon ion radiotherapy. As secondary objectives, imaging response, progression-free survival and overall survival were defined. The trial will be performed as a single-center single-armed Phase I study. Informed consent will be obtained by all patients.

About one third of all patients with the primary diagnosis of pancreatic cancer present with non-metastasized, locally advanced disease. Primary curative resection is rarely possible in these patients, however, it is known that complete resection of the tumor is the strongest prognostic factor. Thus, strategies for downsizing of locally advanced tumors have moved into focus: With radiochemotherapy using photons, about 20-30% of all patients will become resectable [2, 18, 35].

The physical and biological properties of the carbon ion beam promise to improve the therapeutic ratio in patients with pancreatic cancer: Due to the inverted dose profile dose deposition in the entry channel of the beam leads to sparing of normal tissue; the Bragg peak can be directed into the defined target volume, and the sharp dose fall-off thereafter again spares normal tissue behind the target volume. The higher RBE of carbon ions, which has been shown also for pancreatic cancer cell lines in the preclinical setting, is likely to contribute to an increase in local control, and perhaps in OS [28, 30, 36, 37]. Early data from Japanese centers have shown convincing results [30, 37].

Based on the preclinical hypothesis as well as the Japanese data in several indications, the PHOENIX-01 trial was designed to evaluate carbon ion radiotherapy for patients with locally-advanced pancreatic cancer. Beam delivery will be performed using active rasterscanning, and the trial is designed as a dose-escalation trial.

In conclusion, this is the first trial to evaluate actively delivered carbon ion beams in patients with locally advanced pancreatic cancer within a dose-escalation strategy.