Pancreatic cancer is currently the fourth leading cause of cancer-related deaths in the United States [
A common limitation in pancreatic cancer radiation therapy has been the toxicity thought to result from irradiation of nearby organs, most prominently the small bowel and/or stomach. Published studies of various chemoradiotherapy regimens have demonstrated a wide variety of adverse events ranging from occasional nausea to grade 4 vomiting. Evidence suggests that intensity-modulated radiation therapy (IMRT) may provide superior limitation of radiation toxicity, likely due to a dose reduction to organs at risk (OARs). Indeed, recent studies have shown successful OAR dose reduction for other upper abdominal cancers using IMRT as compared to three-dimensional conformal radiotherapy (3D-CRT). For example, in the postoperative treatment of gastric and gallbladder cancer, statistically significant reduced doses and toxicity to adjacent liver and kidneys have been observed [
Whether or not the success that has been seen in the treatment of other malignancies with IMRT is transferable to the management of pancreatic cancer is a subject of ongoing research. For this study, we hypothesized that the incorporation of Dose-Painted IMRT (DP-IMRT) in the management of locally advanced pancreas cancer (LAPC) would reduce side effects due to a decrease in doses to the small bowel, stomach, and liver. DP-IMRT allows for a higher dose per fraction to be delivered to the primary pancreas tumor, while simultaneously treating regional lymph nodes at standard fractionation to doses required for eradication of microscopic disease. Therefore, DP-IMRT may mitigate adverse effects through improved OAR sparing while hypofractionation to the primary tumor may improve clinical response. Here we report on the use of IMRT in the treatment of LAPC with a focus on local control rates, survival, and toxicity. Findings are compared to recently published studies utilizing 3D-CRT in combination with various chemotherapeutic agents [
All patients (
Patient demographics and treatment regimens.
Characteristic |
|
---|---|
Gender | |
Male | 14 (54) |
Female | 12 (46) |
Age (years) | |
Median (range) | 66 (40–87) |
T stage | |
T2 | 1 (4) |
T3 | 12 (46) |
T4 | 13 (50) |
N stage | |
N0 | 10 (38) |
N1 | 10 (38) |
Unknown | 6 (23) |
Induction chemotherapy | |
Gemcitabine | 22 (85) |
Gemcitabine-paclitaxel-capecitabine (GTX) | 1 (4) |
5-Fluorouracil-leucovorin-irinotecan-oxaliplatin (FOLFIRINOX) | 2 (8) |
No induction chemotherapy | 1 (4) |
Concurrent chemotherapy | |
Gemcitabine | 7 (35) |
5-FU | 12 (60) |
Both | 1 (5) |
Tissue constraints were as previously published [
Gross Tumor Volume (GTV) was determined based on CT findings, and margins of 1 cm were added to establish the Clinical Target Volume (CTV). Planning Treatment Volume 1 (PTV1), which included regional pancreaticoduodenal, celiac, para-aortic, and portahepatic lymph nodes, received a total dose of 45 Gy. Dose escalation was delivered to PTV2 (GTV + 0.5 cm) using either 50.4 Gy (
Ten randomly selected DP-IMRT plans for patient treatment were subsequently replanned with 3D-CRT to evaluate the radiation doses to organs at risk (OARs) and the potential relationship of this exposure to the observed adverse events. Of these 10 patients, 6 were treated with dose escalation to 54 Gy. For comparison 3D-CRT plans, both coplanar and noncoplanar plans were designed as previously described [
During the course of their treatment and at routine follow-up visits, patients were asked a standard questionnaire pertaining to commonly occurring toxicities consisting of nausea, vomiting, diarrhea, abdominal pain, anorexia, fatigue, and weight loss. Responses were categorized according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 guidelines. Morbidity was assessed prior to, during, and following treatment. Response Evaluation Criteria In Solid Tumors (RECIST) criteria version 1.1 were used to define progression of disease. Locoregional cancer recurrence was defined as recurrence within the tumor bed and regional lymph nodes. Metastatic recurrence was defined as distant failure outside the radiation field. Preradiation CA 19-9 levels were obtained 4–6 weeks prior to RT during the last cycle of chemotherapy and postradiation CA 19-9 levels were collected at a median of 3 weeks (range 0.5–18) following the completion of RT.
The Fisher exact test and
Of the original 26 patients in our cohort, 77% (
Kaplan-Meier curves for patients without evidence of metastatic disease on CT imaging (
Among the adverse events noted in our study, the most commonly reported were diarrhea, nausea, vomiting, anorexia, abdominal pain, fatigue, and weight loss. The toxicities noted were mostly grade 1 or 2 with the most common being grade 1 nausea (45%) and grade 1 fatigue (45%). One patient exhibited grade 3 nausea and anorexia, and one patient experienced grade 3 anorexia only. Two patients developed grade 3 gastrointestinal (GI) bleeding, one of which was a result of local failure, and 1 patient developed grade 3 bile duct stenosis. There were no grade 4 adverse events noted (Table
Adverse events for patients who received DP-IMRT (
Toxicity |
|
|||
---|---|---|---|---|
Grade 0 | Grade 1 | Grade 2 | Grade 3 | |
Diarrhea | 12 (60) | 4 (20) | 4 (20) | |
Nausea | 4 (20) | 9 (45) | 6 (30) | 1 (5) |
Vomiting | 11 (55) | 6 (30) | 3 (15) | |
Anorexia | 8 (40) | 2 (10) | 8 (40) | 2 (10) |
Abdominal pain | 13 (65) | 6 (30) | 1 (5) | |
Fatigue | 7 (35) | 9 (45) | 4 (20) | |
Weight loss | 13 (65) | 6 (30) | 1 (5) | |
Anemia† | 9 (45) | 9 (45) | 1 (5) | |
Gastrointestinal hemorrhage | 18 (90) | 2 (10) |
DP-IMRT: dose painted.
†Hematologic toxicity for one patient who was lost to followup could not be determined.
Univariate analysis for prognostic factors of OS yielded a statistically significant association with postradiation CA 19-9 levels but not preradiation CA 19-9 levels (
Kaplan-Meier curves for patients receiving Dose-Painted Intensity Modulated Radiation Therapy (DP-IMRT) for (a) pretreatment CA 19-9 levels, (b) posttreatment CA 19-9 levels, and (c) GTV.
Given that the common gastrointestinal (GI) toxicities of RT are thought to be due to irradiation of crucial nearby OARs, it is reasonable to hypothesize that the volume of tumor targeted for irradiation would correlate strongly and directly with the severity of adverse events. Among the adverse events noted in our study, GTVs > 200 cm3 correlated with the degree of abdominal pain (
Comparing radiation exposure of organs at risk between coplanar 3D-CRT and noncoplanar 3D-CRT, treatment with noncoplanar 3D-CRT resulted in significantly lower V12 of the right kidney (
Radiation delivery to OARs.
Organ | Mean IMRT value | Mean coplanar 3D-CRT value | Mean noncoplanar 3D-CRT value |
|
|
---|---|---|---|---|---|
Right kidney V12 (%) | 75.5 | 75.6 | 68.9 |
|
0.3 |
R. kidney V18 (%) | 45.5 | 62.0 | 58.4 | 0.6 | 0.1 |
Left kidney V12 (%) | 69.0 | 48.3 | 41.0 | 0.1 |
|
Left kidney V18 (%) | 43.2 | 35.2 | 26.4 |
|
|
Mean liver dose (Gy) | 16.5 | 20.0 | 19.5 | 0.1 |
|
Small bowel V45 (%) | 7.1 | 12.1 | 12.8 |
|
|
Stomach V45 (%) | 8.7 | 16.6 | 15.8 | 0.2 |
|
Max spinal cord dose | 35.0 | 32.0 | 35.5 |
|
0.8 |
OAR: organ at risk; IMRT: intensity-modulated radiation Therapy; 3D-CRT: three-dimensional conformal radiation therapy;
Axial and coronal images of representative DP-IMRT plans depicting dose wash of ≥42.8 Gy to PTV1 (yellow) and dose escalation of 53.9–57.3 Gy to PTV2 (red). DP-IMRT: Dose-Painted Intensity Modulated Radiation Therapy PTV: Planning Treatment Volume.
The treatment of LAPC is an area of active investigation. Patients are often treated with induction chemotherapy followed by concurrent chemoradiation, although the role of CRT has been questioned [
In the current paper, we investigated the role DP-IMRT may serve in the treatment of LAPC with a focus on improved local control. The OS in our study was approximately 11.6 months, consistent with the published literature, including the RTOG 98-12 and the E4201, which showed overall survivals of 11.2 and 11.1 months, respectively [
The toxicities experienced by the patients were predominantly grades 1 and 2 with one patient experiencing grade 3 nausea and anorexia and one patient exhibiting grade 3 anorexia. Patients in our cohort did not experience any grade 4 adverse events. This is in contrast to published studies utilizing gemcitabine or 5-FU combined with 3D-CRT, in which higher incidences of grade 3 or greater toxicity were observed. Up to 79% grade 3 and 4 adverse events were reported in the Eastern Cooperative Oncology Group (ECOG) trial E4201 which involved 50.4 Gy given concurrently with gemcitabine [
The limitations of the current study are the modest sample size, its retrospective nature, and short follow-up time. Also, while care was taken to ask each patient a standard set of questions regarding adverse events, post hoc analysis inevitably increases the likelihood of underestimating low-grade toxicities. However, the more serious nature of grade 3 and 4 adverse events minimizes the possibility that such levels of toxicity would have been undetected. In addition, the patients in this study received treatments designed and administered by a single institution, minimizing the possibility of institution-based differences.
The present study is novel in its reporting on the treatment of LAPC with DP-IMRT. As a sizable minority of patients die from locoregional progression, the high doses delivered to gross tumor using this technique has the potential to offer improved local control rates leading to OS. DP-IMRT may minimize adverse events by allowing increased sparing of uninvolved tissue. Prospective studies are required to further establish the maximum tolerated dose and degree of LRC offered by DP-IMRT. The feasibility of such future work is suggested by tolerability of DP-IMRT among the patients receiving dose escalation to 54 Gy compared to those receiving 50.4 Gy. Further research in this area can conceivably institute a new treatment regimen with significantly improved local control afforded by the delivery of radiation levels not possible with traditional methods.
In conclusion, DP-IMRT provides a novel approach to the treatment of pancreatic cancer that appears to provide improved local control over conventional methods and results in fewer adverse events, likely due to lower doses of radiation delivered to OARs. Postradiation CA19-9 levels seem predictive of OS in this group of pancreatic cancer patients. Prospective studies are required to compare DP-IMRT to other radiation delivery modalities such as proton therapy or GTV-only irradiation and to investigate the therapeutic advantage of further dose escalation with DP-IMRT both in terms of LRC and improving resection rates.
This paper was funded by a UMDNJ Foundation Grant.