The influence of adjuvant radiotherapy dose on overall survival in patients with resected pancreatic adenocarcinoma

Authors


Abstract

BACKGROUND

Adjuvant radiotherapy (A-RT) for patients with resected pancreatic adenocarcinoma (PAC) is controversial. In the current study, the authors aim to determine whether there is an association between overall survival (OS) and A-RT dose.

METHODS

National Cancer Data Base (NCDB) data were obtained for all patients who received A-RT for resected PAC from 1998 through 2002. Univariate and multivariate survival analyses were performed along with Kaplan-Meier estimates for A-RT levels < 40 grays (Gy), 40 Gy to < 50 Gy, 50 Gy to < 55 Gy, and ≥ 55 Gy.

RESULTS

A total of 1385 patients met the inclusion criteria. The median age of the patients was 64 years (range, 29 years-87 years). All patients underwent surgical resection and A-RT with or without chemotherapy. A total of 231 patients were diagnosed with stage I disease, 273 were diagnosed with stage II disease, 734 were diagnosed with stage III disease, and 126 were diagnosed with stage IVA disease (according to the fifth edition of the American Joint Committee on Cancer); 21 were found to have an unknown stage of disease. The median A-RT dose was 45 Gy (range, 1.63 Gy-69 Gy). The median OS was 21 months (95% confidence interval [95% CI], 19 months-23 months). On multivariate analysis, an A-RT dose < 40 Gy (hazards ratio [HR], 1.30 [95% CI, 1.03-1.66]; P = .031), an A-RT dose of 40 Gy to < 50 Gy (HR, 1.17 [95% CI, 1.00-1.37]; P = .05), and an A-RT dose ≥ 55 Gy (HR, 1.44 [95% CI, 1.08-1.93]; P = .013) predicted worse OS compared with the reference category of 50 Gy to < 55 Gy.

CONCLUSIONS

A-RT doses of < 40 Gy, 40 Gy to < 50 Gy, and ≥ 55 Gy were found to be associated with an inferior OS. The dose of A-RT delivered appears to influence OS and a prospective study evaluating the addition of optimally delivered A-RT for patients with resected PAC is needed. Cancer 2013;119:2350–2357. © 2013 American Cancer Society.

INTRODUCTION

Pancreatic adenocarcinoma (PAC) is a devastating malignancy and the outcomes for this disease remain dismal.[1] The only opportunity for cure from PAC is surgical resection; however, the 5-year overall survival (OS) rate persists at < 20%.[2-7] Furthermore, the majority of patients are not surgical candidates because of locally advanced or metastatic disease at the time of presentation.[6-8]

The primary justification for the use of adjuvant radiotherapy (A-RT) in the United States comes from a trial conducted nearly 3 decades ago by the Gastrointestinal Tumor Study Group (GITSG). The GITSG study demonstrated an improvement in the median OS in patients with resected PAC with the addition of a 40-gray (Gy) split course of A-RT followed by adjuvant chemotherapy.[9, 10] This dose of A-RT is considered inferior to the modern dosing schedule of approximately 50 Gy delivered over 5 to 6 weeks. Results supporting the use of this modern A-RT dose have been presented by single-institution trials[11] as well as large retrospective reviews.[12-15]

The European Organization for Research and Treatment of Cancer (EORTC) was unable to reproduce the findings of the GITSG study using a 40-Gy split course of A-RT.[16] Furthermore, the European Study Group for Pancreatic Cancer-1 (ESPAC-1) trial demonstrated a detrimental effect of A-RT using a range of doses from 40 Gy to 60 Gy.[17, 18] However, conclusions from ESPAC-1 remain controversial.[19, 20]

Although chemotherapeutic variations have been examined in prospective clinical trials,[21] to the best of our knowledge few studies to date have measured the impact of A-RT dose on patient outcomes. One early-phase clinical trial examined the use of an escalated A-RT dose and found no benefit; however, there remains a paucity of such studies.[22] Given the heterogeneity of the prospective trial conclusions, the role of A-RT and the optimal dose range remain controversial.[12-15] The objective of the current study was to determine whether the dose of A-RT influences OS in patients with resected PAC, and to explore whether an optimal A-RT dose range exists.

MATERIALS AND METHODS

The patient population in the current study was obtained from the pancreatic Participant Use Data File (PUF) from the National Cancer Data Base (NCDB), which is the 1 of the world's largest clinical cancer registries.[23] The NCDB is supported by the American College of Surgeons and the American Cancer Society[23] and includes more than 1440 hospitals in the United States. Data available include patient demographics, pathologic characteristics, detailed staging, A-RT dose information, chemotherapy data, and OS data.

Emory University was granted alpha test user site status for the PUF, which includes all incident cases of pancreatic cancer reported to the NCDB for the 5-year period between 1998 and 2002. PUFs are entirely deidentified data files available to selected investigators at Commission on Cancer-approved institutions for the advancement of patient care. Results reported are in compliance with the privacy requirements of the Health Insurance Portability and Accountability Act of 1996 (HIPAA) as described in the Standards for Privacy of Individually Identifiable Health Information; Final Rule (45 CFR Parts 160 and 164). The use and publication of these data have been previously subject to peer review and approval by the NCDB.

There were 94,385 incident cases in the pancreatic PUF for the period between 1998 and2002. Of these, we initially selected 13,580 patients with a primary tumor site in the pancreas who had undergone definitive surgery on the primary site. From this group, we selected those patients who had reported OS data of any duration, which left 12,674 patients. We then selected patients who received external beam A-RT, leaving 5623 patients. We then selected patients for whom the radiation dose was not missing, leaving a total of 1489 patients. Patients with inaccurately coded A-RT doses (defined as inconceivable doses either > 400 Gy or < 1 Gy) were eliminated; this resulted in 11 patients (0.7% of the patients) being eliminated for inconceivable A-RT doses, leaving a total of 1478 patients. The patients with nonmetastatic disease were then selected, leaving a total of 1452 patients. Finally, patients treated with neoadjuvant therapy were excluded, resulting in the current study total of 1385 patients (Fig. 1).

Figure 1.

Patient selection schematic is shown. NCDB indicates National Cancer Data Base; OS, overall survival; A-RT, adjuvant radiotherapy; RT, radiotherapy.

Statistical Analysis

A frequency table for each categorical variable and summary statistics for each continuous variable were calculated to describe the patient-related and disease-specific variables calculated. Univariate survival analysis was performed by assessing the relation between each variable and OS using both the Kaplan-Meier log-rank test and a hazards ratio (HR) (with 95% confidence intervals [95% CIs]) derived through Cox proportional hazards modeling.

Martingale residual plots were used to identify potential nonlinear effects of all continuous covariates on OS. A nonlinear relation was observed for A-RT dose and we further categorized A-RT dose into the 4 different levels based on this relation. The individual association between categorized A-RT dose and each of the other covariates was analyzed using the chi-square test for categorical covariates and analysis of variance for continuous covariates.

Multivariate survival analysis started with all potential confounding variables from the univariate analysis and followed backward elimination steps in a Cox proportional hazards model with an alpha of .05 as removal criteria. On both univariate and multivariate analysis, patients receiving an A-RT dose between 50 Gy and 55 Gy were treated as the reference group. Facility volume was measured as the total number of resected cases in a given facility regardless of facility type with 10 as the unit of incremental increase. Facility types were community cancer programs, comprehensive community cancer programs, and academic/research cancer programs, which includes National Cancer Institute-designated comprehensive cancer centers.

The analyses were conducted using SAS statistical software (version 9.2; SAS Institute Inc, Cary, NC).

RESULTS

A total of 1385 patients were included. The median age of the patients was 64 years (range, 29 years-87 years), 53.1% were male, and 89.8% were white. All patients underwent surgical resection and received A-RT with or without chemotherapy. The majority of the patients (91.3%; 1265 of 1385 patients), received A-RT with concurrent chemotherapy and 7.8% (108 of 1385 patients) received A-RT alone. The use of chemotherapy was unknown in 0.9% of patients (12 of 1385 patients). Disease staging was performed according to the fifth edition of the American Joint Committee on Cancer staging system in which stages I and II included patients with lymph node-negative (N0) T1 to T3 disease, stage III included patients with lymph node-positive (N1) but still resectable disease (T1-T3), and stage IVA included patients with both resectable disease with tumor extension into adjacent organs and patients with unresectable disease. Staging groups for the included patients consisted of 231 with stage I disease, 273 with stage II disease, 734 with stage III disease, 126 with stage IVA disease, and 21 patients for whom the disease stage was missing. The median A-RT dose was 45 Gy (range, 1.63 Gy-69 Gy), and the median duration of treatment was 39 days (range, 1 day-100 days). A total of 164 patients (11.8%) received < 40 Gy, 634 (45.8%) received ≥ 40 Gy to < 50 Gy, 498 (36.0%) received ≥ 50 Gy to < 55 Gy, and 89 patients (6.4%) received ≥ 55 Gy. A detailed summary of patient characteristics is found in Table 1.

Table 1. Baseline Characteristics of All Patients
DemographicsN=1385  
  1. Abbreviations: AJCC, American Joint Committee on Cancer; ARCP, academic research cancer program; CCCP, comprehensive community cancer programs; CCP, community cancer program; Gy, grays; LN, lymph node.

Age, y   
 Mean63.18  
 Median (range)64 (29-87)  
Gender   
 No. of males (%)735 (53.1)  
Race   
 White1231 (89.8)  
 Other140 (10.2)  
 Data missing14  
Treatment Characteristics
 Facility type   
 CCP192 (13.9)  
 CCCP697 (50.3)  
 ARCP496 (35.8)  
Radiation dose, Gy Radiation dose category, Gy 
 Mean45.21<40164 (11.8)
 Median45≥40 to <50634 (45.8)
 Range1.63-69≥50 to <55498 (36.0)
  ≥5589 (6.4)
Radiation duration, d   
 Mean40.18  
 Median (range)39 (1-100)  
 Missing695  
Concurrent chemotherapy1265 (92.1)  
 A-RT alone108 (7.9)  
 Data missing12  
Tumor Characteristics   
 Stage (AJCC 5th ed)   
 I231 (16.9)  
 II273 (20.0)  
 III734 (53.8)  
 IVA126 (9.3)  
 Data missing21  
Tumor size, mm Size groupings, mm 
 Mean35.88≤20269 (21.6)
 Median (range)30.0 (1-750)>20 to ≤ 30399 (32.0)
  >30 to ≤ 40304 (24.4)
  >40274 (22.0)
  Data missing139
No of LNs examined LNs positive 
 Mean9.75Yes800 (61.7)
 Median (range)8 (0-60)No497 (38.3)
 Data missing100Data missing88
Histologic grade   
 Unspecified113 (8.2)  
 1165 (11.9)  
 2655 (47.3)  
 3 and 4452 (32.6)  
Resection margin   
 Negative899 (71.3)  
 Positive361 (28.7)  
 Data missing125  

At a median follow-up of 60 months, the median OS for all patients was 20 months (95% CI, 19 months-22 months). The median OS for patients receiving < 40 Gy was 15 months (95% CI, 11 months-20 months), was 20 months (95% CI, 19 months-22 months) for those patients receiving between 40 Gy to 50 Gy, and was 16 months (95% CI, 14 months-21 months) for those patients receiving > 55 Gy. Patients receiving between 50 Gy to 55 Gy had the longest median OS of 23 months (95% CI, 19 months-25 months). The Kaplan-Meier OS analysis for the entire cohort is shown in Figure 2 and that for each dose level is shown in Figure 3.

Figure 2.

The overall survival (OS) of all 1385 patients included in the analysis is shown. The median OS for all patients was 20 months (95% confidence interval, 19 months-22 months).

Figure 3.

The overall survival (OS) of patients receiving different adjuvant radiotherapy doses is reported. Patients were grouped as receiving < 40 grays (Gy) (median OS, 15 months [95% confidence interval (95% CI), 11 months-20 months]), ≥ 40 Gy to < 50 Gy (median OS, 20 months [95% CI, 19 months-22 months]), ≥ 50 Gy to < 55 Gy (median OS, 23 months [95% CI, 19 months-25 months]), or > 55 Gy (median OS, 16 months [95% CI, 14 months −21 months]).

On the univariate survival analysis, parameters found to be associated with a higher risk of death included an A-RT dose < 40 Gy, an A-RT dose ≥ 40 Gy to < 50 Gy, and an A-RT dose ≥ 55 Gy. Factors found to be significantly associated with a lower risk of death included facility volume, a negative surgical margin, negative lymph nodes, smaller tumor size, lower stage of disease, lower tumor grade, and younger patient age. The results of the univariate analysis can be found in Table 2. The univariate association analysis is summarized in Table 3. It can be observed that the surgical margin status, tumor size, and stage of disease were independent of the A-RT dose.

Table 2. Univariate Survival Analysis
CharacteristicN = 1385HR (95% CI)P
  1. Abbreviations: 95% CI, 95% confidence interval; AJCC, American Joint Committee on Cancer; ARCP, academic research cancer program; CCCP, comprehensive community cancer programs; CCP, community cancer program; Gy, grays; HR, hazards ratio; LN, lymph node.

  2. a

    Facility volume was measured as the total number of resected cases in a given facility regardless of facility type with 10 as the unit of incremental increase.

Age, y   
 <501740.662 (0.519-0.845)<.001
 50 to <655790.749 (0.617-0.910).004
 65 to <754790.802 (0.658-0.978).029
 ≥751531.0
Gender   
 Female6500.905 (0.803-1.020).102
 Male7351.0
Race   
 White12311.0
 Other1400.965 (0.790-1.178).726
Radiation dose, Gy   
 <401641.456 (1.194-1.776)<.001
 40 to <506341.167 (1.020-1.336).025
 50 to <554981.0
 ≥5581.383 (1.080-1.770).010
Concurrent chemotherapy   
 No1081.019 (0.810- 1.281).872
 Yes12651.0
Radiation duration, d   
 ≤351051.201 (0.904-1.595).206
 >35 to ≤402580.934 (0.739-1.182).571
 >40 to ≤451940.972 (0.760-1.245).824
 >451331.0
Surgical margin   
 Negative8990.739 (0.645-0.846)<.001
 Positive3611.0
LN positive   
 No4970.679 (0.596-0.773)<.001
 Yes8001.0
No. of LNs Examined   
 ≤129131.093 (0.952-1.254).207
 >123721.0
Tumor size, mm   
 ≤202690.601 (0.494-0.731)<.001
 >20 to ≤303990.741 (0.624-0.881)<.001
 >30 to ≤403040.859 (0.717-1.029).099
 >402741.0
Stage (AJCC 5th ed)   
 I2310.475 (0.372-0.606)<.001
 II2730.544 (0.431-0.687)<.001
 III7340.789 (0.646-0.964).02
 IVA1261.0
Histologic grade   
 Unspecified1130.592 (0.461-0.759)<.001
 11650.594 (0.482-0.732)<.001
 26550.810 (0.709-0.926).002
 3 and 44521.0
Facility type   
 CCP1921.167 (0.968-1.407).105
 CCCP6971.053 (0.923-1.201).440
 ARCP4961.0
 Facility volume (unit = 10)a13850.975 (0.961-0.989)<.001
Table 3. Variable Association With RT Dose Levels
 <40 Gy, No. (%)≥40 to <50 Gy, No. (%)≥50 to <55 Gy, No. (%)≥55 Gy, No. (%)Parametric Pa
N=164N=634N=498N=89
  1. Abbreviations: AJCC, American Joint Committee on Cancer; ARCP, academic research cancer program; CCCP, comprehensive community cancer programs; CCP, community cancer program; Gy, grays; LN, lymph node; RT, radiotherapy.

  2. a

    The parametric P value was calculated using analysis of variance for numerical covariates and the chi-square test for categorical covariates.

Age, y    .282
 ≤5021 (12.8)66 (10.4)69 (13.9)18 (20.2) 
 >50 to ≤ 6569 (42.1)263 (41.5)216 (43.4)31 (34.8) 
 >65 to ≤ 7556 (34.2)229 (36.1)165 (33.1)29 (32.6) 
 >7518 (11.0)76 (12.0)48 (9.6)11 (12.4) 
Surgical margin    .135
 Negative104 (71.2)425 (73.5)323 (70.4)47 (61.0) 
 Positive42 (28.8)153 (26.5)136 (29.6)30 (39.0) 
LN positive    .529
 No51 (34.0)240 (40.0)177 (38.1)29 (35.4) 
 Yes99 (66.0)360 (60.0)288 (61.9)53 (64.6) 
Tumor size, mm    .508
 ≤2029 (21.5)123 (21.4)100 (21.8)17 (21.8) 
 >20 to ≤ 3042 (31.1)195 (33.9)143 (31.2)19 (24.4) 
 >30 to ≤ 4026 (19.3)143 (24.9)113 (24.7)22 (28.2) 
 >4038 (28.2)114 (19.8)102 (22.3)20 (25.6) 
Concurrent chemotherapy    .032
 No22 (13.7)42 (6.7)37 (7.5)7 (7.9) 
 Yes139 (86.3)588 (93.3)456 (93.5)82 (92.1) 
Stage (AJCC 5th ed)    .826
 I24 (14.9)101 (16.1)94 (19.2)12 (14.1) 
 II29 (18.0)131 (20.8)96 (19.6)17 (20.0) 
 III90 (56.0)337 (53.6)260 (53.2)47 (55.3) 
 IVA18 (11.2)60 (9.5)39 (8.0)9 (10.6) 
Histologic grade    .031
 Unspecified19 (11.6)40 (6.3)45 (9.0)9 (10.1) 
 120 (12.2)66 (10.4)61 (12.3)18 (20.2) 
 265 (39.6)317 (50.0)239 (48.0)34 (38.2) 
 3 and 460 (36.6)211 (33.3)153 (30.7)28 (31.5) 
Facility type    .016
 CCP25 (15.2)86 (13.6)68 (13.7)13 (14.6) 
 CCCP97 (59.2)310 (49.0)236 (47.4)54 (60.7) 
 ARCP42 (25.6)238 (37.5)194 (39.0)22 (24.7) 

On the multivariate survival analysis, an A-RT dose < 40 Gy (HR, 1.30 [95% CI, 1.03-1.66]; P = .031), an A-RT dose ≥ 40 Gy to < 50 Gy (HR, 1.17 [95% CI, 1.00-1.37]; P = .05), and an A-RT dose ≥ 55 Gy (HR, 1.44 [95% CI, 1.08-1.93]; P = .013) were all found to be significantly associated with worse OS. In addition to radiation dose level, age, surgical margin status, stage of disease, tumor size, tumor grade, and facility volume were all found to be significant on multivariate analysis (Table 4).

Table 4. Multivariate Survival Analysis
CharacteristicHR (95% CI)P
  1. Abbreviations: 95% CI, 95% confidence interval; AJCC, American Joint Committee on Cancer; Gy, grays; HR, hazard ratio.

  2. a

    Facility volume was measured as the total number of resected cases in a given facility regardless of facility type with 10 as the unit of incremental increase.

Age, y  
 <500.62 (0.47-0.82)<.001
 50 to ≤650.73 (0.58-0.91).005
 65 to ≤750.78 (0.62-0.97).028
 >751.0
Radiation dose, Gy  
 <401.30 (1.03-1.66).031
 ≥40 to < 501.17 (1.00-1.37).05
 ≥50 to < 551.0
 ≥551.44 (1.08-1.93).013
Surgical margin  
 Negative0.73 (0.63-0.86)<.001
 Positive1.0
Stage (AJCC 5th ed)  
 I0.61 (0.45-0.82)<.001
 II0.64 (0.48-0.85).002
 III0.96 (0.75-1.23).743
 IVA1.0
Tumor size, mm  
 ≤200.55 (0.44-0.69)<.001
 >20 to ≤ 300.73 (0.60-0.89).002
 >30 to ≤ 400.75 (0.62-0.92).006
 >401.0
Histologic grade  
 Unspecified0.52 (0.37-0.73)<.001
 10.58 (0.45-0.69)<.001
 20.83 (0.71-0.97).017
 3 and 41.0
Facility volume: unit = 10a0.98 (0.97-1.00).014

The duration of time over which each of the respective A-RT doses was delivered is summarized in Table 5.

Table 5. Duration of RT Administration
RT Duration, Days<40 Gy, No. (%)40 to <50 Gy, No. (%)≥50 to <55 Gy, No. (%)≥55 Gy, No. (%) 
N=81N=342N=236N=31Parametric Pa
  1. Abbreviations: Gy, grays; RT, radiotherapy.

  2. a

    The parametric P value was calculated using analysis of variance for numerical covariates and the chi-square test for categorical covariates.

<104 (5.0)0 (0)0 (0)0 (0)<.001
10-2018 (22.2)1 (0.3)0 (0)0 (0) 
21-3019 (23.5)6 (1.8)1 (0.4)0 (0) 
31-4019 (23.5)218 (63.7)77 (32.6)0 (0) 
41-5016 (19.8)96 (28.1)132 (55.9)22 (71.0) 
51-602 (2.4)11 (3.2)17 (7.2)4 (12.9) 
>603 (3.7)10 (2.9)9 (3.8)5 (16.1) 
Mean duration30.6239.6242.8451.1 

DISCUSSION

Despite 3 prospective randomized trials, the role of A-RT in patients with resected PAC remains controversial. The majority of the recent series examining A-RT in patients with resected PAC support doses of approximately 50 Gy to 55 Gy, which differs from that used in past prospective trials.[14, 21] The purpose of the current study was to analyze the impact of the A-RT dose on OS in patients with resected PAC and explore whether an optimal A-RT dose exists.

Much of the current rationale for A-RT comes from the landmark GITSG 91-73 analysis.[10] The A-RT dose in the GITSG analysis was 40 Gy delivered over 20 fractions as a split course and resulted in a median OS of 20 months.[9, 10] The role of A-RT in patients with PAC was again examined in EORTC 40891, which also used a 40-Gy split course. The median OS between the 2 arms was not statistically different in this EORTC study, which led to the conclusion that the routine use of A-RT was not warranted.[16, 24] Furthermore, ESPAC-1 indicated a survival detriment when using a 40-Gy to ;60-Gy split course of A-RT.[17, 18] The ESPAC-1 study design has received substantial criticism since its publication given the quality of the A-RT delivery is unknown.[19]

More recently, the Radiation Therapy Oncology Group (RTOG) 9704 trial examined the addition of gemcitabine chemotherapy to 5-fluorouracil (5-FU).[21] High-quality A-RT was delivered at a dose of 50.4 Gy at 1.8 Gy per fraction with continuous infusion of 5-FU in both arms. To the best of our knowledge, this was the first large-scale trial to use a more contemporary A-RT dosing and fractionation schedule.[21] The outcomes were similar to those of the current series and GITSG with a median OS of 20.5 months. On quality assurance review, nearly 50% of the A-RT in RTOG 9704 deviated from protocol guidelines. Abrams et al conducted a secondary analysis of RTOG 9704 and demonstrated that A-RT when not delivered as per protocol was a negative predictor of OS on multivariate analysis.[25] To the best of our knowledge, Abrams et al was the first series to demonstrate that A-RT quality and variation could impact OS.

In addition to prospective trials, several large retrospective analyses have been conducted. The recent Mayo Clinic and Johns Hopkins Hospital collaborative retrospective case series by Hsu et al examined 1045 patients with resected PAC, with 530 patients (50.7%) receiving the combination of 5-FU and RT.[14] The patients in this series also received high-quality A-RT of 50.4 Gy at 1.8 Gy per fraction. Investigators demonstrated that A-RT was associated with an improved OS among all patients and in all subgroups regardless of age, tumor size, surgical margin status, lymph node status, and tumor differentiation.[14]

The current series supports the hypothesis that the dose of A-RT in patients with resected PAC appears to influence OS. Furthermore, it can be seen in the current analysis that patients treated with doses between 50 Gy to 55 Gy had the longest median OS. The current series, along with the secondary analysis of RTOG 9704 by Abrams et al,[25] have provided supportive evidence that A-RT parameters impact OS. These data support the hypothesis that the lack of an A-RT benefit shown in past trials may have been secondary to suboptimal delivery of A-RT. In addition, it should be noted that facility volume did appear to influence OS, thereby reflecting the complexity of PAC management and the importance of facility experience and treatment quality.

The current series demonstrates a significant association between patients treated with A-RT doses < 40 Gy and inferior OS. It is likely these patients did not complete a full course of A-RT because of disease progression, medical comorbidities, or a combination of these factors. Patients treated with A-RT doses > 55 Gy also demonstrated inferior OS compared with the reference cohort that was treated with 50 Gy to 55 Gy. This could potentially be because of increased toxicity or adverse imaging features on computed tomography simulation that motivated doses > 55 Gy. Patients treated to doses of 40 Gy to < 50 Gy also demonstrated an inferior OS when compared with the reference cohort of patients treated with 50 Gy to < 55 Gy. These 2 groups both had the largest patient numbers, comparable patient characteristics, and similar use of chemotherapy. This difference remained significant on multivariate analysis and was independent of tumor size, stage of disease, tumor grade, surgical margin status, and facility volume. This OS difference is supportive evidence that A-RT dose appears to influence OS.

The results of the current analysis should be interpreted with caution because of some important limitations beyond the retrospective design of the study. First, although the number of patients in the current analysis was large, given the total number of patients in the NCDB this was a relatively small fraction. A percentage of the patients in the database did have missing, incomplete, or inaccurately coded information regarding A-RT and consequently were eliminated. Excluding this large number of cases could have introduced a source of selection bias into the analysis. We applied an extensive array of statistical tools in an attempt to offset this bias, including a propensity score weighted analysis and analysis of all characteristics of the eliminated patients, with no significant impact on the overall conclusions noted. In addition, every attempt to minimize the practice of eliminating patients based on perceived coding errors was made, which explains the rather unusual A-RT dose range of 1.63 Gy to 69 Gy. Although those patients with perceived unusual doses could have been excluded, including any conceivable dose was our attempt to present the data as purely as possible.

An additional limitation of the current series is the lack of information regarding the precise use of chemotherapy. It should be noted that approximately 8% of the patients were coded as having received no concurrent chemotherapy with the A-RT, which is likely secondary to prohibitive medical comorbidities, patient refusal, or inaccurate coding. In addition, a lack of a statistically significant difference in OS was demonstrated on the univariate analysis when comparing patients who did not receive chemotherapy with those who did. This finding was attributed primarily to the large discrepancy in patient numbers present in these 2 cohorts, thus making a reliable statistical comparison difficult. It also should be noted that patients treated in different facility types received differing A-RT doses. This likely reflects differences in institutional adoptions of novel A-RT dose recommendations or the experience of the attending radiation oncologist. The lack of other known prognostic factors in the NCDB, such as carbohydrate antigen 19-9 (CA19-9) level and performance status, is also a limitation. In addition, it should be noted that certain variables had large numbers of missing values, including RT duration, number of lymph nodes, tumor size, and surgical margin status. Finally, specific information obtained from the computed tomography simulation scan for the purpose of A-RT planning is unknown. Those patients receiving > 55 Gy could have received this because of residual disease noted on planning computed tomography scan, potentially influencing OS, despite no reported difference in surgical margin status.

Although it is essential to consider the limitations of the current study, it is also important to take note of the strengths and novelty of the analysis. The current series included a large number of patients, with an array of A-RT doses delivered in a variety of facility types. General chemotherapy, A-RT, pathologic parameters, and facility volume differences are known and accounted for. Such a comparison of A-RT dose levels would be difficult to complete with a single institutional database, given the probable absence of a wide range of A-RT doses. Furthermore, it is highly unlikely that the impact of the A-RT dose variation on OS would be addressed in a prospective randomized trial.

In the most general sense, the current series demonstrates that the manner in which A-RT is delivered in patients with resected PAC appears to influence OS, which is also supported by the secondary analysis of RTOG 9704.25 The results of the current series specifically indicate that A-RT dose impacts OS, which is a relatively easily adoptable and verifiable A-RT component. These data bring into further question previously conducted prospective trials examining the addition of A-RT in patients with resected PAC. Finally, this series supports the significance of the current prospective randomized trial, RTOG 0848, which applies 50.4 Gy delivered at 1.8 Gy per fraction of high-quality A-RT to select patients after resection and adjuvant chemotherapy.

Conclusions

The current study presents a large outcomes-based analysis of patients treated with A-RT for resected PAC. Based on these data, the optimal dose of A-RT appears to fall between 50 Gy and 55 Gy. These data support the hypothesis that the characteristics of A-RT delivery influence OS. In addition, these data support the most common and currently used A-RT dose fractionation schedule and underscore the importance of a prospective investigation into the role of A-RT in resected PAC using modern A-RT delivery. Ongoing prospective trials (such as RTOG 0848) will define the true role of high-quality A-RT in patients with resected PAC.

FUNDING SUPPORT

This work was supported in part by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR000454 and TL1TR000456. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

Ancillary