Role of Adjuvant Therapy in Resected Stage IA Subcentimeter (T1a/T1b) pancreatic Cancer

Transcription

1 Role of Adjuvant Therapy in Resected Stage IA Subcentimeter (T1a/T1b) pancreatic Cancer Walid L. Shaib, MD 1 ; Amit Surya Narayan, MD 2 ; Jeffrey M. Switchenko, PhD 3 ; Sujata R. Kane, NP 3 ; Christina Wu, MD 1 ; Mehmet Akce, MD 1 ; Olatunji B. Alese, MD 1 ; Pretesh R. Patel, MD 4 ; Shishir K Maithel, MD 5 ; Juan M. Sarmiento, MD 5 ; David A. Kooby, MD 5 ; and Bassel F. El-Rayes, MD 1 BACKGROUND: The standard of care for patients with resected stage I to stage III pancreatic ductal adenocarcinoma (PDAC) is adjuvant gemcitabine-based chemotherapy. The role of adjuvant treatment in patients with subcentimeter, stage IA PDAC is unknown. The current study evaluated the effect of adjuvant treatment on survival outcomes among patients with American Joint Committee on Cancer/International Union Against Cancer stage IA (T1N0) resected PDAC using the National Cancer Data Base (NCDB). METHODS: A retrospective review of the NCDB was conducted for patients diagnosed with T1 (tumor limited to the pancreas and measuring 2 cm in greatest dimension), lymph node negative (N0), resected PDAC between 2004 and Patient demographics, histology, adjuvant treatment, and survival trends were examined. Kaplan-Meier analysis and log-rank tests were performed to determine the unadjusted association between overall survival (OS), tumor size, and treatment. RESULTS: A total of 876 patients met the inclusion criteria. The patients had a mean age of 66.2 years (range, years); approximately 83.3% were white (730 patients) and 53.1% were female (465 patients). Approximately 45.9% of the patients had moderately differentiated tumor histology (402 patients); 70.0% (613 patients) had tumors measuring 1 to 2 cm (T1c) and 30.0% (263 patients) had tumors measuring <1 cm (T1a/T1b). Approximately 94.2% of patients had negative surgical margins (815 patients) and 46.9% (410 patients) received adjuvant therapy. The median OS was significantly different for patients who received adjuvant therapy compared with patients who did not (70.7 months vs 46.9 months; P=.0001). For patients with tumors measuring <1 cm, survival was not found to be significantly different between patients who received adjuvant treatment compared with those who did not (not reached vs 85.3 months; P=.54). In the multivariable analysis, none of the covariates (treatment group, Charlson-Deyo Score, age, insurance, and facility status) demonstrated significant differences for patients with tumors measuring <1 cm. CONCLUSIONS: The current study is the first to demonstrate no survival benefit for adjuvant therapy in patients with resected subcentimeter PDAC. Cancer 2019;125: American Cancer Society. KEYWORDS: adjuvant therapy, pancreatic cancer, subcentimeter, survival. INTRODUCTION Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest diseases, with a mortality rate approaching its incidence. 1-3 Due to the difficulty in diagnosing the disease at an early stage and the high potential for metastasis, the majority of patients present with an advanced stage of disease, at which time cure is not possible. 2,4 Surgery is the only curative approach, and only 15% to 20% of patients are candidates for surgical resection. 2,4 Approximately 66% to 92% of those patients with resected PDAC develop a disease recurrence within 2 years. 4-6 The 2017 American Joint Committee on Cancer/International Union Against Cancer staging system defines stage IA disease as a classification of T1 N0. 7 T1 is classified further into T1a (tumors measuring <0.5 cm), T1b (tumors measuring cm), and T1c (tumors measuring 1-2 cm). 7 A Surveillance, Epidemiology, and End Results database analysis of 8960 patients demonstrated median survival durations for patients with stage IA, stage IB, stage IIA, stage IIB, and stage III disease of 38 months, 24 months, 18 months, 17 months, and 14 months, respectively. 8 In this analysis, approximately 5% of the patients had stage IA (T1N0) disease. Adjuvant therapy has been an area of research over the past several decades. 9 Adjuvant therapy typically is recommended for patients with PDAC after surgical resection. Two landmark trials, the Adjuvant chemotherapy with gemcitabine and long-term outcomes among patients with resected pancreatic cancer (CONKO-001) randomized trial. and the Corresponding author: Walid L. Shaib, MD, Department of Hematology and Oncology, Winship Cancer Institute, Emory University, 1365 Clifton Rd NE, Atlanta, GA 30322; wshaib@emory.edu 1 Department of Hematology and Oncology, Winship Cancer Institute, Emory University, Atlanta, Georgia; 2 Department of Internal Medicine, Emory University, Atlanta, Georgia; 3 Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, Atlanta, Georgia; 4 Department of Radiation Oncology, Winship Cancer Institute, Emory University, Atlanta, Georgia; 5 Department of Surgical Oncology, Winship Cancer Institute, Emory University, Atlanta, Georgia. The first 2 authors contributed equally to this article. DOI: /cncr.31787, Received: May 28, 2018; Revised: August 4, 2018; Accepted: August 7, 2018, Published online November 20, 2018 in Wiley Online Library (wileyonlinelibrary.com) 57

2 European Study Group for Pancreatic Cancer (ESPAC- 4), established comparison of adjuvant gemcitabine and capecitabine with gemcitabine monotherapy in patients with resected pancreatic cancer (ESPAC-4): a multicentre, open-label, randomised, phase 3 trial. 2,10 These trials included patients across all locally resected disease stages, but the majority of patients had stage II and stage III disease. In the CONKO-001 trial, patients with resected PDAC were randomized to receive adjuvant single-agent therapy with gemcitabine versus observation. 2 The 5-year overall survival (OS) rates were 20.7% versus 10.4 %, respectively. 2 Of the 179 patients who received gemcitabine, 4% (7 patients) had T1 disease and 82% (146 patients) had T3 PDAC. Of the 175 patients who were randomized to the observation arm, 4% (7 patients) had T1 disease and 83% (146 patients) had T3 PDAC. 2 The results of this trial supported the use of adjuvant single-agent gemcitabine (P=0.01), and defined the standard of care for the adjuvant treatment of patients with resected PDAC. 2,10 The ESPAC-4 trial was a multicenter, open-label, randomized, phase 3 trial that randomized patients to adjuvant gemcitabine and capecitabine combination therapy compared with gemcitabine single-agent adjuvant therapy. 10 The median OS for patients who received adjuvant gemcitabine and capecitabine was 28.0 months compared with 25.5 months for patients treated with gemcitabine alone (P=.032). The median tumor size was 3 cm. Of the 364 patients who received adjuvant gemcitabine and capecitabine, 4% (15 patients) had stage I PDAC and 90% (326 patients) had stage III disease. Among the patients treated with gemcitabine alone, 2% (7 patients) had stage I disease. 10 The ESPAC-4 trial demonstrated that adjuvant treatment with the combination of gemcitabine and capecitabine is superior to single-agent gemcitabine, thereby establishing the current standard of care for the adjuvant treatment of patients with PDAC. 10 The major adjuvant therapy trials, including ESPAC-1, ESPAC-3/ version 1, ESPAC-3/version 2, ESPAC-4, and CONKO- 001, with the percentage of patients with stage 1 disease assessed are summarized in Table 1. 2,10-13 Patients with resected stage I PDAC are underrepresented in the PDAC adjuvant trials. This most likely is because of the low prevalence of this disease at the early stage. To our knowledge, there currently are no data supporting the use of adjuvant therapy specifically in patients with resected stage I PDAC. 14 The relatively small sample size of patients with stage IA PDAC in the CONKO-001 and ESPAC-4 trials precludes a meaningful subgroup analysis with which to answer the question of whether adjuvant treatment in patients with stage IA disease would confer a survival benefit. To address this question, the current study used the National Cancer Data Base (NCDB) analysis to evaluate the effect of adjuvant treatment on survival among patients with subcentimeter, resected PDAC. TABLE 1. Summary of Adjuvant Treatment Trials for Patients With Resected Pancreatic Cancer Study Treatment Arms Primary Outcome, Months P No. No. of T1Nx Cases ESPAC a Chemotherapy vs CRT OS, 20.1 vs vs 27 ESPAC-3 v ,a Chemotherapy (5-FU/gemcitabine) OS, 43.1 vs vs 19 vs 34 b vs observation ESPAC Gemcitabine vs gemcitabine plus OS, 28 vs vs 15 (T1N0) capecitabine CONKO Gemcitabine vs observation DFS, 13.4 vs 6.9 < vs 7 RTOG Gemcitabine plus CRT vs 5-FU OS, 20.5 vs vs 43 c plus CRT GITSG Observation vs CRT followed by OS, 11 vs Not published chemotherapy EORTC ,a Observation vs CRT OS, 19 vs vs 13 CONKO Gemcitabine vs gemcitabine plus DFS, 11.4 vs Not published erlotinib JSAP Gemcitabine vs observation DFS, 11.4 vs vs 4 JASPAC S1 vs gemcitabine 5-y OS, 44.1% vs 24.4% <.0001 d vs 8 Abbreviations: 5-FU, 5-fluorouracil; CONKO, Adjuvant Chemotherapy with Gemcitabine and Long-term Outcomes among Patients with Resected Pancreatic Cancer; CRT, chemoradiation; DFS, disease-free survival; EORTC, European Organization for Research and Treatment of Cancer; ESPAC, European Study Group for Pancreatic Cancer; GITSG, Gastrointestinal Tumor Study Group; JASPAC, Japanese Adjuvant Study Group of Pancreatic Cancer; JSAP, Japanese Study Group of Adjuvant Therapy for Pancreatic Cancer; OS, overall survival; RTOG, Radiation Therapy Oncology Group. a The ESPAC-1 study noted that some of the size data were missing. The ESPAC-3 and EORTC trials examined periampullary cancers, including pancreatic ductal adenocarcinoma. b Numbers represent stage I in patients with periampullary cancers. c Included T1 and T2 disease classifications. d Noninferiority trial. 58

3 Adjuvant Therapy for Subcentimeter PDAC/Shaib et al MATERIALS AND METHODS Data Source and Study Population A retrospective analysis was performed using the NCDB for patients with resected, T1, stage IA PDAC (ie, tumor confined to the pancreas and measuring 2 cm in dimension). Eligibility criteria included adenocarcinoma of the pancreas, stage I disease, no preoperative therapy, and treatment received between 2004 and 2013 (Table 2). Patient demographics and tumor characteristics such as race, sex, age, surgical margins, tumor grade, histology, treatment modalities, and survival trends were examined. Histology codes of (ICD-O3): 8010, 8020, 8140, 8143, 8210, 8211, 8230, 8260, 8261, 8262, 8263, 8490, 8500, 8514, 8521, 8570, and 8572 were identified. The histology code 8453 for intraductal papillary mucinous neoplasm (IPMN) was used to identify patients with IPMN with coexistent codes. Patients were stratified into those with a tumor size of <1 cm and those with a tumor size of 1 to 2 cm, and by the type of adjuvant treatment received (no therapy, chemotherapy, concurrent chemotherapy/ radiotherapy, or radiotherapy). Patients who underwent neoadjuvant therapy or had an unknown treatment sequence were excluded from the current analysis. TABLE 2. Diagram of Study Population Selection for Stage IA Pancreatic Cancer Selection and Exclusion Criteria Sample Size Excluded NCDB pancreas PUF cancer cases 309,709 Include pancreatic adenocarcinoma 254,562 55,147 and PDAC Include patients with primary tumor 194,674 59,888 site of the pancreatic head, body, or tail Include pathologic T classification ,540 Include pathologic N classification Include tumor size 0-2 cm Include patients with resected cancer Exclude if missing information regarding survival time Include sequence number = 0 or Exclude class of case = Exclude unknown chemotherapy/ unknown RT cases, unknown chemotherapy/no RT cases, no chemotherapy/unknown RT cases, any neoadjuvant chemotherapy or neoadjuvant RT cases, chemotherapy sequence unknown/ no RT cases, chemotherapy sequence unknown/adjuvant RT cases, no chemotherapy/rt sequence unknown cases, and any unknown chemotherapy or unknown RT Abbreviations: NCDB, National Cancer Data Base; PDAC, pancreatic ductal adenocarcinoma; PUF, Participant User Files; RT, radiotherapy. Statistical Analysis Descriptive statistics were generated for each covariate including the mean, median, range, and standard deviation for continuous variables and the frequency and percentage for categorical variables. The endpoint, OS, was defined as the time from diagnosis to the time of death or last known time alive, at which point surviving patients were censored. OS was estimated using the Kaplan-Meier method, and patient groups (those undergoing adjuvant therapy vs surgery alone) were compared using log-rank tests. Univariate Cox proportional hazards models were fit for OS as a function of treatment group (adjuvant treatment vs surgery alone), sex, race (black, white, or other), age, tumor grade, Charlson-Deyo Score ( 1 or 0), surgical margins (positive vs negative), tumor site (pancreatic head, body, or tail), insurance status (private, government insurance, or not insured/unknown), facility type (nonacademic vs academic), lymphovascular invasion (LVI), and tumor size (<0.5 cm, cm, and 1-2 cm). Multivariable Cox models were considered that included treatment, controlling for the above covariates, with the exception of LVI due to its level of missingness. In addition, the analysis was stratified by tumor size into subcentimeter versus 1 to 2 cm in size. Model assumptions were checked and verified. The analysis was performed using SAS statistical software (version 9.4; SAS Institute Inc, Cary, North Carolina), and significance was assessed at the.05 level. RESULTS Patient Characteristics A total of 1589 patients with stage IA PDAC underwent surgical resection between 2004 and Of these, 876 patients met the criteria for study inclusion (Table 2). The mean age at the time of diagnosis was 66.2 years (range, years). The majority of the patients were white (730 patients; 83.3%) and 10.5% were black (92 patients). Females constituted 53.1% (465 patients) of the study population. The majority of patients had moderately differentiated tumors (402 patients; 45.9%), followed by poorly/undifferentiated tumors (196 patients; 22.4%), well-differentiated tumors (174 patients; 19.9%), and undetermined differentiation (104 patients; 11.9%). Patients with tumors measuring 1 to 2 cm constituted 70.0% of the study sample (613 patients), whereas patients with subcentimeter tumors constituted 30.0% (263 patients). The majority of patients had negative surgical margins (815 patients; 94.2%). Patient characteristics across treatment categories are summarized in 59

4 TABLE 3. Characteristics of Patients (N = 876) Variable Level No. % Treatment type No chemotherapy, no RT No chemotherapy, adjuvant RT Adjuvant chemotherapy, no RT Adjuvant chemotherapy, adjuvant RT Tumor size, cm Race White Black Other/unknown Sex Male Female Charlson-Deyo Score Grade Well differentiated Moderately differentiated Poorly differentiated/undifferentiated Cell type not determined Lymphovascular invasion Not present Present Missing data 514 Surgical margins Negative Positive Missing data 11 Primary tumor site in pancreas Head Body Tail Facility type Community cancer program/comprehensive community cancer program/integrated network cancer program Academic/research program Missing data 11 Insurance status Not insured/unknown Private Medicaid/Medicare/other government Age at diagnosis, y Mean Median 67 Minimum 32 Maximum 90 SD Missing data 0 Abbreviations: RT, radiotherapy; SD, standard deviation. Table 3. Eighty one of the 876 patients (9.2%) had coexistent diagnosis of IPMN: 38 of 613 patients (6.2%) with tumors measuring 1 to 2 cm and 43 of 263 (16.3%) patients with subcentimeter tumors. Patients treated with surgical resection alone (ie, without adjuvant treatment) constituted 53.2% of the study population (466 patients). Adjuvant chemotherapy without radiotherapy was used in 27.9% of patients (244 patients), radiotherapy without chemotherapy was used in 1.3% of patients (11 patients), and adjuvant chemoradiation was used in 17.7% of patients (155 patients). 60

5 Adjuvant Therapy for Subcentimeter PDAC/Shaib et al Survival Outcomes For patients with stage IA (T1N0) PDAC who received adjuvant therapy, the median OS was 70.7 months compared with 46.9 months for patients who did not receive adjuvant therapy (P=.0001); the 12-month, 24-month, and 60-month OS rates for patients who received adjuvant therapy were 93.6%, 77.5%, and 53.8%, respectively, compared with 79.4%, 67.7%, and 41.2%, respectively, for those who did not (P=.0001) (Fig. 1). When stratified for a tumor size of 1 to 2 cm, the median OS was 64.5 months for patients who received adjuvant therapy compared with 30.8 months for patients who did not receive adjuvant therapy (P<.0001); the 12-month, 24-month, and 60-month OS rates were 93.0%, 75.1%, and 52.7%, respectively, for patients who received adjuvant therapy compared with 72.5%, 61.0%, and 31.0%, respectively, for those who did not (P<.0001). For patients with subcentimeter tumors (T1a/T1b), the median OS for those who received adjuvant therapy was not reached compared with 85.3 months for patients who did not receive any adjuvant therapy (P=.54); the 12-month, 24-month, and 60-month OS rates were 95.5%, 86.4%, and 57.2%, respectively, for patients who received adjuvant therapy compared with 91.2%, 79.1%, and 59.1%, respectively, for those who did not (Fig. 2). For patients with stage IA disease who received adjuvant chemotherapy alone (no radiotherapy), the median OS was 78.7 months compared with 46.9 months for Figure 1. Kaplan-Meier curve of patients with resected, stage IA pancreatic cancer comparing the overall survival of patients who received adjuvant treatment with that of patients who did not receive adjuvant treatment (P=.0001). 95% CI indicates 95% confidence interval. 61

6 Figure 2. Kaplan-Meier curve of patients with resected, stage IA subcentimeter (T1a/T1b) pancreatic cancer comparing the overall survival of patients who received adjuvant treatment with that of patients who did not receive adjuvant treatment (P=.54). 95% CI indicates 95% confidence interval. patients who did not receive adjuvant therapy (P<.0001). The 12-month, 24-month, and 60-month OS rates were 95.8%, 84.2% and 58.3%, respectively (P<.0001). When stratified for tumor sizes of 1 to 2 cm, the median OS for patients who received adjuvant chemotherapy alone was 78.4 months (vs 30.8 months; P<.0001); the 12-month, 24-month, and 60-month OS rates were 95.5%, 82.1%, and 56.3%, respectively (P<.0001). For patients with subcentimeter tumors, the median OS was not reached for those who received adjuvant chemotherapy without radiotherapy (vs 85.3 months; P=.078); the 12-month, 24-month, and 60-month OS rates were 96.8%, 91.0%, and 64.6%, respectively (P=.078). For patients with stage IA disease who received any adjuvant radiotherapy, the median OS was 45.6 months (vs months; P=.24). The 12-month, 24-month, and 60-month OS rates were 90.3%, 68.2%, and 47.1%, respectively, versus 79.4%, 67.7%, and 41.2%, respectively for those who did not receive adjuvant radiotherapy (P=.24). When stratified for tumor sizes of 1 to 2 cm, the median OS for patients who received adjuvant radiotherapy was 56.7 months (vs 30.8 months; P=.002). The 12- month, 24-month, and 60-month OS rates were 90.0%, 66.5%, and 48.0%, respectively, which was statistically significant (P=.002). For patients with subcentimeter tumors, the median OS was 40.3 months for patients who 62

7 Adjuvant Therapy for Subcentimeter PDAC/Shaib et al received adjuvant radiotherapy (vs 85.3 months; P=.18). The 12-month, 24-month, and 60-month OS rates were 92.3%, 76.9%, and 44.7%, respectively, which were not significantly different when compared with patients who did not receive adjuvant radiotherapy (P=.18). Univariate and Multivariable Analyses Univariate analysis of the patients with stage IA PDAC who underwent surgical resection demonstrated a significant survival difference for patients receiving adjuvant therapy versus those treated with surgery alone (hazard ratio [HR], 0.68; 95% confidence interval [95% CI], [P<.001]), those with a tumor size of 1 to 2 cm versus <0.5 cm (P<.001), those with a Charlson- Deyo Score of 1 versus 0 (HR, 1.33; 95% CI, [P=.005]), tumor grade (P<.001), those with positive surgical margin status versus negative surgical margin status (HR, 2.05; 95% CI, [P<.001]), those with a primary tumor location in the head versus the tail of the pancreas (HR, 1.66; 95% CI, [P<.001]), having private insurance versus government insurance (HR, 0.7; 95% CI, [P<.001]), and age (HR, 1.02; 95% CI, [P=.005]). There were no survival differences noted for race, sex, LVI, and facility type (Table 4). Univariate analysis stratified to a tumor size of <1 cm demonstrated no survival difference for adjuvant treatment (HR, 0.86; 95% CI, [P=.54]), race (P=.9), Charlson-Deyo Score of 1 versus 0 (HR, 1.50; 95% CI, [P=.056]), tumor grade (P=.32), positive surgical margin status versus negative surgical margin status (P=.48), primary tumor location in the head versus the tail of the pancreas (P=.38), and having private insurance versus government insurance (P=.17). A survival difference was detected for LVI (HR, 4.94; 95% CI, [P=.004]) and age (HR, 1.02; 95% CI, [P=.016]). The multivariable analysis of patients with stage IA PDAC who underwent surgical resection demonstrated a significant survival difference for patients receiving adjuvant therapy versus surgery alone (HR, 0.69; 95% CI, [P<.001]), those with a Charlson-Deyo Score of 1 versus 0 (HR, 1.31; 95% CI, [P=.008]), those with poorly differentiated tumors versus tumors in which a cell type was not determined (HR, 1.55; 95% CI, [P=.013]), those with a positive surgical margin versus a negative surgical margin (HR, 2.13; 95% CI, [P<.001]), those with a primary tumor location in the head versus the tail of the pancreas (HR, 1.61; 95% CI, [P=.002]), and age (HR, 1.02; 95% CI, [P=.007]). There was no survival difference noted with regard to race (P=.28), sex (P=.68), facility type (nonacademic vs academic; P=.13), and insurance status (not insured/unknown vs government insurance; P=.73). In a multivariable analysis stratified for a tumor size of <1 cm, there was no survival difference noted with regard to adjuvant therapy versus surgery alone (HR, 0.84; 95% CI, [P=.53]), race (P=.75), sex (P=.63), Charlson-Deyo Score of 1 versus 0 (HR, 1.45; 95% CI, [P=1]), well-differentiated tumor grade versus cell type not determined (P=.82), positive surgical margin versus negative surgical margin (P=.4), tumor location in the head versus the tail of the pancreas (P=.29), age (HR, 1.01; 95% CI, [P=.37]), facility type (nonacademic vs academic: HR, 1.35; 95% CI, [P=.21]), and insurance status (not insured/unknown vs government insurance; P=.98) (Table 5). DISCUSSION The current standard of care for patients with resected PDAC is adjuvant combination chemotherapy. 10 Previous adjuvant trials (Table 1) have evaluated patients across all stages of PDAC after surgical resection. A minority of the patients in these trials had stage I disease (including T1 and T2 disease). 2,10 To the best of our knowledge, there are no randomized trials to date evaluating adjuvant treatment specifically for patients with stage IA PDAC because the incidence of this stage of disease at the time of diagnosis is very low. 2,4 Thus, designing a prospective trial to evaluate adjuvant treatment in patients with stage IA disease would not be feasible. The current study evaluated adjuvant treatment in patients with T1 PDAC using the NCDB. Patients were stratified by tumor size into those with subcentimeter tumors (T1a/T1b) and those with tumors measuring 1 to 2 cm (T1c). Patients with lymph node disease and those who received neoadjuvant treatment were excluded. The limitations of the current study include the retrospective nature of the analysis, missing data, and miscoding of the disease and treatments. Data analyzed did not include the details of therapy such as the type of radiotherapy, the total dose, the timing and location of the radiotherapy delivered, the dose and type of chemotherapy, the number of cycles, and the toxicities and morbidities of the treatments used. There also is potential selection bias. Some of the patients likely were not prescribed adjuvant treatment due to comorbidities and/or postoperative complications. The study excluded patients who were diagnosed after 2013 due to a lack of survival 63

8 TABLE 4. Univariate Analysis for OS of Patients With Stage IA, Resected Pancreatic Cancer OS (Months From Diagnosis) Covariate Level No. HR (95% CI) HR P a Log-rank P a Treatment group Adjuvant treatment ( ) <.001 <.001 No adjuvant treatment 466 Tumor size, cm ( ) <.001 < ( ) Race Other/unknown ( ) Black ( ).328 White 730 Sex Female ( ) Male 411 Charlson-Deyo Score ( ) Grade Well differentiated ( ).068 <.001 Moderately differentiated ( ).717 Poorly differentiated/ ( ).020 undifferentiated Cell type not determined 104 Lymphovascular invasion Present ( ) Not present 325 Surgical margins Positive ( ) <.001 <.001 Negative 815 Primary tumor site in pancreas Head ( ) <.001 <.001 Body ( ).874 Tail 133 Facility type Community cancer program/ ( ) comprehensive community cancer program/integrated network cancer program Academic/research program 503 Insurance status Not insured/unknown ( ) Private ( ) <.001 Medicaid/Medicare/other government 506 Age at diagnosis ( ) <.005 Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio; OS, overall survival. a Bold type indicates statistical significance. data and the need for 5-year follow-up. Furthermore, a total of 1589 patients with stage IA disease were identified in the database; 55% were included in the analysis, excluding those patients with missing survival data, a history of neoadjuvant treatment (part of the exclusion criteria), and an unknown treatment sequence. The combination of gemcitabine and capecitabine was less likely to be used given the time frame of this analysis, which might have affected the outcomes. 10 This may affect the extent to which comparisons can be adjusted. In the CONKO-001 trial, serious treatment-related adverse effects of adjuvant therapy were reported in approximately 14% of patients. 2 Of the 359 patients in the ESPAC-4 trial who received adjuvant combination therapy with gemcitabine and capecitabine, 63% had grade 3 and grade 4 toxicities, which included neutropenia, leukopenia, hand-foot syndrome, fatigue, diarrhea, and anemia, among others. 10 The cost for 1 month of the combination therapy of gemcitabine and capecitabine is approximately US$25, The results of the 64

9 Adjuvant Therapy for Subcentimeter PDAC/Shaib et al TABLE 5. Multivariate Analysis for OS of Patients With Resected, Subcentimeter Pancreatic Cancer a OS (Months From Diagnosis) Covariate Level HR (95% CI) HR P Treatment group Adjuvant treatment 0.84 ( ) b.527 b No adjuvant treatment Race Other/unknown 1.14 ( ).749 Black 1.17 ( ).754 White Sex Female 1.12 ( ).624 Male Charlson-Deyo Score ( ) Grade Well differentiated 0.92 ( ).821 Moderately differentiated 1.00 ( ).995 Poorly differentiated/undifferentiated 1.75 ( ).123 Cell type not determined Surgical margins Positive 1.53 ( ).401 Negative Primary tumor site in pancreas Head 1.37 ( ).289 Body 0.81 ( ).615 Tail Age at diagnosis 1.01 ( ).374 Facility type Community cancer program/comprehensive community cancer program/integrated network cancer program Academic/research program 1.35 ( ).210 Insurance status Not insured/unknown 0.99 ( ).982 Private 0.77 ( ).354 Medicaid/Medicare/other government Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio; OS, overall survival. a The number of observations in the original data set was 263. The number of observations used was 252. b Backward selection with an α level of removal of.1 was used. No variables were removed from the model. current study demonstrated that adjuvant therapy does not confer an OS benefit for patients with resected, stage IA, subcentimeter T1a/T1b PDAC. Thus, repercussions of the cost of therapy and the side effects of treatment may outweigh the benefits in this group of patients with PDAC. The role of radiotherapy in the adjuvant setting for patients with PDAC remains controversial. 16 The ESPAC-1 trial compared patients with resected PDAC who received adjuvant concurrent chemotherapy and radiotherapy compared with chemotherapy alone, chemotherapy followed by chemoradiation, and observation only. 11,16 The study demonstrated an OS benefit for adjuvant chemotherapy compared with concurrent chemoradiation (20.1 months vs 15.9 months; P=.009), and demonstrated that concurrent chemoradiation has a deleterious effect on survival, with shorter rates of recurrence-free survival (10.7 months vs 15.2 months; P=.04) and higher rates of adverse events (6% vs 4%) when compared with adjuvant chemotherapy. 11,16 In the current study, among patients with resected, subcentimeter (T1a/ T1b) stage IA PDAC, radiotherapy appeared to have a nonsignificant trend toward shorter survival compared with patients who were treated with no radiotherapy. IPMNs are precancerous lesions with a risk of developing into invasive carcinoma that approaches 30%. To the best of our knowledge, there are no radiographic or clinical characteristics that reliably predict malignancy in patients with IPMNs. 23 It was expected that the majority of patients with subcentimeter PDACs would have a preoperative diagnosis of IPMN. In this analysis, only 9% of all patients with stage I PDAC had a coexisting diagnosis of IPMN; 16% of patients with stage IA PDAC disease had IPMN. Contrary to the expectations, this number is relatively low because subcentimeter PDACs are expected to be incidental after surgical resection for IPMN. 65

10 The current study is the first to evaluate adjuvant therapy in a small subgroup of patients with stage IA, resected PDAC, which is underrepresented in prospective adjuvant trials. Chemotherapy and radiotherapy both have significant cost and adverse effects. Based on this analysis, adjuvant therapy for all patients with stage I, resected pancreatic cancer demonstrated benefit; patients with subcentimeter (T1a/T1b) tumors are unlikely to benefit from adjuvant therapy. This likely could be related to the number of patients involved in the current analysis and the limitations of using retrospective data. Evaluation of the effects of adjuvant therapy in this group of patients using prospective randomized trials was not feasible given the very low incidence of this stage of disease at the time of diagnosis. 2,4 Based on these results, a review of the standard of care for patients with resected PDAC may be warranted, wherein adjuvant therapy is not recommended for patients with resected T1a/T1b PDAC with negative surgical resection margins. FUNDING SUPPORT Supported in part by the Biostatistics and Bioinformatics Shared Resource of Winship Cancer Institute of Emory University and the National Institutes of Health/National Cancer Institute under award number P30CA 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 Walid L. Shaib has received honoraria for acting as a consultant and member of the advisory board for Ipsen Pharmaceuticals, and has received clinical funding for trials for which he was a principal investigator from Covance, ArQule Inc, Eli Lilly and Company, Boston Biomedical Pharma Inc, Merck Sharp and Dohme Corporation, and the Hoosier Cancer Research Network Inc for work performed outside of the current study. Bassel F. El-Rayes has received clinical trial funding grants from Aveo Pharmaceuticals, Boston Biomedical Pharma Inc, Bristol-Myers Squibb, Genentech, the Hoosier Cancer Research Network Inc, Merck, Novartis, and Taiho Oncology for work performed outside of the current study. AUTHOR CONTRIBUTIONS Walid L. Shaib: Planned, wrote, revised, and supervised the writing and concept of the article. Amit Surya Narayan: Wrote and edited the article. Jeffrey M. Switchenko: Analysis of data and statistics. Sujata Kane: Reviewed and edited the article. Christina Wu: Reviewed and edited the article. Mehmet Akce: Reviewed and edited the article. Olatunji B. Alese: Reviewed and edited the article. Pretesh Patel: Reviewed and edited the article. Shishir K Maithel: Reviewed and edited the article. Juan M. Sarmiento: Reviewed and edited the article. David Kooby: Reviewed and edited the article with comments. Bassel F. El-Rayes: Edited the article. REFERENCES 1. Chakraborty J, Langdon-Embry L, Cunanan KM, et al. Preliminary study of tumor heterogeneity in imaging predicts two year survival in pancreatic cancer patients. PLoS One. 2017;12:e Oettle H, Neuhaus P, Hochhaus A, et al. Adjuvant chemotherapy with gemcitabine and long-term outcomes among patients with resected pancreatic cancer: the CONKO-001 randomized trial. JAMA. 2013;310: Kleeff J, Korc M, Apte M, et al. Pancreatic cancer. Nat Rev Dis Primers. 2016;2: Gong J, Tuli R, Shinde A, Hendifar AE. Meta-analyses of treatment standards for pancreatic cancer. Mol Clin Oncol. 2016;4: Abrams RA, Lowy AM, O Reilly EM, Wolff RA, Picozzi VJ, Pisters PW. Combined modality treatment of resectable and borderline resectable pancreas cancer: expert consensus statement. Ann Surg Oncol. 2009;16: Castellanos JA, Merchant NB. Intensity of follow-up after pancreatic cancer resection. Ann Surg Oncol. 2014;21: Bertero L, Massa F, Metovic J, et al. Eighth Edition of the UICC Classification of Malignant Tumours: an overview of the changes in the pathological TNM classification criteria what has changed and why? Virchows Arch. 2018;472: Kamarajah SK, Burns WR, Frankel TL, Cho CS, Nathan H. Validation of the American Joint Commission on Cancer (AJCC) 8th Edition Staging System for Patients with Pancreatic Adenocarcinoma: a Surveillance, Epidemiology and End Results (SEER) analysis. Ann Surg Oncol. 2017;24: Wolfgang CL, Herman JM, Laheru DA, et al. Recent progress in pancreatic cancer. CA Cancer J Clin. 2013;63: Neoptolemos JP, Palmer DH, Ghaneh P, et al; European Study Group for Pancreatic Cancer. Comparison of adjuvant gemcitabine and capecitabine with gemcitabine monotherapy in patients with resected pancreatic cancer (ESPAC- 4): a multicentre, open-label, randomised, phase 3 trial. Lancet. 2017;389: Neoptolemos JP, Stocken DD, Friess H, et al. European Study Group for Pancreatic Cancer. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med. 2004;350: Neoptolemos JP, Moore MJ, Cox TF, et al. European Study Group for Pancreatic Cancer. Effect of adjuvant chemotherapy with fluorouracil plus folinic acid or gemcitabine vs observation on survival in patients with resected periampullary adenocarcinoma: the ESPAC-3 periampullary cancer randomized trial. JAMA. 2012;308: Neoptolemos JP, Stocken DD, Bassi C, et al. European Study Group for Pancreatic Cancer. Adjuvant chemotherapy with fluorouracil plus folinic acid vs gemcitabine following pancreatic cancer resection: a randomized controlled trial. JAMA. 2010;304: American Cancer Society. American Joint Committee on Cancer (AJCC) TNM staging system S, cancer/pancreaticcancer/detailedguide/pancreatic-cancer-staging. Accessed April 5, Tam VC, Ko YJ, Mittmann N, et al. Cost-effectiveness of systemic therapies for metastatic pancreatic cancer. Curr Oncol. 2013;20:e90-e Ahn DH, Williams TM, Goldstein DA, El-Rayes B, Bekaii-Saab T. Adjuvant therapy for pancreas cancer in an era of value based cancer care. Cancer Treat Rev. 2016;42: Regine WF, Winter KA, Abrams R, et al. Fluorouracil-based chemoradiation with either gemcitabine or fluorouracil chemotherapy after resection of pancreatic adenocarcinoma: 5-year analysis of the U.S. Intergroup/ RTOG 9704 phase III trial. Ann Surg Oncol. 2011;18: Gastrointestinal Tumour Study Group. Further evidence of effective adjuvant combined radiation and chemotherapy following curative resection of pancreatic cancer. Cancer. 1987;59: Klinkenbijl JH, Jeekel J, Sahmoud T, et al. Adjuvant radiotherapy and 5-fluorouracil after curative resection of cancer of the pancreas and periampullary region: phase III trial of the EORTC Gastrointestinal Tract Cancer Cooperative Group. Ann Surg. 1999;230: ;discussion Sinn M, Bahra M, Liersch T, et al. CONKO-005: Adjuvant Chemotherapy With Gemcitabine Plus Erlotinib Versus Gemcitabine Alone in Patients After R0 Resection of Pancreatic Cancer: a multicenter randomized phase III trial. J Clin Oncol. 2017;35:

11 Adjuvant Therapy for Subcentimeter PDAC/Shaib et al 21. Ueno H, Kosuge T, Matsuyama Y, et al. A randomised phase III trial comparing gemcitabine with surgery-only in patients with resected pancreatic cancer: Japanese Study Group of Adjuvant Therapy for Pancreatic Cancer. Br J Cancer. 2009;101: Uesaka K, Boku N, Fukutomi A, et al.; JASPAC 01 Study Group. Adjuvant chemotherapy of S-1 versus gemcitabine for resected pancreatic cancer: a phase 3, open-label, randomised, non-inferiority trial (JASPAC 01). Lancet. 2016;388: Tanaka M, Chari S, Adsay V, et al.; International Association of Pancreatology. International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas. Pancreatology. 2006;6: