Abstract
Objectives:
Patients with locally advanced pancreatic adenocarcinoma (PDAC) receive induction chemotherapy +/− radiation, with the goal of R0 resection and improving survival. Herein, we evaluate outcomes of patients who presented with stage III PDAC and received induction FOLFIRINOX.
Methods:
An institutional database was queried for consecutive patients who received induction FOLFIRINOX for locally unresectable PDAC between 2010–2016. Clinical and radiographic parameters were assessed pre- and post-treatment, and clinical outcomes were evaluated.
Results:
There were 200 patients who met inclusion criteria. The median number of cycles of FOLFIRINOX was 8, 70% (n=140) received radiation, and 18% (n=36) underwent resection. Median OS in resected patients was 36 months (95%CI:24–56), and this group had improved OS compared to patients that did not undergo resection (HR(95%CI); 0.41(0.26–0.64), p<.001). Patients (n=112) who did not progress on induction therapy but remained unresectable had a median OS of 23.9 months (95%CI:21.1–25.4).
Conclusion:
Nearly 20% of patients with locally advanced PDAC responded sufficiently to induction FOLFIRINOX to undergo resection, which was associated with improved OS compared to patients that did not undergo resection. Patients with stable disease who remain unresectable represent a group of patients with locally advanced PDAC who may benefit from optimization of additional nonoperative treatment.
Keywords: locally advanced pancreatic cancer, FOLFIRINOX, induction chemotherapy
Introduction
Approximately 15–20% of patients with pancreatic ductal adenocarcinoma (PDAC) present with resectable disease, while an additional 20–30% have borderline resectable or locally advanced tumors at diagnosis.[1, 2] Patients who present with locally advanced PDAC receive induction chemotherapy with or without radiation, with the goal of converting the tumor to resectable and improving overall survival (OS). Modified FOLFIRINOX (5-fluorouracil, leucovorin, irinotecan, and oxaliplatin) has emerged as the preferred induction chemotherapy regimen for patients with locally advanced or borderline resectable PDAC with good performance status. The use of mFOLFIRINOX in this setting was initially based on data demonstrating improved outcomes in patients with metastatic disease.[3] Subsequently, mFOLFIRINOX has demonstrated improved survival as compared to gemcitabine in the adjuvant setting.[4] Additionally, a number of retrospective studies have shown that FOLFIRINOX in the neoadjuvant setting is associated with improved resection rates and progression free survival as compared to gemcitabine-based regimens.[5–7]
Previously reported retrospective studies examining the role of induction FOLFIRINOX in patients who present with locally unresectable disease show that approximately 30% of patients achieve resection.[8–13] Furthermore, in borderline and locally advanced tumors, resection has been associated with both improved progression-free and overall survival.[6, 10, 14–16] A previous study from MSKCC evaluated 101 patients with stage III PDAC who received induction FOLFIRINOX between 2010 and 2013.[8] In this cohort, 31% of patients underwent resection, and median survival was not reached at the time of publication. Of those who remained unresectable, median survival of patients who progressed during induction FOLFIRINOX therapy was 11 months and 26 months for those who did not.
In this study, we evaluated all stage III PDAC patients who presented to MSKCC between 2010 and 2016 and received induction FOLFIRINOX as initial therapy. The primary aims were to analyze response to chemotherapy +/− radiation and resectability rate in this cohort of patients with locally advanced PDAC. Additionally, factors associated with unresectable disease, recurrence rates and survival outcomes were evaluated. Finally, we examined the subset of patients who were not converted to resectable but did not progress on induction therapy.
Methods
Patient Cohort
An institutional database was queried for all treatment-naïve patients who presented to MSKCC with locally advanced PDAC between July 2010 and December 2016 and were treated with induction mFOLFIRINOX. Approval for this study was obtained from the Institutional Review Board (IRB) at MSKCC. The definition of stage III disease used in this study was based on the National Comprehensive Cancer Network (NCCN) definition[17] and was consistent with our previously published data.[8] Locally advanced disease was defined by encasement of the superior mesenteric artery (SMA) of greater than 180 degrees, any celiac axis (CA) abutment or encasement, inferior vena cava (IVC) abutment, unreconstructable superior mesenteric vein/portal vein (SMV/PV) encasement or occlusion, and aortic invasion or encasement. Patients with radiographic N0 or N1 disease were included. Clinical stage was adjudicated on pretreatment cross-sectional imaging, and additionally, all patients were assessed by the attending surgeon and/or medical oncologist and confirmed to have locally unresectable tumors. Furthermore, individual cases were reviewed at a multidisciplinary tumor board to finalize treatment strategy. Exclusion criteria included patients who had metastatic disease at diagnosis, started treatment prior to presentation or had incomplete treatment history. Additional clinical and pathologic data were obtained from the institutional database and review of the electronic medical record (EMR).
mFOLFIRINOX was administered according to institutional protocol, with doses of 5-fluorouracil (5-FU) of 1200 mg/m2/day × 2 days and bolus 5-FU 300mg/m2, leucovorin 300 mg/m2, oxaliplatin 65–85 mg/m2, and irinotecan 150–180mg/m2 administered every two weeks. Dose reductions and/or delays were at the discretion of the treating medical oncologist. For the purpose of this study, dose reductions or delays were defined as a decrease in dose of chemotherapeutic agent, delayed dose, or omission of an agent from at least one cycle of treatment. Per institutional protocol, restaging CT was obtained following induction mFOLFIRINOX and again following radiation, if received. Radiographic response was prospectively evaluated by a radiologist at the time of imaging and reviewed at a multidisciplinary tumor board. All eligible patients by inclusion criteria had T4 tumors at diagnosis, and downstaging was defined as stage T3 or lower on post-treatment imaging. Serum CA19-9 was recorded at the time of diagnosis and following completion of mFOLFIRINOX chemotherapy. Adverse events were reviewed in the EMR, and toxicity from chemotherapy was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 4.0.[18]. Data lock occurred on July 17th, 2019.
Statistical Analysis
Patient characteristics were described overall and by resection status. Continuous variables were expressed as median (IQR) and categorical variables were expressed as frequency (percent). Comparisons between resected and unresected patients, as well as subgroups of resected patients, were made with Fisher’s Exact test or Wilcoxon Rank Sum test, where appropriate. SMA, CA, and SMV/PV involvement pre-treatment were entered into a multivariable logistic model to assess the additive relationship of these factors with resection. Given sample size constraints, a separate multivariable logistic regression model was built using significant (p<0.05) univariable post-treatment factors. Backward selection with exit criteria of 0.05 was applied. For variables that were not independent by definition, such as absolute and percent change in size as well as T-stage and downstaging of T-stage, only percent change in size and downstaging of T-stage were included in the model.
OS was measured from the start of mFOLFIRINOX until death, and patients alive at last follow-up were censored. Cross-sectional imaging was reviewed for the first site of progression (unresectable) or recurrence (resectable). In all patients, progression-free survival (PFS) was estimated from the start of mFOLFIRINOX until radiologic progression or death, while in resected patients, recurrence-free survival (RFS) was estimated from the date of operation to the date of recurrence. Patients alive and disease free at last follow-up were censored. Additionally, in patients who underwent resection, OS was also estimated from the date of operation, and for patients who were unresected and progressed on FOLIFIRINOX, OS was estimated from date of progression. Kaplan Meier plots were used to visualize survival outcomes separately by resection status.
Univariable time-dependent Cox regression was used to assess the relationship between resection status and OS. To help visualize this association, landmark Kaplan Meier plots were drawn starting one year after initiating mFOLFIRINOX therapy, as most resections occurred by one year. Patients who had resections by one year were considered resected, and those that did not were considered unresectable. Only patients who were alive with greater than one year of follow up were included. Two-sided p-values less than 0.05 were considered statistically significant. All analyses were performed with SAS version 9.4 T1M5 (The SAS Institute, Cary, NC).
Results
During the study period, there were 200 patients who presented with stage III PDAC and received induction mFOLFIRINOX. Median age was 63 years (IQR 56–68), and the cohort was 52% female (n=104). Most patients had excellent or good performance status [ECOG 0–1 (97%, n=193)]. The majority of tumors were located in the head of the pancreas (n=108, 54%) and approximately one-third of patients had pretreatment biliary drainage (n=71, Table 1). Treatment details of all patients are presented in Figure 1.
Table 1.
Baseline clinical and radiographic characteristics of all patients with Stage III PDAC who underwent induction FOLFIRINOX, as well as those that did and did not undergo resection. Continuous variables expressed as median (IQR) and categorical variables expressed as number (%).
| All Patients n = 200 |
Resection n = 36 |
No Resection n = 164 |
p-value | |
|---|---|---|---|---|
| Age, years | 63 (56–68) | 62 (57–68) | 63 (56–68) | >0.95 |
| Gender | 0.36 | |||
| Female | 104 (52) | 16 (44) | 88 (54) | |
| Male | 96 (48) | 20 (56) | 76 (46) | |
| Pretreatment ECOG | 0.74 | |||
| 0 | 33 (17) | 6 (17) | 27 (16) | |
| 1 | 160 (80) | 28 (78) | 132 (80) | |
| 2 | 7 (4) | 2 (6) | 5 (3) | |
| Pretreatment Biliary Drainage | 71 (36) | 13 (36) | 58 (35) | 0.473 |
| Tumor Location | 0.91 | |||
| Head | 108 (54) | 21 (58) | 87 (53) | |
| Body | 79 (40) | 13 (36) | 66 (40) | |
| Tail | 13 (7) | 2 (6) | 11 (7) | |
| Tumor size, cm⁺ | 3.7 (3.0–4.6) | 3.4 (3.0–4.0) | 3.8 (3.0–4.8) | 0.13 |
| Clinical N-stage | 0.85 | |||
| N0 | 116 (58) | 20 (56) | 96 (59) | |
| N1 | 84 (42) | 16 (44) | 68 (41) | |
| SMA | 0.014 | |||
| No Involvement | 92 (46) | 20 (56) | 72 (44) | |
| Abutment | 20 (10) | 7 (19) | 13 (8) | |
| Encasement | 88 (44) | 9 (25) | 79 (48) | |
| Celiac Axis | 0.31 | |||
| No Involvement | 101 (51) | 21 (58) | 80 (49) | |
| Abutment | 27 (14) | 6 (17) | 21 (13) | |
| Encasement | 72 (36) | 9 (25) | 63 (38) | |
| HA | 0.17 | |||
| No Involvement | 128 (64) | 26 (72) | 102 (62) | |
| Abutment | 15 (8) | 4 (11) | 11 (7) | |
| Encasement | 57 (29) | 6 (17) | 51 (31) | |
| SMV/PV | 0.010 | |||
| No Involvement | 47 (24) | 7 (19) | 40 (24) | |
| Abutment | 29 (15) | 12 (33) | 17 (10) | |
| Encasement/Occlusion | 119 (60) | 17 (47) | 102 (62) | |
| Thrombus | 5 (3) | 0 (0) | 5 (3) | |
| Basis for Unresectability | ||||
| Arterial | 148 (74) | 21 (58) | 127 (77) | 0.022 |
| Venous | 153 (77) | 29 (81) | 124 (76) | 0.67 |
| Pretreatment CA19-9 (Units/mL) | 206 (54–621) | 180 (43–607) | 222 (57–621) | 0.55 |
| Pretreatment CEA (ng/mL) | 4.0 (2.4–7.3) | 3.0 (1.7–6.7) | 4.3 (2.5–7.8) | 0.037 |
one patient with no distinct tumor on initial scan
Figure 1.
Flowchart of patients included in study.
When comparing patients who underwent resection (n=36, 18%) and those who did not (n=164, 82%), the two cohorts did not differ with respect to age, pretreatment performance status, initial tumor size, or clinical nodal stage (Table 1). Both resected and unresected groups received a median of 8 cycles of mFOLFIRINOX. Overall, 59% of patients had dose reductions or delays, and 21% of patients had grade 3 or 4 toxicity, which did not differ significantly between groups (p=0.14). There were 140 patients (70%) who received radiation, and of these, 84% received concurrent fluoropyrimidine therapy with radiation (n=118). There was no significant difference in pretreatment serum CA19-9 between groups (p=0.55), however, for the patients with available data (n=161), unresectable patients had significantly higher CEA (median, 4.3 vs 3.0, p=0.037). A decrease in CA19-9 following completion of FOLFIRINOX was greater in the resected group as compared to the unresected group, although this difference was not statistically significant (65% vs 51%, p=0.07).
Comparing pretreatment imaging between the two groups, arterial involvement was associated with unresectability (77%, n=127 vs 58%, n=21, p=0.022), and specifically, encasement of the SMA (48%, n=79 vs 25%, n=9, p=0.014). PV/SMV involvement was also associated with unresectable disease (encasement: 62%, n=102 vs 47%, n=17; p=0.010). There was no significant difference in the two groups with respect to celiac axis or hepatic artery involvement (p=0.17–0.31). Additionally, patients with multivessel involvement were less likely to undergo resection (51%, n=83 vs 19%, n=7, p<0.001). Multivariable analysis demonstrated that SMA encasement was associated with unresectability and SMV/PV involvement was independently associated with resectability (Table 3). Following completion of the protocol (mFOLFIRINOX +/− (chemo)radiation), patients who underwent resection had significantly smaller tumors (2.5 cm vs. 3.4 cm, p<0.001) and lower clinical T-stage (T4, 25%, n=9 vs. 90%, n=148, p<0.001), however there was no significant difference in clinical nodal status (17% vs 24%, p=0.39) (Table 2). Multivariable selection-based analysis of post-protocol factors demonstrated that decreased T-stage was associated with resectable disease and SMA encasement was associated with unresectable disease (Table 3).
Table 3.
Multivariable analysis of pre- and post-treatment factors associated with resection.
| Pre-Treatment | OR (95% CI) | p-value |
|---|---|---|
| SMV/PV | ||
| No Involvement | ref | 0.004 |
| Abutment | 3.72 (1.16–11.94) | |
| Encasement/Occlusion | 0.53 (0.17–1.63) | |
| Thrombus | NE | |
| SMA | ||
| No Involvement | ref | 0.004 |
| Abutment | 2.74 (0.90–8.32) | |
| Encasement | 0.32 (0.12–0.83) | |
| Celiac Axis | ||
| No Involvement | ref | 0.12 |
| Abutment | 0.53 (0.16–1.78) | |
| Encasement | 0.33 (0.12–0.95) | |
| Post-Treatment | OR (95% CI) | p-value |
| T-Stage Downstaged | ||
| No | ref | <.001 |
| Yes | 25.74 (6.26–105.78) | |
| SMA | ||
| No Involvement | ref | 0.022 |
| Abutment | 0.22 (0.03–1.46) | |
| Encasement | 0.14 (0.03–0.62) |
OR>1, odds of undergoing resection are higher; OR<1, odds of undergoing resection are lower; NE: not able to estimate
Table 2.
Post-treatment clinical and radiographic characteristics of all patients following induction FOLFIRINOX +/− radiation. Continuous variables expressed as median (IQR) and categorical variables expressed as number (%).
| All Patients n = 200 |
Resection n = 36 |
No Resection n = 164 |
p-value | |
|---|---|---|---|---|
| FOLFIRINOX cycles | 8 (6–9) | 8 (5–8) | 8 (6–10) | 0.44 |
| FOLFIRINOX duration, weeks | 14.3 (10.0–20.1) | 14.1 (11.7–17.3) | 14.5 (10.0–20.7) | 0.84 |
| Dose reduction/delayed dose | 118 (59) | 16 (44) | 102 (62) | 0.06 |
| Toxicity Grade | 0.14 | |||
| None | 26 (13) | 5 (14) | 21 (13) | |
| Grade 1 or 2 | 133 (67) | 28 (78) | 105 (64) | |
| Grade 3 | 37 (19) | 3 (8) | 34 (21) | |
| Grade 4 | 4 (2) | 0 (0) | 4 (2) | |
| Radiation | 140 (70) | 23 (64) | 117 (71) | 0.42 |
| No chemosensitizer | 22 (16) | 3 (13) | 19 (16) | |
| 5FU-based | 76 (54) | 12 (52) | 64 (55) | |
| Gemcitabine-based | 42 (30) | 8 (35) | 34 (29) | |
| Post-FOLFIRINOX | ||||
| Tumor size, cm** | 3.3 (2.5–4.1) | 2.6 (2.2–3.3) | 3.4 (2.7–4.2) | 0.002 |
| Size reduction, % | −12.8 (−26.1–0.0) | −18.1 (−32.8–3.5) | −11.1 (−24.3–0.0) | 0.046 |
| T-stage | <.001 | |||
| ycT2 | 2 (1) | 2 (6) | 0 (0) | <.001 |
| ycT3 | 20 (10) | 12 (33) | 8 (5) | |
| ycT4 | 178 (89) | 22 (61) | 156 (95) | |
| T-stage downstaged | 22 (11) | 14 (39) | 8 (5) | <.001 |
| N-stage | 0.57 | |||
| N0 | 124 (62) | 24 (67) | 100 (61) | 0.57 |
| N1 | 76 (38) | 12 (33) | 64 (39) | |
| N-stage downstaged | 17 (9) | 4 (11) | 13 (8) | 0.52 |
| Post-FOLFIRINOX CA19-9 | 78 (36–215) | 52 (28–142) | 85 (39–235) | 0.10 |
| Decrease CA19-9, % | −54.3 (−80.4–−5.6) | −64.7 (−83.9–25.5) | −51.2 (−80.3–0.0) | 0.07 |
| Post-FOLFIRINOX CEA | 4 (3–7) | 3 (2–5) | 5 (3–8) | 0.008 |
| Decrease CEA, % | 0.0 (−32.1–50.0) | −7.5 (−32.9–38.1) | 4.3 (−31.4–50.0) | |
| Post-Protocol* | ||||
| Tumor size, cm** | 3.1 (2.4–4.1) | 2.5 (1.9–3.0) | 3.4 (2.4–4.3) | <.001 |
| Size reduction, % | −17.1 (−32.6–1.5) | −29.3 (−42.8–13.8) | −15.2 (−29.8–0.0) | 0.003 |
| T-stage | ||||
| ycT2 | 2 (1) | 2 (6) | 0 (0) | <.001 |
| ycT3 | 33 (17) | 23 (64) | 10 (6) | |
| ycT4 | 157 (79) | 9 (25) | 148 (90) | |
| T-stage downstaged | 34 (17) | 25 (69) | 9 (5) | <.001 |
| SMA | <.001 | |||
| No Involvement | 73 (37) | 24 (67) | 49 (30) | |
| Abutment | 21 (11) | 6 (17) | 15 (9) | |
| Encasement | 98 (49) | 4 (11) | 94 (57) | |
| Celiac Axis | <.001 | |||
| No Involvement | 94 (47) | 23 (64) | 71 (43) | |
| Abutment | 19 (10) | 8 (22) | 11 (7) | |
| Encasement | 79 (40) | 3 (8) | 76 (46) | |
| HA | 0.032 | |||
| No Involvement | 126 (63) | 26 (72) | 100 (61) | |
| Abutment | 8 (4) | 3 (8) | 5 (3) | |
| Encasement | 58 (29) | 5 (14) | 53 (32) | |
| SMV/PV | <.001 | |||
| No Involvement | 48 (24) | 15 (42) | 33 (20) | |
| Abutment | 41 (21) | 14 (39) | 27 (16) | |
| Encasement/Occlusion | 101 (51) | 5 (14) | 96 (59) | |
| Thrombus | 2 (1) | 0 (0) | 2 (1) | |
| Basis for Unresectability | ||||
| Arterial | 140 (70) | 14 (39) | 126 (77) | <.001 |
| Venous | 103 (52) | 5 (14) | 98 (60) | <.001 |
| N-stage | ||||
| N0 | 146 (73) | 28 (78) | 118 (72) | 0.39 |
| N1 | 46 (23) | 6 (17) | 40 (24) | |
| N-stage downstaged | 19 (10) | 6 (17) | 13 (8) | 0.11 |
8 patients without post-protocol scan; 2 patients resected, 6 patients with clinical deterioration precluding imaging
tumor size unavailable in 7 patients post-FOLFIRINOX and 5 post-protocol
150 patients with pre-treatment elevated CA19-9 included; 48 patients with pre-treatment elevated CEA included
Fifty-two patients (26%) underwent exploration, and 36 patients (18%) underwent resection (Figure 1). Vascular resection/reconstruction was performed in 39% (n=14) of patients; 6 of these patients had a portal venous resection, 7 had resection of the celiac axis/hepatic artery, and one patient had both a celiac axis resection and a portal vein resection. In patients who underwent resection, 23 individuals received radiation while 13 did not. There were no significant differences in pathologic variables when comparing patients who received radiation prior to resection and those who did not, although the rate of R0 resection was lower in those that received radiation (48%, n=11 vs. 85%, n=11, p=0.07, Supplemental Table 1). Furthermore, patients who received radiation were significantly more likely to have a vascular resection (52%, n=12 vs 15%, n=2, p=0.039). The overall R0 resection rate was 61% (n=22) (Table 4).
Table 4.
Pre-treatment and post-treatment factors associated with R0/R1 resection. Continuous variables expressed as median (IQR) and categorical variables expressed as number.
| All Patients n=35ǂ |
R0 Resection n=22 |
R1 Resection n=13 |
|
|---|---|---|---|
| Pre-treatment | |||
| Tumor size, cm | 3.5 (3.0–4.1) | 3.6 (3.1–4.5) | 3.0 (3.0–3.5) |
| SMA | |||
| No Involvement | 20 (57) | 14 (64) | 6 (46) |
| Abutment | 7 (20) | 4 (18) | 3 (23) |
| Encasement | 8 (23) | 4 (18) | 4 (31) |
| Celiac Axis | |||
| No Involvement | 21 (60) | 14 (64) | 7 (54) |
| Abutment | 5 (14) | 3 (14) | 2 (15) |
| Encasement | 9 (26) | 5 (22) | 4 (31) |
| HA | |||
| No Involvement | 25 (71) | 17 (77) | 8 (62) |
| Abutment | 4 (12) | 3 (14) | 1 (7) |
| Encasement | 6 (17) | 2 (10) | 4 (31) |
| SMV/PV | |||
| No Involvement | 6 (17) | 3 (14) | 3 (23) |
| Abutment | 12 (34) | 7 (32) | 5 (38) |
| Encasement/Occlusion | 17 (49) | 12 (54) | 5 (38) |
| Pre-treatment CA19-9, Units/mL | 168 (42–585) | 104 (30–585) | 295 (76–577) |
| Post-treatment | |||
| Tumor size, cm | 2.5 (2.0–3.0) | 2.6 (2.0–3.0) | 2.5 (2.0–2.8) |
| SMA | |||
| No Involvement | 24 (73) | 16 (80) | 8 (62) |
| Abutment | 5 (15) | 2 (10) | 3 (23) |
| Encasement | 4 (12) | 2 (10) | 2 (15) |
| Celiac Axis | |||
| No Involvement | 23 (70) | 16 (80) | 7 (54) |
| Abutment | 7 (21) | 2 (10) | 5 (38) |
| Encasement | 3 (9) | 2 (10) | 1 (8) |
| HA | |||
| No Involvement | 25 (76) | 18 (90) | 7 (54) |
| Abutment | 3 (9) | 1 (5) | 2 (15) |
| Encasement | 5 (15) | 1 (5) | 4 (32) |
| SMV/PV | |||
| No Involvement | 14 (42.5) | 8 (40) | 6 (46) |
| Abutment | 14 (42.5) | 9 (45) | 5 (38) |
| Encasement/Occlusion | 5 (15) | 3 (15) | 2 (15) |
| Post-FOLFIRINOX CA19-9, Units/mL** | 50 (28–142) | 45 (25–108) | 138 (28–275) |
one patient excluded as no pathologic data available
two patients with no post-treatment imaging;
5 patients with no post-FOLFIRINOX CA19-9
Median follow-up of survivors was 36 months (range: 2–84), and median OS of all patients was 21.3 months (95%CI: 19.1–23.9) (Figure 2). At 5-years post-mFOLFIRINOX, OS of patients who underwent resection was 27.1% (95%CI: 12.7–43.9%), with a median OS of 35.7 months (95%CI: 24.3–56.4, Figure 3a), while 5-year OS in the unresected group was 5.2% (95%CI: 1.5–12.5%), with a median OS of 20 months (95%CI: 16.8–21.9, Figure 3b). OS was directly compared between these two groups using a time-dependent covariate, and patients who underwent resection had significantly improved OS as compared to those who remained unresectable (HR (95% CI), 0.41 (0.26–0.64), p<0.001, Figure 3c).
Figure 2.
Overall survival of entire cohort.
Figure 3.
Overall survival and progression free survival in resected patients (a) and unresected patients (b). Landmark curve of overall survival at one-year post-FOLFIRINOX based on resection status (c).
In resected patients, the 5-year PFS was 19.6% (95%CI: 8.2–34.6%), and median PFS was 18 months (95%CI: 14.2–23.7) from the start of mFOLIFIRNOX (Figure 3a). From the time of surgery, the median OS was 27.3 months (95%CI: 17.0–49.8) and median RFS was 7.6 months (95%CI: 5.2–17.1). By the end of follow-up, 23 patients developed a recurrence. Distant recurrence as the site of first recurrence occurred in 8 patients, isolated local recurrence occurred in 11 patients, and 4 patients developed both local and distant recurrences.
In the 164 patients who did not undergo resection, median PFS was 10.1 months (95%CI: 9.2–11.7) and 5-year PFS was only 1.0% (95%CI: 0.1–4.7%, Figure 3b). Initial progression occurred locally in the pancreas in 51 patients, at distant sites in 80 patients, and at both local and distant sites in 15 patients. There were 52 patients who progressed during induction chemotherapy +/− (chemo)radiation, and 42 of these patients progressed at distant sites. Median OS from the time of progression in this group was 8.1 months (95%CI: 7.3–9.6).
There were 112 patients who did not progress during induction therapy yet remained unresectable, and median survival in this group was 23.9 months (95%CI: 21.1–25.4). The majority of these patients demonstrated disease progression during follow up; 43 progressed locally within the pancreas, 42 progressed at distant sites, 9 progressed at both local and distant sites and 13 patients had no progression during follow up. The majority of these patients (n=81, 72%) received additional treatment following completion of induction FOLFIRINOX +/− radiation. There were 31 patients (28%), who continued on maintenance chemotherapy in the absence of progressive disease, and a total of 72 patients (64%) underwent additional systemic treatment after local and/or distant progression.
Discussion
Stage III PDAC is defined by significant vascular involvement precluding surgical resection. Induction mFOLFIRINOX is the mainstay of treatment for patients who present with locally unresectable disease and have good performance status. A subset of these patients will receive radiation following induction chemotherapy for improved local control. Previous data have demonstrated that approximately 25–40% of these patients ultimately undergo resection,[8–11] and that patients who undergo resection have improved survival compared to those that do not.[6, 10, 15, 16] In patients who remain with unresectable disease but do not progress on induction therapy, survival is improved as compared to those patients whose disease progresses.[8]
In this cohort of 200 patients with locally unresectable PDAC who received induction mFOLFIRINOX +/− (chemo)radiation, 26% underwent surgical exploration and 18% were successfully resected. For the resected subset of patients, the median survival from the start of mFOLFIRINOX was 36 months and the median PFS was 18 months. In patients with unresectable tumors, the median OS was 20 months, and PFS was 10 months from the commencement of FOLFIRINOX. Previous data have demonstrated that patients with locally advanced disease who undergo resection following induction chemotherapy have improved survival compared to those with persistently unresectable disease.[6, 10, 14–16] The results of our study further support this observation, and the median OS of 36 months is consistent with other studies.
Of all patients treated with induction mFOLFIRINOX, 26% were explored and 18% resected. This resection rate is lower than previously reported, however there is variability among studies that assess resectability rates. Specific induction treatment regimen, use of (chemo)radiation, surgeon willingness to perform vascular resection, and the definition of borderline and locally advanced tumors can all influence resectability rates and are difficult to control for in this and other retrospective studies. As an institutional practice, we perform arterial resections selectively and very rarely in combination with PV/SMV resections, given the poor results,[19–21] and this is likely a contributing factor to the resection rate in the current study. Additionally, when evaluating feasibility of resection in stage III patients, assessment of post-treatment imaging to determine likelihood of R0 resection is the primary consideration. Previous studies have cited an R0 resection rate of 40–90% for patients who receive mFOLFIRINOX for locally advanced disease.[6, 8–10, 15, 22–24] The R0 resection rate in the current study was 61%, which is in the range of prior data.
A previous study by Ferrone et al. demonstrated that unresectability on post-neoadjuvant imaging does not correlate with successful resection in patients with both borderline and locally advanced tumors, noting that 92% of patients had R0 resection when only one third of patients converted to resectable after treatment as assessed on cross-sectional imaging.[25] Consistent with these data, our series demonstrated that only a subset of patients (69%) who underwent resection had their tumors downstaged on post-treatment imaging. Furthermore, within the group that remained unresectable, there was a 15% decrease in tumor size post-protocol, yet more than 50% of patients had significant SMA, CA or SMV/PV involvement on post-treatment imaging. Although tumors may decrease in size, vascular involvement may remain unchanged and still preclude resection.
This study evaluated the extent of vascular involvement on cross-sectional imaging as a predictor of resection. Our results showed that patients who have SMA encasement on pre-treatment imaging, as well as persistent encasement of the SMA following induction therapy, are significantly less likely to be resected. Extent of venous involvement was also associated with resectability, however, SMV/PV abutment was associated with an increased likelihood of resection as compared to no involvement. This could be explained by the fact that SMV/PV abutment was associated with multivessel involvement, and the true reason for unresectability was not accounted for in this model. Furthermore, vascular resection in patients with SMV/PV abutment could account for successful resection in a subset of patients.
The group of patients who remained with locally advanced and unresectable tumors after induction chemotherapy +/− (chemo)radiation had a median OS of 24 months from the commencement of FOLFIRINOX. This is also comparable to previously published data,[8] and indicates that there is a group of patients with localized, non-metastatic disease who have relatively favorable outcomes without resection. Further studies should focus on this cohort of patients, in terms of evaluating biomarkers to predict if there are a subset of these patients who could derive benefit from additional treatment.
Ablative dose radiation therapy has recently emerged as a promising treatment modality in patients with locally advanced pancreatic cancer. Early data demonstrated improved OS in patients treated with high-doses of radiation (defined as a biologically effective dose >70Gy) in patients with unresectable PDAC.[26, 27] At Memorial Sloan Kettering, an ablative dose radiotherapy strategy has been used routinely for locally advanced PDAC patients since 2016. In our data, comparison of ablative dose radiation therapy with surgical resection in patients with locally advanced and borderline resectable PDAC did not demonstrate a significant difference in overall survival or locoregional progression between the two groups [Jossilant et al, unpublished]. These data suggest that there is a group of patients with locally advanced disease who may benefit from multimodal nonoperative therapy, in an effort to maximize OS and minimize morbidity. However, future prospective trials are warranted to evaluate these treatment options.
There are two main strengths to this study that allow for detailed examination of patients who present with locally advanced disease. First, this study captures the denominator of patients who received induction mFOLFIRINOX at our institution. Many studies have focused primarily on patients who underwent surgical exploration or resection following induction therapy,[16, 22] missing the true number of patients who present with locally advanced disease. The other strength of this of this study is that it includes only patients with truly locally unresectable tumors while excluding those with borderline resectable disease. All patients had a pretreatment scan demonstrating T4 disease, documentation from an attending surgeon or medical oncologist describing a locally unresectable tumor and consensus multi-disciplinary review. While a few prior studies include only stage III patients, many studies encompass both borderline and locally advanced PDAC.
There are several limitations to this study. While all treatment was given according to an institutional protocol, there may have been variation among treating physicians regarding specifics of the mFOLFIRINOX regimen and willingness to attempt resection, which is difficult to capture in a retrospective study. The use of radiation was decided following evaluation of response to induction chemotherapy, and therefore we were unable to control for the differences in the patients who received radiation and those who did not given the retrospective nature of this study. Additionally, the number of cycles of mFOLFIRINOX, as well as the radiation regimen used, was not pre-defined.
Conclusion
In conclusion, we evaluated all patients who presented to MSKCC over a six-year period with locally unresectable PDAC and received induction mFOLFIRINOX. Of this cohort, 18% of patients underwent surgical resection. Survival in the resected group was significantly improved as compared to those who remained unresectable, approaching the survival rates of patients who present with initially resectable tumors. Further trials will need to build upon the current paradigm of mFOLFIRINOX +/− (chemo)radiation in patients with stage III PDAC, and additionally, future studies should aim to optimize multimodality treatment options in patients with stable disease following induction therapy.
Supplementary Material
Synopsis.
This study evaluated 200 patients with locally unresectable pancreatic ductal adenocarcinoma (PDAC) who received induction FOLFIRINOX +/− radiation. In this cohort, 26% of patients underwent surgical exploration, and the overall resection rate was 18%; surgical resection was associated with improved survival.
Footnotes
Conflicts of Interest: No disclosures related to this work
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1.Hidalgo M Pancreatic cancer. N Engl J Med. 2010;362(17):1605–17. [DOI] [PubMed] [Google Scholar]
- 2.Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet. 2004;363(9414):1049–57. [DOI] [PubMed] [Google Scholar]
- 3.Conroy T, Desseigne F, Ychou M, Bouche O, Guimbaud R, Becouarn Y, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817–25. [DOI] [PubMed] [Google Scholar]
- 4.Conroy T, Hammel P, Hebbar M, Ben Abdelghani M, Wei AC, Raoul JL, et al. FOLFIRINOX or Gemcitabine as Adjuvant Therapy for Pancreatic Cancer. N Engl J Med. 2018;379(25):2395–406. [DOI] [PubMed] [Google Scholar]
- 5.Chapman BC, Gleisner A, Rigg D, Messersmith W, Paniccia A, Meguid C, et al. Perioperative and Survival Outcomes Following Neoadjuvant FOLFIRINOX versus Gemcitabine Abraxane in Patients with Pancreatic Adenocarcinoma. JOP. 2018;19(2):75–85. [PMC free article] [PubMed] [Google Scholar]
- 6.Gemenetzis G, Groot VP, Blair AB, Laheru DA, Zheng L, Narang AK, et al. Survival in Locally Advanced Pancreatic Cancer After Neoadjuvant Therapy and Surgical Resection. Ann Surg. 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yoo C, Kang J, Kim KP, Lee JL, Ryoo BY, Chang HM, et al. Efficacy and safety of neoadjuvant FOLFIRINOX for borderline resectable pancreatic adenocarcinoma: improved efficacy compared with gemcitabine-based regimen. Oncotarget. 2017;8(28):46337–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sadot E, Doussot A, O’Reilly EM, Lowery MA, Goodman KA, Do RK, et al. FOLFIRINOX Induction Therapy for Stage 3 Pancreatic Adenocarcinoma. Ann Surg Oncol. 2015;22(11):3512–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Suker M, Beumer BR, Sadot E, Marthey L, Faris JE, Mellon EA, et al. FOLFIRINOX for locally advanced pancreatic cancer: a systematic review and patient-level meta-analysis. Lancet Oncol. 2016;17(6):801–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Chatzizacharias NA, Tsai S, Griffin M, Tolat P, Ritch P, George B, et al. Locally advanced pancreas cancer: Staging and goals of therapy. Surgery. 2018;163(5):1053–62. [DOI] [PubMed] [Google Scholar]
- 11.Nitsche U, Wenzel P, Siveke JT, Braren R, Holzapfel K, Schlitter AM, et al. Resectability After First-Line FOLFIRINOX in Initially Unresectable Locally Advanced Pancreatic Cancer: A Single-Center Experience. Ann Surg Oncol. 2015;22 Suppl 3:S1212–20. [DOI] [PubMed] [Google Scholar]
- 12.Marthey L, Sa-Cunha A, Blanc JF, Gauthier M, Cueff A, Francois E, et al. FOLFIRINOX for locally advanced pancreatic adenocarcinoma: results of an AGEO multicenter prospective observational cohort. Ann Surg Oncol. 2015;22(1):295–301. [DOI] [PubMed] [Google Scholar]
- 13.Faris JE, Blaszkowsky LS, McDermott S, Guimaraes AR, Szymonifka J, Huynh MA, et al. FOLFIRINOX in locally advanced pancreatic cancer: the Massachusetts General Hospital Cancer Center experience. Oncologist. 2013;18(5):543–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rangelova E, Wefer A, Persson S, Valente R, Tanaka K, Orsini N, et al. Surgery Improves Survival After Neoadjuvant Therapy for Borderline and Locally Advanced Pancreatic Cancer: A Single Institution Experience. Ann Surg. 2019. [DOI] [PubMed] [Google Scholar]
- 15.Blazer M, Wu C, Goldberg RM, Phillips G, Schmidt C, Muscarella P, et al. Neoadjuvant modified (m) FOLFIRINOX for locally advanced unresectable (LAPC) and borderline resectable (BRPC) adenocarcinoma of the pancreas. Ann Surg Oncol. 2015;22(4):1153–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Michelakos T, Pergolini I, Castillo CF, Honselmann KC, Cai L, Deshpande V, et al. Predictors of Resectability and Survival in Patients With Borderline and Locally Advanced Pancreatic Cancer who Underwent Neoadjuvant Treatment With FOLFIRINOX. Ann Surg. 2019;269(4):733–40. [DOI] [PubMed] [Google Scholar]
- 17.Lowery MA, Kelsen DP, Stadler ZK, Yu KH, Janjigian YY, Ludwig E, et al. An emerging entity: pancreatic adenocarcinoma associated with a known BRCA mutation: clinical descriptors, treatment implications, and future directions. Oncologist. 2011;16(10):1397–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cheng DT, Mitchell TN, Zehir A, Shah RH, Benayed R, Syed A, et al. Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT): A Hybridization Capture-Based Next-Generation Sequencing Clinical Assay for Solid Tumor Molecular Oncology. J Mol Diagn. 2015;17(3):251–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bockhorn M, Burdelski C, Bogoevski D, Sgourakis G, Yekebas EF, Izbicki JR. Arterial en bloc resection for pancreatic carcinoma. Br J Surg. 2011;98(1):86–92. [DOI] [PubMed] [Google Scholar]
- 20.Mollberg N, Rahbari NN, Koch M, Hartwig W, Hoeger Y, Buchler MW, et al. Arterial resection during pancreatectomy for pancreatic cancer: a systematic review and meta-analysis. Ann Surg. 2011;254(6):882–93. [DOI] [PubMed] [Google Scholar]
- 21.Zettervall SL, Ju T, Holzmacher JL, Huysman B, Werba G, Sidawy A, et al. Arterial, but Not Venous, Reconstruction Increases 30-Day Morbidity and Mortality in Pancreaticoduodenectomy. J Gastrointest Surg. 2019. [DOI] [PubMed] [Google Scholar]
- 22.Hackert T, Sachsenmaier M, Hinz U, Schneider L, Michalski CW, Springfeld C, et al. Locally Advanced Pancreatic Cancer: Neoadjuvant Therapy With Folfirinox Results in Resectability in 60% of the Patients. Ann Surg. 2016;264(3):457–63. [DOI] [PubMed] [Google Scholar]
- 23.Pietrasz D, Marthey L, Wagner M, Blanc JF, Laurent C, Turrini O, et al. Pathologic Major Response After FOLFIRINOX is Prognostic for Patients Secondary Resected for Borderline or Locally Advanced Pancreatic Adenocarcinoma: An AGEO-FRENCH, Prospective, Multicentric Cohort. Ann Surg Oncol. 2015;22 Suppl 3:S1196–205. [DOI] [PubMed] [Google Scholar]
- 24.Rombouts SJ, Walma MS, Vogel JA, van Rijssen LB, Wilmink JW, Mohammad NH, et al. Systematic Review of Resection Rates and Clinical Outcomes After FOLFIRINOX-Based Treatment in Patients with Locally Advanced Pancreatic Cancer. Ann Surg Oncol. 2016;23(13):4352–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ferrone CR, Marchegiani G, Hong TS, Ryan DP, Deshpande V, McDonnell EI, et al. Radiological and surgical implications of neoadjuvant treatment with FOLFIRINOX for locally advanced and borderline resectable pancreatic cancer. Ann Surg. 2015;261(1):12–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Krishnan S, Chadha AS, Suh Y, Chen HC, Rao A, Das P, et al. Focal Radiation Therapy Dose Escalation Improves Overall Survival in Locally Advanced Pancreatic Cancer Patients Receiving Induction Chemotherapy and Consolidative Chemoradiation. Int J Radiat Oncol Biol Phys. 2016;94(4):755–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Rudra S, Jiang N, Rosenberg SA, Olsen JR, Roach MC, Wan L, et al. Using adaptive magnetic resonance image-guided radiation therapy for treatment of inoperable pancreatic cancer. Cancer Med. 2019;8(5):2123–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.