April 12, 2019
3 min read

Three chemotherapy-related factors linked to longer pancreatic cancer survival

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Photo of Mark Truty
Mark Truty

Three neoadjuvant chemotherapy-related factors appeared associated with improved survival among patients with advanced pancreatic cancer who are candidates for tumor resection, according to study results published in Annals of Surgery.

Researchers found that patients who had more cycles of chemotherapy before surgery, normal carbohydrate antigen 19-9 levels after chemotherapy, and whose tumors showed a major pathologic response to chemotherapy had significantly longer RFS and OS than those who did not.

“We now have more advanced surgical techniques and more effective chemotherapy and radiation therapy. We can take all of these advances and put them together to get the outcomes we are looking for,” Mark Truty, MD, surgical oncologist in the department of surgery at Mayo Clinic, said in a press release.

Researchers from Mayo Clinic conducted a retrospective study of the institution’s pancreatic ductal adenocarcinoma surgical database to find factors associated with improved survival after surgery. The study included 194 patients (median age at surgery, 64.1 years; 55% men) who underwent total neoadjuvant therapy before resection for pathologist- and biopsy-confirmed borderline resectable or locally advanced pancreatic ductal adenocarcinoma between 2010 and 2017.

Researchers excluded patients who received intraoperative or adjuvant radiotherapy.

Truty and colleagues evaluated patients’ baseline carbohydrate antigen 19-9 (CA19-9) tumor marker levels and compared them with levels after receiving chemotherapy but before subsequent preoperative chemoradiation followed by resection.

Researchers defined optimal CA19-9 response as having normal CA19-9 levels (< 37 U/mL) after chemotherapy regardless of the prechemotherapy level.

Study participants received preoperative chemotherapy with FOLFIRINOX (leucovorin calcium, fluorouracil, irinotecan hydrochloride and oxaliplatin) or gemcitabine and nab-paclitaxel. Chemotherapy was followed by 5 weeks of photon/proton external beam chemoradiotherapy, including concurrent radiosensitizing chemotherapy with oral capecitabine (69%), 5-FU infusion (23%) or IV gemcitabine (8%).

Patients underwent surgery within a median of 5 weeks of final chemoradiotherapy treatment.

Sixty-three percent of tumors were considered borderline resectable and 37% were locally advanced. Patients who underwent resection had tumors involving venous (71%), arterial (70%) or both structures (40%) at diagnosis. Most patients (81%) had elevated CA19-9 levels ( 37 U/mL).

Among all patients, median RFS, excluding operative mortalities, was 23.5 months and median OS was 58.8 months.

Results showed patients who received six or more cycles of chemotherapy had significantly longer RFS (27.3 months vs. 10.3 months, P < .001) and OS (60.1 months vs. 23.9 months, P < .001) compared with those who received fewer than six chemotherapy cycles.


Patients who had an optimal postchemotherapy CA19-9 response also had significantly longer RFS (29.3 months vs. 10.5 months, P < .001) and OS (60.5 months vs. 30.2 months, P < .001), as did those whose tumors had a major pathologic response to preoperative chemotherapy (RFS, not reached vs. 12.1 months, P < .001; OS, 72.1 months vs. 34.5 months, P < .001).

“Of the three factors associated with improved outcomes, pathologic treatment response was statistically the most predictive for RFS and OS,” the researchers noted.

The investigators also evaluated the combination of these prognostic factors and found that the 29% of patients who achieved all three had superior median RFS (> 69 months vs. 6.9 months), as well as RFS rates at 1 year (97% vs. 25%), 2 years (90% vs. 13%), and 3 years (72% vs. 0%) compared with the 10% who achieved none of these factors.

The same held true for OS (median, > 77 months vs. 18.5 months; 1 year, 100% vs. 84%; 2 year, 100% vs. 25%; 3 year, 95% vs. 8%).

“Although all three of the factors we identified were independently predictive of survival, they were also related to one another in that they were all chemotherapy-related, more specifically, related to ‘effective’ chemotherapy,” Truty and colleagues wrote.

“Simply extending the duration of ineffective chemotherapy will likely not lead to any significant survival benefit, thus the need to have objective measures of chemotherapeutic response, that is, CA19-9 or PET, and maximizing them before resection,” the investigators wrote. “Until we can identify more predictive molecular markers of response, we are left with these current clinical surrogates.”

This retrospective study was limited by possible selection bias, the investigators noted, and should not be generally applied to all patients who have borderline resectable or locally advanced pancreatic ductal adenocarcinoma.

“If we postulate neoadjuvant therapy ultimately prolongs survival, we must demonstrate chemotherapeutic response preoperatively, otherwise suboptimal outcomes are anticipated,” Truty and colleagues wrote. “Future prospective studies should focus on defining response endpoints; the critical question is not whether we should consider neoadjuvant therapy, but how it is best accomplished.” – by Drew Amorosi

Disclosures: The authors report no relevant financial disclosures.