Treatment Informed by PET Imaging Linked to Better Response Rates in Esophageal Cancer

Karyn A. Goodman, MD

Karyn A. Goodman, MD

Switching neoadjuvant chemotherapy regimens based on PET imaging after induction chemotherapy improved pathologic complete response (pCR) rates in patients with esophageal and gastroesophageal junction (GEJ) cancers, according to findings from the phase II CALGB 80803 study presented on a presscast held ahead of the 2017 Gastrointestinal Cancers Symposium.

Among patients who switched to an alternative chemotherapy regimen based on PET imaging, the pCR rate was 18%. In previous studies, pCR rates were only 5% among patients who continued the same chemotherapy in the neoadjuvant phase despite a PET scan showing their tumors were non-responsive to the induction regimen.

“We have shown that a short course of induction chemotherapy followed by early response assessment using PET imaging to determine whether to switch from ineffective chemotherapy to alternative chemotherapy during preoperative chemoradiation was effective in improving pCR rate,” lead study author Karyn A. Goodman, MD, a radiation oncologist at the University of Colorado School of Medicine, said when presenting the findings on the presscast.

“CALGB 80803 shows the benefit of a new paradigm of using metabolic imaging in esophageal and GEJ cancer to individualize multimodality therapy and improve outcomes in this poor prognosis population,” added Goodman.

The phase II CALGB 80803 study included 257 patients with stage II/III esophageal and GEJ cancers. Patients had a baseline PET scan and were then randomized to standard induction chemotherapy consisting of either modified FOLFOX (mFOLFOX-6; oxaliplatin, leucovorin, 5-FU; n = 129) on days 1, 15, and 29 or carboplatin/paclitaxel on days 1, 8, 22, and 29 (n = 128). A second PET scan was performed during days 36 to 42, following 2 or 3 cycles of chemotherapy.

“If the scan showed that the maximum standard uptake value (SUV-max) had decreased by ≥35% from baseline, we considered the patient a PET responder and they continued on the same chemotherapy during the standard chemoradiation portion [chemotherapy + concurrent radiation (5040 cGy in 180 cGy fx)]. If the SUV-max had decreased by <35% from baseline, the patient was considered a PET non-responder and they crossed over to the alternative chemotherapy offered in the opposing study arm plus concurrent radiation (5040 cGy in 180 cGy fx),” explained Goodman.

Surgery occurred approximately 6 weeks after completion of chemoradiotherapy. The primary endpoint of the trial is to improve pCR of PET non-responders from 5% to 20% by changing from ineffective therapies.

Among 129 patients assigned to mFOLFOX-6, 73 (57%) were PET responders, 39 (30%) were PET non-responders, 13 (10%) were off-study, and the data were missing for 4 (3%). The overall pCR rate for the mFOLFOX-6 cohort was 31% (95% CI, 0.22-0.41). The pCR rate among PET responders was 37.5% (24/64). The pCR rate among the patients who switched to carboplatin/paclitaxel during the chemoradiotherapy phase was 19% (7/37).

Goodman said it was notable that, “the pCR rate in patients who were PET responders and received both induction and concurrent FOLFOX was 38%, which looks quite promising, but this study was not powered to do a head-to-head comparison of FOLFOX and carboplatin/paclitaxel.”

Among 128 patients randomized to carboplatin/paclitaxel, there were 64 (50%) PET responders, 49 (38%) PET non-responders, 13 (10%) were off-study, and the data were missing for 2 (2%). The overall pCR rate for the carboplatin/paclitaxel cohort was 14.4% (95% CI, 0.08-0.23). The pCR among PET responders was 12.5% (7/56). The pCR among patients who switched to mFOLFOX-6 during the chemoradiotherapy phase was 17% (7/41).

Across the entire study population, the pCR rate among all PET non-responders was 18% (14/78; 95% CI, 0.10-0.28). “So the efficacy criteria of changing chemotherapy based on PET response was met for the improvement in pCR,” said Goodman.

The pCR rate among all PET responders across both study arms was 26% (31/120; 95% CI, 0.18-0.35). Among all patients in the entire study, the pCR rate was 22.7% (45/198; 95% CI, 0.17-0.29).

Responding to a question on the potential for improved survival in patients who changed chemotherapy regimens, Goodman said, “We have survival data that is maturing and we’ll be reporting that in the future; however, we do know from prior studies that pCR rates do correlate with survival outcomes…We will be reporting disease-free survival, which was one of the secondary endpoints, probably within the next 6 months to a year.”

Goodman KA, Niedzwiecki D, Hall N, et al. Initial results of CALGB 80803 (Alliance): a randomized phase II trial of PET scan-directed combined modality therapy for esophageal cancer. Presented at: 2017 Gastrointestinal Cancers Symposium; January 19-21, 2017; San Francisco, California; Abstract 1.