Phase II Trial: Proton beam therapy offers lower toxicities without harming progression-free survival in patients with esophageal cancer versus intensity-modulated radiation therapy

By Darcy Lewis

A Phase IIB randomized trial for proton beam therapy (PBT) vs intensity-modulated radiation therapy (IMRT) in patients with esophageal cancer has found that the dosimetric superiority of PBT translates to clinically significant reductions in severe adverse events following chemoradiation therapy (CRT), according to a single-institution study presented at the 2019 ASTRO Annual Meeting in Chicago.

The authors, led by Steven H. Lin, MD, PhD, a radiation oncologist at MD Anderson Cancer Center in Houston, hypothesized that PBT could reduce the total toxicity burden (TTB) experienced by esophageal cancer patients treated with chemoradiation therapy without affecting clinical outcomes like progression-free survival (PFS). Thus, TTB and PFS were the trial’s co-primary endpoints.

The posterior probability that mean TTB was lower for PBT vs IMRT was 0.9989 with 66% of the data gathered, which exceeded the prespecified stopping boundary at 67% follow-up of 0.9942. IMRT had posterior mean TTB 2.3 times higher [39.9 (95% DI 26.2-54.9) vs. PBT’s 17.4 (95% DI 10.5-25.0)]. The mean postoperative complications severity score for IMRT was 7.6 times higher [19.1 (95% DI 7.3-32.3) than PBT’s 2.5 (95% DI 0.3-5.2)].

With a median follow-up time of 44.1 weeks, PFS was comparable among CRT cohorts. The median PFS in the IMRT cohort was 18.1 months (range: 10 months – not reached), while it was 28.5 months in the PBT cohort (range: 7.1 months – not reached).

The trial’s secondary endpoints were overall survival (OS), patient quality of life assessed by the EQ5D standard instrument, and tumor pathologic response, all of which were found to be comparable between groups.

Lin, who won ASTRO’s second annual Steven A. Leibel Memorial award for outstanding abstract in the Clinical Trials Session for this trial, explained his group’s unusual TTB endpoint. “We defined TTB as a severity-weighted sum over 11 specific severe adverse events (AE) that can occur from start of CRT up to 12 months, including postoperative complications for patients who undergo surgery,” he said. “We assigned each AE grade a weight quantifying its severity on a scale of 0-100.”

Given the increased costs associated with PBT, Lin also explained key facets of the trial’s accrual and early stoppage. In March 2019, the investigators closed the Phase IIB trial prior to activation of the Phase III NRG-GI006 trial (NCT03801876). “At that point, 145 pts were randomized, 72 to IMRT and 73 to PBT, with 105 pts evaluable (61 IMRT and 44 PBT),” he said. “For the 29 unevaluable pts on the PBT arm, 22 (76%) were treated off protocol due to insurance denial. For the 11 unevaluable pts on the IMRT arm, eight (72%) withdrew consent due to wanting PBT only.”

Lin noted several limitations, including the fact that this trial was based on a single institution and was relatively small in size. He also noted that 80% of the PBT patients received Passive Scattering PBT, meaning that the results cannot be applied to intensity-modulated proton therapy. Additionally, the widespread insurance denial of PBT introduced a potential selection bias, particularly since some IMRT-group patients chose to receive PBT instead.

Discussant Charles Simone, MD, chief medical officer, New York Proton Center, agreed with Lin’s assessment of the study limitations. “The total toxicity burden may be clinically meaningful, but it's not a validated endpoint,” he said, adding that he appreciates TTB’s ability to reflect all toxicities’ cumulative impacts. “By focusing on a binary endpoint, you may underestimate the potential benefit of proton therapy. TTB as an endpoint does allow for improved statistical power by looking at a variety of potential endpoints, but, again, this is not a validated endpoint and something to be considered cautiously.”

Simone noted that Lin had omitted a slide from his talk that showed that the mean lymphocyte count and rate of grade four lymphopenia during chemoradiation improved with proton therapy. “This is important because these have previously been shown to be a predictor of overall survival in esophageal cancer,” he said. “With more patients, this may prove to be an important finding.”

Simone concluded by placing this trial in its larger context: “Protons were shown in this study to better spare organs at risk compared with IMRT,” he said. “This is the first trial across any disease type showing that proton therapy can reduce toxicities in a randomized fashion compared with IMRT, and the study showed it in several different ways.”

Simone closed by urging colleagues to accrue patients to the Phase III trial NRG-GI006: “Ultimately, even if GI006 is positive, we know that not all esophageal patients are going to benefit from proton therapy, and we need to better understand which patients and which subsets of patients can benefit the most.”