'Pick the Winner' Trial in Rectal Cancer: Chemoradiation First

Liam Davenport

July 05, 2019

BARCELONA, Spain — Patients with locally advanced rectal cancer are more likely to achieve a pathologic complete response (pCR) if they are given chemoradiotherapy before chemotherapy as part of preoperative total neoadjuvant therapy (TNT), conclude German investigators.

The findings come from a randomized phase 2 "pick the winner" trial, known as CAO/ARO/AIO-12, that compared two TNT sequences based on oxaliplatin (Eloxatin, Sanofi Aventis) in more than 300 patients.

The trial showed that giving chemoradiotherapy before chemotherapy led to significant improvements in pCR rates over those seen historically, at 25% vs 15%.

Giving chemotherapy followed by chemoradiotherapy prior to surgery did not improve response rates. In addition, those patients who underwent chemoradiotherapy before surgery were more likely to experience grade 3/4 chemoradiotherapy-related toxicity.

The sequence of TNT also affected compliance with therapy.

Patients given chemotherapy first were less likely to receive all of their chemoradiotherapy, whereas those who received chemoradiotherapy first were less likely to complete their subsequent chemotherapy.

The results were presented here at the World Conference on Gastrointestinal Cancer (WCGC) 2019 (abstract O-011) by lead investigator Ralf-Dieter Hofheinz, MD, Interdisciplinary Tumor Center, University Hospital Mannheim, Germany.

"This is the first randomized trial comparing both of these potential strategies of total neoadjuvant therapy," he told the meeting.

"It turns out that, in our study, chemoradiotherapy followed by consolidation chemotherapy reached the primary endpoint and is now the winner," he said.

He added that this approach has been chosen for an upcoming phase 3 trial (ACO/ARO/AIO-18.1) that will compare preoperative oxaliplatin-based chemoradiotherapy and consolidation chemotherapy with standard fluorouracil (5-FU)–based chemoradiotherapy in intermediate- and high-risk rectal cancer.

Commenting on the study, Eric van Cutsem, MD, PhD, from the University Hospital Gasthuisberg, Leuven, Belgium, who is cochair of the congress, said that "the field of locally advanced rectal cancer is evolving, that is clear."

He told Medscape Medical News: "What we see is that there is a clear move towards total neoadjuvant treatment, including chemotherapy and chemoradiotherapy. We don't know yet completely what is the best strategy."

van Cutsem said the current study "builds on the strategy of rectal cancer."

He underlined that although the trial did not move the field forward per se, it offered greater understanding of treatment sequencing and how that affects compliance.

He personally believes that for advanced local tumors, probably not all patients should be irradiated and that more patients should be given combination chemotherapy.

van Cutsem continued: "Probably the best is going to be that, to personalize that a bit more, maybe in the future give more chemotherapy and then, in some patients, after combination chemotherapy, irradiate them if the mesorectal fascia is still involved."

Study Details

Hofheinz began his presentation by indicating that the rationale for TNT is that compliance is better and toxicity is lower compared with adjuvant therapy.

Moreover, early systemic treatment potentially improves control of micrometastases and increases tumor regression. This could, in turn, enhance resection rates and improve patient selection for organ preservation.

He said that TNT "has been investigated in a number of trials, and you have two potential options: to start with chemotherapy or to start with chemoradiotherapy."

The researchers therefore conducted the CAO/ARO/AIO-12 to compare the two approaches.

Although the trial was not designed to demonstrate a significant difference between the two sequences, the hypothesis was that TNT would increase the pCR rate from the historical rate of 15% achieved with standard chemoradiotherapy to 25%, with the better sequence being designated the winner, he explained.

The trial enrolled adults with histologically confirmed rectal adenocarcinoma that had been shown on MRI to be of stage T3c/d (with mesorectal fat spread >5 mm) or T4. The patients' ECOG performance status was 0 or 1.

The chemotherapy regimen consisted of three cycles of FOLFOX (folinic acid [leucovorin], 5-FU, and oxaliplatin). Chemoradiotherapy comprised standard radiotherapy plus 5-FU and oxaliplatin.

Patients randomly assigned to group A initially received three cycles of FOLFOX, followed by a rest period, followed by chemoradiotherapy, followed by a second rest period, and then underwent surgery, which was performed approximately 5 weeks after the last chemotherapy or radiotherapy.

Patients in group B received chemoradiotherapy, followed by a rest period, followed by chemotherapy and a second rest period. They underwent surgery 5 weeks after completing TNT.

The discussant for this study, Ali I. Shamseddine, MD, the American University of Beirut Medical Center, Lebanon, pointed out that there was a difference in the time to surgery between the two groups. He said that the difference in pCR seen between the two arms "may be due to delaying the surgery up to 90 days compared to only 45 days in group A".

Two of the 306 patients who began treatment achieved complete remissions with TNT and so did not undergo surgery. Several others also did not have surgery, for various reasons. A total of 284 patients underwent surgery.

The median age of the patients who started treatment was 60 years, and 65% were men. The circumferential resection margin was ≤1 mm in 31% of the patients in group A and 22% of the patients in group B.

Compliance and Toxicity Differences

Compliance with treatment was affected by the sequence in which patients were treated.

The proportion who received the total radiotherapy dose was comparable between the groups, at 91% for group A and 97% for group B.

However, patients in group A were less likely than those in group B to receive the full dose of either 5-FU, at 78% vs 87%, or oxaliplatin, at 76% vs 93%.

Compliance to the FOLFOX chemotherapy regimen was comparable in both groups of patients. In group A, 93% completed all three cycles of 5-FU; in group B, 90% completed all three cycles.

However, patients in group A were less likely than those in group B to receive no 5-FU chemotherapy at all, at 0% vs 7%. Patients in group A were more likely to complete all three cycles of oxaliplatin, at 92% vs 87%.

Hofheinz said that patients in group A were more likely than those in group B to experience grade 3/4 chemoradiotherapy toxicity, at 37% vs 27%.

In both groups, 22% experienced grade 3/4 chemotherapy-related toxicity.

Primary Endpoint Results

With respect to the primary endpoint, the researchers found that on an intention-to-treat analysis of all 306 patients, giving chemotherapy first in group A did not lead to an improvement in pCR over historical rates, at 17% (P = .210).

In contrast, group B patients, who received chemoradiotherapy first, experienced a significant increase in the rate of pCR over historical rates, at 25% (P = .0002).

The proportion of patients who achieved a pathologic and clinical complete response was 21% in group A and 28% in group B.

Hofheinz noted that "if we add the couple of patients who were not resected to the complete remissions, it turns out that in group B...one may achieve about 30% with complete clinical and pathological remissions."

Tumor regression of grade 4, indicating greater than 50% tumor regression, was observed in 20% and 28% of group A and group B patients, respectively.

Postoperative morbidity was comparable between the two groups. The majority (54% in group A and 66% in group B) experienced no complications with respect to the Clavien-Dindo classification.

Points for Discussion

Invited to discussing the findings, Shamseddine pointed out that TNT became of interest as a result of the shortcomings of postoperative adjuvant chemotherapy.

He discussed a 2012 study by Polina Khrizman, MD, and colleagues that showed that compliance with postoperative adjuvant chemotherapy is suboptimal — 27% of eligible patients with locally advanced rectal cancer patients never undergo it.

In addition, fewer than half of those who do undergo the treatment receive the full dose or complete the course without interruptions or delays.

That trial also showed that delaying adjuvant chemotherapy beyond 12 weeks postoperatively is associated with worse outcomes, with each additional 4-week delay increasing mortality by 14% — hence, the interest in intensifying therapy before surgery.

Shamseddine said that although the results from CAO/ARO/AIO-12 trial are "positive," there are several points that warrant discussion.

The first is the difference in time to surgery between the two groups — the group with better pCR underwent surgery sooner (within 45 days, vs 90 days).

The second concerns use of oxaliplatin, which is not a standard treatment for locally advanced rectal cancer and may have contributed to the higher toxicity observed in patients in group B.

Finally, he said that questions remain concerning the use of short- vs long-course chemoradiotherapy.

Shamseddine said that nevertheless, on the basis of these findings, the suggested risk-adapted treatment strategy for patients aged 75 years and younger who have a good performance status and T3c/d (>5 mm) or T4 disease is to administer preoperative short- or long-course chemoradiotherapy followed by chemotherapy.

For stage Ts low-lying tumors, radiotherapy or chemotherapy alone can be used, followed by surgery.

The use of postoperative doublet chemotherapy, Shamseddine added, should depend on the tumor pathology.

Future trials, he concluded, will provide answers to questions regarding the use of short-course chemoradiotherapy, chemotherapy alone without radiotherapy, selective elimination of radiotherapy, selective surgery, and the testing of new agents alongside radiotherapy.

Different Approach: Nivolumab Up Front

In the same session, another approach to the treatment of locally advanced rectal cancer was described by Takayuki Yoshino, MD, National Cancer Center Hospital East, Kashiwa, Japan, who presented data from the VOLTAGE study (abstract O-010).

This trial examined the efficacy and safety of nivolumab (Opdivo, Bristol-Myers Squibb) immunotherapy and subsequent radical surgery following preoperative chemotherapy in locally advanced rectal cancer.

Yoshino focused on 39 patients who had microsatellite stable disease. Of those patients, 28% achieved a pCR. This percentage rose even higher in selected groups of patients: pCR was achieved by 60% of patients with baseline programmed death ligand 1 (PD-L1) expression ≥1% and by 62% of patients with a CD8/eTreg ratio ≥2.

Six patients had PD-L1 expression ≥1% and an increased CD8/eTreg ratio at baseline. In this small group, the pCR rate was 83%.

However, none of the seven patients who had both lower PD-L1 expression and a lower CD8/eTreg ratio achieved a pCR< the rate in these patients was 0%.

van Cutsem told Medscape Medical News that these new data on immunotherapy are "preliminary" and that it is "not yet clear" what role it will play in locally advanced rectal cancer.

He said that it is known that checkpoint inhibitors such as nivolumab and pembrolizumab (Keytruda, Merck) are "radiosensitizers."

van Cutsem added, however, that "we don't have, let's say, a clear recommendation for the clinicians on how to integrate that" finding into clinical practice.

Josep Tabernero, MD, PhD, from Vall d'Hebron University Hospital, Barcelona, who is cochair of the congress, emphasized that the sequencing of drugs is one of the major challenges in treating patients.

He said: "Clinical trials provide the right questions and the right answers, but not everything is so simple, especially in the field of immunotherapy, where you see all these kinds of combinations that are really good."

However, he added, "You never know the right sequence, and unfortunately, the preclinical models in the field of immunotherapy are not really very predictive of what is going to happen in the human being."

That is why he and his colleagues "always propose more academic studies to study the right sequence."

Tabernero said: "This is usually something that is very difficult to do in initial pivotal phase 3, pharma-sponsored studies, but this is one of the places were academia can contribute a lot to generate more and more knowledge that is really fundamental, especially in the field of immunotherapeuticals."

The study by Yoshino and colleagues was funded by Ono Pharmaceutical. No other funding was declared. Hofheinz has received honoraria for lectures and consulting from Amgen, Bayer, Bristol-Myers Squibb, Boehringer, Merck, Merck Sharp & Dohme, Lilly, Roche, Saladax, Sanofi, and Servier. Yoshino has received research funding from Novartis, MSD, Sumitomo Dainippon Pharma, Chugai Pharmaceutical, Sanofi, Daiichi Sankyo, Parexel International, Ono Pharmaceutical. Shamseddine is a member of the advisory boards of Roche, Sanofi, Pharmamed, MSD, Bayer, Lilly, Janssen, Pierre Fabre, and Merck. He has received research grants from Roche, Sanofi, Novartis, GlaxoSmithKline, Merck, aned Bristol-Myers Squibb and educational materials and honoraria from Roche, Sanofi, Merck Sharp & Dohme, Amgen, Merck, and Pierre Fabre. No other relevant financial relationships have been disclosed.

World Conference on Gastrointestinal Cancer (WCGC) 2019: Abstracts O-011 and 0-010. Presented July 4, 2019.

For more from Medscape Oncology, follow us on Twitter: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: