Patients who had advanced gastroesophageal cancer but were considered unsuitable for full-dose chemotherapy because of their advanced age and/or frailty “had an improved patient experience with no significant detriment in cancer control” when treated with reduced-intensity chemotherapy in the phase III randomized GO2 trial. “The GO2 clinical trial shows that the goals of palliative chemotherapy in the older and/or frail population, including but not limited to cancer control, may be better achieved using treatment at doses well below those currently regarded as standard,” the trial investigators concluded in a report in JAMA Oncology.1
“The intensity of the chemotherapy schedules developed over the years doesn’t really match the population who are most at risk of getting cancers of the gastrointestinal tract.”— Matthew T. Seymour, MD
Tweet this quote
Chemotherapy schedules that have been developed for gastrointestinal cancers, including esophageal, colon, and pancreatic, “are effective and well proven. However, when you look at the populations of patients who participate in the studies that validated those schedules, they don’t match the demographics of the real-life population of patients who suffer from the disease,” the study’s corresponding author, Matthew T. Seymour, MD, said in an interview with The ASCO Post. Dr. Seymour is Professor of Gastrointestinal Cancer Medicine, Institute of Oncology, St. James University Hospital, University of Leeds, United Kingdom.
Matching the Population
Particularly for tumors of the gastrointestinal tract, “the intensity of the chemotherapy schedules developed over the years doesn’t really match the population who are most at risk of getting the cancers,” Dr. Seymour added. As noted in the study, more than half of deaths due to gastroesophageal cancer occur in people over 75 years, “many of whom are frail and with comorbidities. But evidence guiding treatment for vulnerable older patients is poor: standard chemotherapy regimens were developed in trials involving predominantly nonfrail, noncomorbid patients of a median age younger than 65.”
In a survey prior to the GO2 study, many oncologists reported having to adapt the evidence from the standard clinical trials “to suit the patients they are actually treating,” Dr. Seymour said. So, GO2 investigators decided to “study the patients who are frail and older because those are the people who are getting chemotherapy for this disease, and we need to know the best way of doing that,” Dr. Seymour said. “The most important question for us was, ‘Is if feasible to reduce the intensity of chemotherapy to make the treatment less toxic without compromising substantially the effectiveness of the treatment?’” The results showed it was feasible.
No Benefit From Higher Doses
Of the 514 patients randomly assigned to one of three dose levels of oxaliplatin plus capecitabine, those who received the highest dose, level A (oxaliplatin at 130 mg/m2 on day 1, capecitabine at 625 mg/m2 twice daily on days 1–21, on a 21-day cycle), did not have significantly longer progression-free survival compared with those who received an intermediate dose, level B (doses 0.8 times those in level A), or the lowest dose, level C (doses 0.6 times those in level A). “Noninferior progression-free survival was confirmed for levels B vs A (hazard ratio [HR] = 1.09, 95% confidence interval [CI] = 0.89–1.32) and C vs A (HR = 1.10, 95% CI = 0.90–1.33),” the investigators reported. In addition, “level C produced fewer toxic effects” than A or B and better overall treatment utility.” (Overall treatment utility is a measure combining efficacy, toxic effects, quality of life, and patient value/acceptability.)
Patients could enter a second phase of the trial, comparing level C chemotherapy with best supportive care, “if the patient and clinician agreed the indication for chemotherapy was uncertain,” the investigators noted. Among these 45 patients, “overall survival was nonsignificantly longer with chemotherapy: median 6.1 months vs 3.0 months (HR = 0.69, 95% CI = 0.32–1.48, P = .34),” the researchers reported. The patients in this second randomization “were frailer,” Dr. Seymour pointed out. “So, perhaps not surprisingly, the level C chemotherapy produced more toxicity in those patients than it did in the patients who had gone into the main randomization.”
No Subgroup Benefited
Patients were recruited from oncology clinics in the United Kingdom. “In the absence of established objective frailty thresholds and given the complex interrelations of frailty and advanced age, we used oncologists’ clinical judgment in selecting patients,” the authors stated. The median age of study patients was 76 years; 385 (75%) were men, and 299 (58%) were severely frail.
Patients had locally advanced and/or metastatic gastroesophageal cancer and had not received chemotherapy for advanced disease, although some patients “could have received chemotherapy at the time of their original surgery,” Dr. Seymour said. “There was a mixture of patients who had disease that was already advanced at the time of diagnosis and some patients who had prior surgery and then relapsed. We didn’t exclude patients who had prior adjuvant chemotherapy, but this had to be their first treatment for noncurable advanced disease.”
“Should we start to look at lower-dose chemotherapy treatments in patients who would have been too young and fit to go into the GO2 study?”— Matthew T. Seymour, MD
Tweet this quote
“No subgroup benefited from higher doses,” the investigators reported. The lowest dose level “produced better overall treatment utility even in younger or less frail patients.”
“That was a real surprise,” Dr. Seymour told The ASCO Post. “In a trial like this, where you have a mixed collection of patients, we thought we might find a group, perhaps the younger or fitter patients or those with fewer comorbidities, that benefited more from having a high dose of chemotherapy. But, in practice, we didn’t find that. Even the younger and fitter patients seemed to have a better outcome if they had level C.”
More Reliable Treatment Delivery
Chemotherapy could be “stopped in the event of radiologic or clinical [disease] progression, unacceptable toxic effects, or patient choice,” the authors noted. Patients most likely to stop were those receiving the highest doses of chemotherapy. The toxicities most likely to result in treatment discontinuation were diarrhea, “sometimes nausea, and more often just a feeling of tiredness, a sense of lethargy, and a lack of joy, which makes it really hard to enjoy life with your family and carry on with your normal activities,” Dr. Seymour added.
“The rate of toxic effects fell with reducing dose levels, and treatment delivery was more reliable,” the authors noted. Patients receiving the lowest dose levels were more likely to complete their first three cycles without dose reduction or stoppage.
“This wasn’t a fixed-duration course,” Dr. Seymour explained. Patients were not told they would have to stop after a set number of cycles but could continue chemotherapy if it seemed to be of benefit. “In practice, what we found is that patients who had been allocated a lower dose ended up receiving more cycles.”
Questioning the Cytotoxic Approach
“It is accepted practice in oncology that when we treat patients with a standard dose of treatment, a proportion of patients will have toxicity that requires them to delay or perhaps stop treatment for a while. That can be a traumatic experience for patients,” Dr. Seymour commented. “They have a period when they are not on chemotherapy at all. Then they end up on a dose reduction, and that reduction may be down to a level that is lower than they could have tolerated had they started. They’ve got all the worry and emotional turmoil that goes with the experience, and some people never end up restarting treatment after a dose reduction. Part of the benefit we saw in the trial was simply that if you start people on a lower dose, they are going to be able to receive without those interruptions, then they are more likely to benefit.”
Whether or not this benefit would be seen with other chemotherapy regimens and with other types of cancer remains unknown. “It is dangerous to extrapolate too much from a single trial,” Dr. Seymour said. “But it certainly begs the question of whether the general philosophy of cytotoxic chemotherapy practice, which was developed back in the 1960s and 1970s, very much along the lines of maximum tolerated dose as the way to go, may, in fact, be the wrong approach, particularly for the older and frailer population. What we found is that we can reduce the doses and make life better for patients without compromising their survival.”
Overall Treatment Utility
Patient experience was compared using overall treatment utility, “a novel composite endpoint,” the authors noted, “combining clinical efficacy, tolerability, and the patient’s own assessment of treatment value and acceptability.” Overall treatment utility was developed for the FOCUS2 study among frail and older patients with colorectal cancer, which was reported in 2011.2
“We have used overall treatment utility only three times,” Dr. Seymour noted. “We used it in that FOCUS 2 trial. We then developed it a bit more with the 321GO trial,”3 a randomized feasibility trial for patients with gastroesophageal cancer, “sort of a pilot study before GO2,” Dr. Seymour noted. “Then we used it here in GO2. Each time, we learned a little more. In the new version that we used in GO2, there is more emphasis placed on quality-of-life changes than there was previously.”
In the GO2 trial, overall treatment utility for patients receiving chemotherapy was scored at 9 weeks after initiation and included computed tomography (CT), clinical assessment of cancer progression status, toxic effects and serious adverse events, and patient value/acceptability, ascertained in answers to two questions in a questionnaire before patients received their scan results: “Since you started chemotherapy, how worthwhile do you think your treatment has been?” and “How much has your chemotherapy interfered with your normal daily activities?” For both questions, the choices were “not at all, a little, quite a bit, or very much.”
“Overall treatment utility, assessed in all 514 patients by intention to treat, was good in 196 patients (38%), intermediate in 149 (29%), and poor in 169 (33%),” the authors reported. There were more good (43%) and fewer poor (29%) overall treatment utility outcomes in the group receiving the lowest intensity chemotherapy, but the difference among the groups was not statistically significant.
The conclusion of the GO2 study report recommends “further development of overall treatment utility to capture the complexity of a multidimensional aspect of assessing the outcome of cancer treatment, including its value to patients and its adverse effects.” That is occurring within other ongoing studies by research groups in Europe and elsewhere, Dr. Seymour noted.
“When you do a composite endpoint, which is taking in elements from different types of measurement, it is necessary to examine the relevant importance of those different elements,” Dr. Seymour stated. “We always need to be aware that these composite endpoints should be closely examined and can sometimes be improved.”
Other endpoints, such as progression-free survival, may be “convenient for oncologists,” Dr. Seymour said, but “progression-free survival itself is not an endpoint that is of any actual direct relevance to patients. It doesn’t tell you how long you are symptom-free. It doesn’t tell you how long you live. It doesn’t tell you anything very much, except how long your CT scan was stable, and that is not usually hugely important for the way you enjoy your life.”
Probability Model
“In a multivariate analysis in 522 patients with all variables available, baseline frailty, quality of life, and neutrophil-to-lymphocyte ratio were independently associated with overall treatment utility,” the investigators noted. “These factors can be used to calculate a predictive score: (0.27 if not severely frail) + (0.39 if EQ-VAS ≥ 50) + (0.34 if neutrophil-to-lymphocyte ratio ≤ 4.0).” EQ-VAS stands for EuroQOL visual analog scale and neutrophil-to-lymphocyte ratio. “This score (range 0–1) translates into the probability of good, intermediate, or poor overall treatment utility at 9 weeks,” the authors noted. “Thus, a slightly frail patient with baseline EQ-VAS = 55 and neutrophil-to-lymphocyte ratio = 3.0 (predictive score = 1) has a 44% probability of good and a 27% probability of poor overall treatment utility. Conversely, a severely frail patient with baseline EQ-VAS = 45 and neutrophil-to-lymphocyte ratio = 5.0 (score = 0) has only an 18% probability of good overall treatment utility but a 57% probability of poor overall treatment utility.” Details on using and interpreting the predictive model are included in the study appendices.
“When you derive a model from a specific patient population, you have to be careful that you apply that model only for patients who would have fit into that same patient population,” Dr. Seymour stressed. “Within the population of frail and older patients, it is quite helpful” to be able to measure the baseline quality of life, the neutrophil-to-lymphocyte ratio, and performance base to get an idea from those criteria whether the patient is likely to have a good overall treatment utility with chemotherapy,” Dr. Seymour said. “I wouldn’t want anybody to take our model and apply it to a fit 30-year-old. That wouldn’t make sense.”
Future Phases of Research
“In the GO2 trial, all of our assessments at baseline were observation. We took a lot of measurements, but we didn’t try to change anything,” Dr. Seymour said. The next phase will be to see “what happens if we take somebody who we predict is going to have a poor outcome, but we actually do intensive work to get that patient into a better state. Then, can we correct things, and, if we do that, can we then get a better outcome?”
Also planned is a “trial using the principles from GO2 but applying them to older, frailer patients with potentially operable colon cancer, giving them relatively low-dose chemotherapy prior to their operation, using that opportunity to get people into the best possible fitness for their surgery,” Dr. Seymour said. He expects that trial to be opening for recruitment internationally later in 2021.
A somewhat surprising result of GO2 was that “even the younger and fitter patients seemed to benefit more from the lower dose of chemotherapy,” Dr. Seymour noted. “That begs the question of whether we should start to look at lower dose chemotherapy treatments in patients who would have been too young and fit to go into the GO2 study.”
That could apply not only to gastrointestinal cancers, “but across a range of other cancers as well. Certainly, when we are looking at other conditions for which we use cytotoxic drugs, then we need to ask that question,” Dr. Seymour said. He cautions, however, that “GO2 doesn’t tell us anything about the principles of using noncytotoxic medications. The principles of dosing for targeted therapies and immunotherapies are completely different from those for cytotoxic therapies. So, I think we should be careful about over extrapolating.”
DISCLOSURE: Dr. Seymour received grants from Cancer Research UK during the conduct of the study and institutional research funding from Amgen and AstraZeneca.
REFERENCES
1. Hall PS, Swinson D, Cairns DA, et al: Efficacy of reduced-intensity chemotherapy with oxaliplatin and capecitabine on quality of life and cancer control among older and frail patients with advanced gastroesophageal cancer: The GO2 phase 3 randomized clinical trial. JAMA Oncol 7:869-877, 2021.
2. Seymour MT, Thompson LC, Wasan HS, et al; FOCUS2 Investigators; National Cancer Research Institute Colorectal Cancer Clinical Studies Group: Chemotherapy options in elderly and frail patients with metastatic colorectal cancer (MRC FOCUS2): An open-label, randomised factorial trial. Lancet 377:1749-1759, 2011.
3. Hall PS, Lord SR, Collinson M, et al: A randomised phase II trial and feasibility study of palliative chemotherapy in frail or elderly patients with advanced gastroesophageal cancer (321GO). Br J Cancer 116:472-478, 2017.
