Global study sets new risk-based standard to personalise chemotherapy for colon cancer after surgery

Share :
Published: 4 Jun 2017
Views: 3347
Dr Axel Grothey - Mayo Clinic Cancer Centre, Rochester, USA

Dr Grothey presents, at a press conference at ASCO 2017, analysis from six medical trials with over 12,800 patients, finding that 3 months of chemotherapy was nearly as effective as 6 months in patients with relatively lower recurrence risk and caused fewer side effects, particularly nerve damage.

Read the news story for more. 


Thank you for giving me a chance to present this late breaking abstract on behalf of the co-authors of the IDEA collaboration, particularly Dr Qian Shi who is going to be the lead author at the plenary session presentation.

When you go back to the treatment of stage 3 colon cancer nothing had really changed since 2004 when six months of oxaliplatin based therapy with FOLFOX or CAPOX, capecitabine oxaliplatin versus 5FU oxaliplatin, regimen had become standard of care with curative potential for patients. The problem that we are facing as clinicians, and patients in particular, is that longer duration of oxaliplatin based therapy is associated with dose dependent neurotoxicity that is debilitating for many patients short term and long term. It’s nerve damage in the form of numbness, tingling, pain which can really persist for the rest of the patient’s life. So a shorter duration of treatment without compromising on efficacy would really be of benefit for patients and healthcare resources.

We started this academic collaboration more than a decade ago. Six different phase III trials conducted in twelve countries, really a worldwide effort, to eventually include 12,834 patients with stage 3 disease, interrogating this duration question. We needed this high number of patients to really have a precise estimate of how much efficacy we would compromise with a shorter duration of therapy and really have a so-called non-inferiority design of clinical trial. So this is where these studies were conducted, really truly a worldwide effort.

Now, what we are really talking about here is balancing less toxicity and potential lack of a higher risk of recurrence. The question is, of course, how much loss of efficacy, meaning anti-tumour efficacy, would we be willing to compromise for much better toxicity outcomes, lower toxicity, so which difference is acceptable? That’s the non-inferiority question.

So the studies, all of these studies, six studies with these 12,800 patients, randomised patients to six months of therapy or three months of therapy and FOLFOX or CAPOX, the actual chemotherapy regimen was up to the investigators and patients. This was not a comparison between these two regimens. The objective was really to reduce side effects without giving up too much efficacy. When you put this into statistical terms in this non-inferiority design you need to look at the upper limit of the hazard ratio 1.12. Dr Shi will actually explain this in her plenary session presentation more and more. So anything that is below 1.12, the upper limit of the hazard ratio, would be considered non-inferior according to our assessment. We actually went to patient advocates and physicians and said, ‘How much are we willing to compromise on outcome, potentially on outcome, for less toxicity?’

These are the results. First of all the toxicity and you can see that for both chemotherapy regimens, FOLFOX and CAPOX, just highlighting on neurotoxicity, we have a dramatic reduction of neurotoxicity. About a third less patients or three times more toxicity when you use six months compared to three months, highly statistically significant. We know that three months’ duration is much less toxic.

Now, what about efficacy? We look at Kaplan-Meier curves for disease free survival and this is the outcome. Three months versus six months which seems to track very much on top of it. When you do the math actually we compromise, we reduce, potentially reduce, efficacy and disease free survival by 0.9% at three years. Unfortunately the hazard ratio crossed the upper limit, crossed 1.12. So for all patients combined, based on the assessment that we made, non-inferiority was not achieved but the difference in efficacy is very small. Now we come into clinical judgement – how much do we really take out of this? How much are we willing to push patients into six months of therapy for a very minimal difference in disease free survival?

Also we have 12,800 patients so we could do subgroup analysis, pre-planned subgroup analysis, and I want to highlight two which I think are clinically relevant. Number one, we divided patients into low risk tumours and patients with high risk tumours identified by tumour T stage and lymph node involvement. As you can see here, the three year disease free survival rate between the low risk and high risk tumours differs dramatically; it’s 83% for low risk tumours, 63% for high risk tumours. It was very clear to see that three months’ treatment duration was adequate for patients who had a low risk tumour burden, meaning 60% of patients. So those patients should not receive more than three months of therapy and that’s the majority of patients with stage 3 disease.

There was also an interesting treatment related effect. It seems to be that three months of capecitabine oxaliplatin were adequate for all patients independent of stage whereas for FOLFOX for the high risk tumours you needed six months of therapy. It will be further explained in the plenary session presentation.

So the IDEA group comprised of investigators around the world came up with a consensus statement which I hope will make inroads into guidelines and clinical practice. And this is how I have embraced my clinical practice. For the majority of patients with stage 3 disease, 60% low risk group, we do not recommend more than three months of adjuvant chemotherapy with an oxaliplatin based regimen. For the higher risk patients we can make an assessment based on tolerability of therapy, patient preference, risk of recurrence and the choice of therapy, how long we should push beyond the three months of therapy. That’s really a discussion we can have with patients and the IDEA collaboration really gave us the data to really provide a framework for discussions on risks and benefits for adjuvant therapy here. A shorter duration of therapy is clearly associated with a decreased risk of neurotoxicity. As I said, low risk cancers this is the immediate consequence here – 60% of stage 3 disease should not receive more than three months of therapy with an oxaliplatin based regimen. That applies to 20,000 patients a year in the United States and you can easily see that translates into hundreds of thousands of patients worldwide. There is not a direct comparison between these two different regimens. It seems to be that if you are using CAPOX, capecitabine oxaliplatin, three months is adequate.

Let me finish by giving some very important inordable [?? 6:40] facts. You already heard IDEA represents an international academic collaboration that was conducted over more than a decade. It’s by far the largest prospective study conducted in the history of colorectal cancer research. It was funded by public funds and by philanthropy. There is not a company interest involved and who is interested in shortening the duration of therapy from a commercial perspective? Those trials which have immediate impact for patient care, which are critical for patient care, can only be conducted with a publically funded clinical trial system. This is really what we stand for here, NIH and other mechanisms around the world that really contributed funding for the study. Thank you.