Volker Heinemann says early tumor shrinkage may be a better predictor than progression-free survival in colorectal cancer.
Early tumor shrinkage (ETS) may be a more effective surrogate marker than progression-free survival (PFS) for patients with metastatic colorectal cancer (mCRC), according to Volker Heinemann.
In an interview with CURE, Heinemann, director of the Comprehensive Cancer Center at Ludwig Maximilian University of Munich, discusses the connection between OS, depth of response, ETS, its role in several clinical trials, and what it could mean for the future of mCRC treatment.
What can we learn from ETS in metastatic colorectal cancer?
What we tried to find is a surrogate parameter that we can measure early on, which relates response to OS. The first step was to investigate only tumor response, and we found that early tumor shrinkage was in fact related to OS. Patients reaching tumor shrinkage of equal to or greater than 20 percent had a prolonged survival. This was, to some extent, not understood. Then, we looked at depth of response and we could demonstrate that early tumor shrinkage was a very good predictor of chemo-sensitivity and, thereby, of depth of response. Depth of response in and of itself was very well correlated to OS in CRC.
Most importantly, the depth of response goes one step beyond RECIST. RECIST will differentiate stable disease from partial response and from complete response, which categorically differentiates different response if you're looking for categories. Now, depth of response is a continuous parameter, which goes from, let's say, 0 percent to 100 percent, with 100 percent being complete response and 0 percent being no response at all. Therefore, by looking at depth of response as a continuous parameter, we can for a single patient, more precisely demonstrate how effective treatment was. Partial response just categorically defines a patient that has a tumor reduction of more than 30 percent.
What is the need for a surrogate marker to show OS in mCRC other than PFS?
PFS has been, until now, the most popular surrogate parameter predictive of OS, and has been used also in registration trials. With the advent of targeted agents, PFS in itself is losing its predictive power to some extent, specifically because patients are surviving for a very prolonged period of time.
PFS is now typically only useful for 10 months. In fact, there is a seeding effect that we cannot extend PFS to much more than 12 months in some studies, and for that reason we believe that PFS is no longer a very good surrogate parameter anymore for OS.
We believe that if you look specifically at first-line treatment response rate, or parameters of response such as ETS or depth of response, that is much more appropriate.
How is depth of response being used to analyze patient data?
What we could show specifically is that with regard to treatment intensification, this could mean either intensification with a triple chemotherapy like FOLFOXIRI or by adding an anti-EGFR agent to a chemotherapy doublet in RAS wild-type patients. In these kind of settings, we want to demonstrate that not only response by itself, but early tumor shrinkage and depth of response were significantly correlated to OS.
In these kinds of studies, these concepts have been evaluated and found relevant.
Why is it valuable to use depth of response?
In the most recent presentation that we had at the 2016 ASCO Annual Meting and the 2016 World GI Congress in Barcelona, Spain, we evaluated the SIRFLOX study in regards to depth of response. The first step was that we analyzed tumor burden, and that specifically with regard to tumor burden in the liver. That's because SIRT is active in the liver, so we looked at the liver tumor burden. Using a finite mixed modeling, which was also supported by an ROC analysis, we could define a 12 percent cutoff for tumor burden, which was interesting. We differentiated patients with more than a 12 percent tumor burden in the liver than those with equal or less than 12 percent.
We have two different groups. One group was more extensive with their tumor burden in the liver, and one was a smaller extent. We could understand the effect of SIRT in addition to chemotherapy was markedly greater in the patients with the larger tumor burden. This was evaluated with regard to depth of response. We could demonstrate that when we added SIRT to chemotherapy in patients with a greater tumor burden than 12 percent, we could show that depth of response was 77 percent, as compared to 57 percent in the comparator arm.
In other words, we demonstrated that by the addition of SIRT to chemotherapy, we could increase depth of response by roughly 20 percent. This was not standing alone, but was accompanied by a marked increase in PFS in the liver in this particular group.
I think it is clinically relevant and it gives us the hope that this improvement in PFS to some extent also relates to improvement in OS, which is finally the most important topic.
What are the broader implications of these results in colorectal cancer?
If we see a major improvement in PFS in a vital organ such as the liver, we conclude at the moment that there is a high probability that this will translate to prolonged survival. We have comparable data from the CLOCK study. The CLOCK study was a trial, which compared chemotherapy versus chemotherapy plus radio frequency ablation. Radio frequency ablation can be interpreted as a treatment-intensification with regard to liver metastases. Investigators could demonstrate that PFS could be markedly improved in the liver and overall, that improved PFS and also related to improved OS.