One of the basic concepts of oncology is that you treat patients with different drugs once they've shown progression on a treatment, rather than continue that a patient has presumably become resistant to. However, there are some exceptions to this: many or most women with breast cancer continue the antibody herceptin (trastuzumab) even after progression, adding it to one chemo and then the next, and the same is often done with avastin in colon cancer and sometimes lung cancer as well. In the past few years, there has been interest in whether the patients who respond well to an EGFR inhibitor like tarceva (erlotinib) or iressa (gefitinib) should continue on it for a while or even forever after showing the first evidence of progression on an EGFR inhibitor.

It may help for us all to take a step back and remember that the goal of improving survival and slowing the progression of a tumor may occur not only if the cancer is shrinking or even if it's holding steady. If the cancer might potentially be growing quickly, even slowing the progression may translate to an improvement in how a patient does. In the lab, basic scientists examine the growth of a cancer in lab animals and consider it beneficial if the cancer progression is slower over time when a new treatment is added, compared with a placebo or some alternative approach. But in the grading system oncologists use, we don't discriminate between slow progression and faster progression -- it's just considered a disappointment and time to move on.

(Click to enlarge)

The point is that imperfect brakes is better than no brakes. In fact, in some settings of especially effective treatments for other cancers, some investigators noticed that patients who were progressing slowly on a previously very effective treatment showed a rapid rebound progression when they stopped the treatment (as if they jumped from the blue line to the yellow line in the figure above). So a few years ago, the folks at Memorial Sloan Kettering Cancer Center (MSKCC)studied a small number of patients with either EGFR mutations or a prolonged response to EGFR inhibitors who were showing some progression (abstract here). They did CT scans and PET scans right before and right after patients took a planned three week break from their EGFR inhibitors, and they also assessed how these patients were feeling. After restarting their EGFR inhibitors and repeating scans and checking patient symptoms, they added another novel agent called everolimus, an mTor inhibitor (see prior post for discussion of the novel agent approach of mTor inhibitors alone and combined with EGFR inhibitors). For the more visually oriented, this is the overall design of the complicated trial:

The trial enrolled only 13 patients, of whom only 10 had the complete collection of scans and follow-up as outlined. The study provided no firm conclusions, but the investigators noted that several patients who had been progressing slowly on an EGFR inhibitor appeared to note a symptomatic decline when they stopped them, and some (but not all) stabilized or even improved when they restarted their EGFR inhibitor:

The imaging results pretty much mirrored the symptomatic results (a rather soft measurement that could be marred by the placebo effect), with a few patients showing a signal of a worsening (upward lines) by CT and PET, then a more horizontal or even downward slope with restarting the EGFR inhibitor.

Interestingly, the investigators from MSKCC pretty much considered the addition of the mTor inhibitor everolimus as a dismal failure in the paper, because no patient achieved a significant response, but I think that the addition of the mTor inhibitor looked at least as helpful as resuming the EGFR inhibitor. I thought that their conclusion from this somewhat provocative but rather scant work that patients should continue on an EGFR TKI indefinitely was rather bold, and by that I mean tenuous or downright questionable.

With all due respect, I think it's helpful to remember that this is still murky ground and that none of us has all of the answers, including the folks at MSKCC, who happen to be unusually wedded to EGFR-based treatments compared to many other experts in the lung cancer field. I don't think that the majority of experts feel that patients who responded to EGFR inhibitors should remain on them forever. The idea they promote is that patients with sensitivity to EGFR inhibitors and then show progression probably only have resistance in a subset of their cancer cells, so it makes sense to keep the EGFR TKI going even when they've progressed. I think it makes no sense to say that about an EGFR inhibitor but not about a standard chemo agent on which a patient has had a nice response, except that we haven't identified the key molecular factors associated with chemo responsiveness. But we often see progression in one area with a response or stable disease in another area. With chemo, we less frequently consider continuing on a treatment where someone has progressed.

In addition, there are many experts who think that there is a strong possibility that giving chemo and an EGFR inhibitor concurrently may lead to a negative interaction of antagonism, so just adding chemo and continuing tarceva after progression may just hobble your new chemo (see prior post). The folks at MSKCC have tended to be on the very far end of the pro-EGFR TKI (with just about anything) spectrum, but even there, several of the experts don't believe in giving chemo and EGFR inhibitors together.

My general principles about continuing current treatment vs. switching for progressive disease are basically these:

1) I definitely don't think it's necessary to always abandon a treatment on which a patient has evidence of progression

2) The better tolerated the therapy, the more I'd be inclined to continue a treatment

3) The more subtle the progression (bulk and rate of change), the more I'd be inclined to continue a treatment

4) The fewer remaining treatment options available, the more I'd be inclined to continue a treatment

The last point is an important one. The curve at the very top of the post compares slow progression on current treatment with faster progression on no treatment at all. But what if the alternative is a new treatment that would work better than the current treatment? I'd say it's time to switch. What if continuing the current treatment while giving a new treatment negated the benefit of the new treatment? I wouldn't continue the old treatment, which is why I still don't want to give an EGFR TKI along with chemo. Even if there's some residual benefit from continuing an EGFR inhibitor many months beyond first progression (a debatable possibility, not a firm conclusion), you need to compare that against the anticipated benefit of something new.

Enough for now. This is controversial stuff, so we can continue to hammer on these ideas if people have questions, comments, or objections. And again, as I emphasize that nobody has all of the answers, whether they're from MSKCC or any other renowned institution, that includes me. This work is my interpretation of a complicated topic.