Over the last year or two, I have seen an increasing number of patients with multiple myeloma who are deeply worried that they have “failed” treatment because they are not in “complete response” (CR).  This phe­nom­e­non is gaining further steam with recent interest in “minimal re­sid­ual disease” (MRD).

In fact, with numerous educational programs, daily emails, and ubiqui­tous lectures touting a new regimen with even higher complete re­sponse rates, I am now almost as worried as them. Of course, the cause of my worry is not that patients have not achieved the magical complete re­sponse or minimal residual disease-negative status, but at how mis­in­ter­pre­ta­tion of data can lead to needless concern, unnecessary chemo­ther­a­py, increased side effects and cost of care, and even harm.

A lot of the recent hype surrounding complete response and minimal residual disease-negative status is the result of two factors.

The first is that prior to Revlimid ^[1] (lenalidomide), Velcade ^[2] (bortezomib), and other new drugs, a complete response was seen in less than 5 percent of patients treated with standard-dose therapy.  Now, this rate is climbing steadily with each new study, making investigators giddy with optimism.

The second factor is more subtle. It is based on a proliferation of studies in which patients who achieve a complete response or minimal residual disease-negative status are reported to live longer than those who do not achieve these levels of response.

On the surface, these two factors suggest that we can achieve complete response in more patients than before, and we should try our best to get to that point because patients in complete response live longer than those not in complete response. However, there are numerous caveats to this superficial interpretation of the data.

In newly diagnosed patients treated with induction, stem cell transplantation, and Revlimid maintenance therapy, the proportion of patients who get to a complete response is approximately 30 percent. Does this mean that 70 percent of myeloma patients have “failed”?

Even with more expensive triplet induction regimens such as Revlimid, Velcade, and dexamethasone ^[3] (Decadron) (abbreviated as RVD), only 40 percent of newly diagnosed myeloma patients achieve complete response. An aggressive seven-drug induction regimen followed by two back-to-back transplants, and three years of maintenance used in the “Total Therapy 3” regimen results in a complete response rate of 55 percent.

Should one-half of newly diagnosed myeloma patients panic that they have not reached a complete response? That would be over 10,000 worried patients each year in the United States alone.  The short answer is no; absolutely not.

Multiple myeloma is a remarkably heterogeneous disease; the outcomes vary dramatically depending on the patient’s chromosomal abnormalities. The type of myeloma one patient has may be completely different than the myeloma another patient has; it may not even be the same disease.

I am sure patients participating in support groups have seen and pondered about the marked differences in how fellow patients seem to respond to the exact same therapy.

Thus, the first point I would like to make is that myeloma is way too complex to have one set of rules. In some patients, a complete response is critical for long-term survival; while in many, lack of a complete response has no impact whatsoever on long-term outcome. There are countless patients with myeloma who have a persistent monoclonal (M) protein after initial therapy who live for years without progression and without needing additional therapy.

A classic study ^[4] by Dr. Bart Barlogie’s group looked at the impact of complete response in patients according to their chromosomal abnormalities in their myeloma cells.  His group found that most (more than 85 percent) of patients with myeloma lived the same length of time whether they achieved a complete response or not. Only in very high-risk patients, as defined by gene expression profiling, could one see the possible importance of complete response.

How then do we interpret the numerous studies that find superior survival in patients who achieve a complete response? In simple terms, what these studies show is that the achievement of complete response identifies patients with biologically better-risk myeloma.

These studies cannot be interpreted to mean that modifying (or adding) therapy in order to achieve complete response prolongs life. Such studies are yet to be done.

It is one thing to say that achievement of complete response is a prognostic marker, but quite another to take that to mean we need to treat patients until they reach complete response.  It is also incorrect from a statistical standpoint to compare the outcomes of patients in complete response versus those not in complete response, since all early deaths that occur before patients have time to respond will be included in the “no complete response” group. Even landmark analysis methods, which attempt to correct this problem, may not fully rectify the biases that occur.  Therefore, most study designs assessing the importance of complete response are fundamentally flawed.

I continue to maintain that in the absence of data from appropriately designed randomized trials, targeting complete response (or minimal residual disease-negative status) as a therapeutic goal for individual patients remains an unproven strategy for most patients with myeloma. It is a strategy worth testing in clinical trials, but it is not considered the goal of therapy in standard clinical practice.

Patients with low-risk myeloma can do just as well without achieving complete response. In addition, patients with a long history of monoclonal gammopathy of undetermined significance (MGUS) or smoldering multiple myeloma (SMM) prior to the diagnosis of myeloma often do not achieve a complete response with therapy, but usually do very well regardless.

Studies are ongoing now to refine the meaning of complete response in myeloma by increasing the stringency of the definition. For example, modern flow cytometric and molecular methods can be used to detect residual malignant myeloma cells even in patients who are in conventional complete response. The term “minimal residual disease-negative” refers to absence of residual cancer cells by these newer methods. However, this is a moving target; minimal residual disease-negative in myeloma does not automatically mean zero myeloma cells; it just means that no cells are detected up to the limits of resolution of the test.

I am actively involved in projects looking at innovative methods to detect minimal residual disease, and in studying the value of minimal residual disease-negative status. But I would be the first to acknowledge that we are far from having proven the importance or the specific role of these endpoints. In addition, one should be careful in separating an interesting hypothesis that needs to be tested in clinical trials from proven fact that is ready for incorporation in clinical practice.

Why not try and achieve complete response or minimal residual disease-negative status in all patients, whether or not the value of such metrics is proven? After all, why would anyone not want to eradicate all traces of myeloma?

One good reason is that we want to avoid unnecessary chemotherapy.

Let us consider a patient who has achieved a partial or very good partial response after a course of therapy. Modifying therapy or adding drugs in order to achieve a complete response in such a patient often results in more side effects and lower quality of life.

Similarly, choosing regimens for standard clinical practice based on “the highest complete response rate” would do the same: there is no free lunch. These strategies would be acceptable if there is clear proof that such an approach prolongs survival, but they would be unacceptable if driven by incorrect interpretation of data.

At this point, there is reasonable evidence that newly diagnosed myeloma patients with high-risk chromosomal abnormalities should be treated with sustained complete response as a clear therapeutic goal. But data supporting a similar approach in most other patients with myeloma is weak.

Second, regimens that produce higher complete response rates are typically more expensive, and are more cumbersome.

In many countries, patients and their families pay the entire cost of cancer drugs; some may go bankrupt trying to obtain these treatments. If there is clear evidence of meaningful survival improvement, it would be justifiable not to spare any reasonable costs. But it would be sad to have patients and their families cope with dual (health and financial) setbacks based on surrogate endpoints and inadequate data.

Finally, it is just not worth the mental worry.

We hematologists have no reservations recommending additional therapy to large cell lymphoma patients who fail to achieve complete response with standard therapy. But in that disease the meaning of complete response is quite different than in myeloma.

For most patients with myeloma, not achieving a complete response should not be construed as failure.  Part of the reason for this paradox is that the term complete response as defined in myeloma carries nowhere near the same implication as a complete response in large cell lymphoma or childhood leukemia. Part of it, more importantly, is because we are dealing with a heterogeneous condition ^[5] in terms of chromosomal abnormalities.

To argue against complete response is not an easy task. My goal is not to minimize the reassurance that patients in complete response get with that knowledge. Achieving a complete response, especially a sustained one, is indeed a good prognostic marker. But that does not mean that patients who fail to achieve that benchmark should panic; neither does it mean that they need to change course and try a different treatment regimen.

In fact, a large subset of patients who do not achieve complete response will live just as long as patients who do.

Dr. S. Vincent Rajkumar is a professor of medicine and chair of the Myeloma Amyloidosis Dysproteinemia Group at the Mayo Clinic in Rochester, Minnesota. His research focuses on clinical, epidemiological, and laboratory research for myeloma and related disorders.