Survival Of Nonsecretory Multiple Myeloma Patients Improves Over Last Decade
Results of a recent retrospective study conducted at the Mayo Clinic indicate that the survival of patients with nonsecretory multiple myeloma has improved over the last decade.
Nonsecretory multiple myeloma is a rare form of myeloma in which a patient’s disease cannot be diagnosed or tracked by the presence of monoclonal protein in the blood or urine. This is not the case for most myeloma patients, who have “secretory” disease which can be tracked using lab results such as a patient’s “M-spike” or free light chain levels.
The authors of the new study found that, among multiple myeloma patients diagnosed between 1973 and 2001, the median overall survival was similar for patients with nonsecretory myeloma (3.6 years) and secretory myeloma (3.5 years).
Overall survival improved in both secretory and nonsecretory myeloma patients diagnosed between 2001 and 2012 compared to the earlier period. However, the improvement in survival was greater in patients with nonsecretory multiple myeloma. The median overall survival for nonsecretory patients diagnosed between 2001 and 2012 was 8.3 years compared to 5.4 years for secretory patients diagnosed in the same time period.
The researchers note that the improvement in survival in the later period occurred during the time when the newer, “novel” therapies for multiple myeloma began to see widespread use. It is not clear, however, why the introduction of these newer therapies had a greater impact on the survival of nonsecretory myeloma patients than secretory patients.
Background
Multiple myeloma is a cancer in which the body overproduces plasma cells. In healthy people as well as people with myeloma, plasma cells produce antibodies – also known as immunoglobulins – that help the body fight infection.
Each plasma cell produces one type of immunoglobulin, and the body produces many types of plasma cells. Together, the different types of plasma cells produce the range of different kinds of immunoglobulin needed by the body.
The excess plasma cells in myeloma patients, however, typically overproduce a single type of plasma cell. When this happens, it leads to overproduction of the immunoglobulin associated with that type of plasma cell. The accumulation of this immunoglobulin is what is detected in lab tests that track, for example, a patient’s M-spike.
In patient with nonsecretory multiple myeloma, on the other hand, the plasma cells do not produce (“secrete”) excess immunoglobulins. As a result, the disease cannot be diagnosed or tracked using standard blood and urine tests.
Nonsecretory myeloma can be detected, however, through bone marrow biopsies, PET scans, and through its impact on a patient’s bones, hemoglobin levels, and other signs of the disease (for more information on nonsecretory myeloma, please see this physician column by Dr. Bijay Nair).
According to the authors of the new study, there is currently limited information available about the clinical course and prognosis of patients with nonsecretory multiple myeloma. To shed more light on the topic, the researchers conducted a retrospective analysis of data collected at their institution.
Study Design
The authors searched the Mayo Clinic multiple myeloma database to identify nonsecretory myeloma patients diagnosed between January 1973 and June 2012. Patients were identified as having nonsecretory myeloma if their serum and urine immunofixation did not show any monoclonal protein at diagnosis and at all subsequent check-ups.
A total of 124 nonsecretory patients were identified. The median age of the patients was 62, and 71 percent of the patients were diagnosed before 2001.
Data on initial therapy was available for 121 patients, and 116 of them received drug-based therapy as part of their initial treatment. Of those, 77 percent received conventional chemotherapy and 23 percent received treatment with one or more novel agents, such as thalidomide (Thalomid), Velcade (bortezomib), and Revlimid (lenalidomide). Also, 27 percent of the patients who were initially treated with either conventional chemotherapy or novel agents went on to have an autologous stem cell transplant following their initial therapy.
Results of bone marrow biopsies from the time of diagnosis were available for 119 patients; 70 percent of these patients had a bone marrow plasma cell percentage of 10 percent or greater.
Serum free light chain testing was performed in 29 patients at the time of diagnosis. The free light chain ratio was abnormal in 65 percent of these patients.
The median follow-up time was 102 months.
Survival results for the nonsecretory patients were compared to those for 7075 secretory myeloma patients in the Mayo Clinic database who were diagnosed during the time period covered by the study. The study authors note that the number of nonsecretory and secretory patients in the database indicates that nonsecretory myeloma accounted for 1.7% of all multiple myeloma diagnoses in their center's database for the study period.
Study Results
The researchers looked first at how nonsecretory patients responded to treatment and changes in their survival over time.
They found that the overall response rate to initial treatment did not differ significantly between the nonsecretory patients who received conventional chemotherapy (88 percent) and those who received novel agents (84 percent) as initial treatment.
The median progression-free survival after initial therapy was 28.6 months during the entire period covered by the study, and the median overall survival was 49.3 months.
The median overall survival of the nonsecretory patients improved significantly over time. It was 43.8 months for patients diagnosed prior to 2001, increasing to 99.2 months for patients diagnosed between 2001 and 2012.
The median overall survival also was significantly longer for the nonsecretory patients who received novel agents as part of their initial therapy (not yet reached), compared to those who received conventional chemotherapy (46.1 months).
The authors do not report, however, if there was any difference in progression-free survival between patients treated with conventional chemotherapy and those treated with novel agents, nor do they report whether there was any change in progression-free survival over time.
Overall survival was higher in nonsecretory patients who had a normal free light chain ratio at diagnosis than in those who had an abnormal ratio. This result also has been found in previous studies that have looked at the impact of the free light chain ratio on the prognosis of myeloma patients in general (not just nonsecretory patients).
The researchers turned next to a comparison of overall survival trends in nonsecretory and secretory myeloma patients.
They found that, among patients diagnosed before 2001, the median overall survival was similar in patients with nonsecretory myeloma (3.6 years) and patients with secretory myeloma (3.5 years).
Overall survival also increased for both nonsecretory and secretory patients in the 2001-2012 time period compared to the earlier period. The improvement in survival that occurred during the later period, however, was greater for nonsecretory patients.
The median overall survival for nonsecretory patients diagnosed between 2001 and 2012 was 8.3 years, compared to 5.4 years for secretory patients diagnosed in the same time period.
The researchers considered several potential explanations for the observed divergence in overall survival, but were not able to develop a satisfactory explanation.
Differences in disease stage at diagnosis, for example, could not account for the divergence. There also was not enough data to determine if differences in chromosomal abnormalities at the time of diagnosis could be the cause of the divergence in survival. Only nine of the nonsecretory patients had their chromosomal abnormalities checked at the time of diagnosis.
The authors did note, however, that among the nine nonsecretory patients for whom chromosomal abnormality information was available for the time of diagnosis, the types of abnormalities that were present did not seem different from those normally found in newly diagnosed secretory myeloma patients.
For more information, please refer to the study by Chawla, S. et al., “Clinical Course and Prognosis of Non-Secretory Multiple Myeloma,” in the European Journal of Haematology, February, 2015 (abstract).
Related Articles:
- Early Use Of Radiation Therapy Associated With Shorter Survival In Multiple Myeloma
- Importance Of Factors Affecting Multiple Myeloma Survival Changes With Patient Age
- Sustained Complete Response To Initial Treatment Associated With Substantial Survival Benefit In Multiple Myeloma
- Revlimid, Velcade, and Dexamethasone, Followed By Stem Cell Transplantation, Yields Deep Responses And Considerable Overall Survival In Newly Diagnosed Multiple Myeloma
- Stem Cell Transplantation May Be Underutilized In Multiple Myeloma Patients In Their 80s
Very interesting article, thank you. I would disagree with the authors about the note that the nine nonsecretory patients, whose genetic information was available, had similar chromosomal abnormalities as normally present in myeloma patiens. In their study there were only two moderate or high risk nonsecretory patients (del 17) and all the other seven nonsecretory patients had standard risk chromosomal abnormalities (trisomies, normal, or t11;14). But as they said, there were not enough patients who had their chromosomal abnormalities checked to draw any conclusions.
There are only 1.7% of non secretors in the sample, with a vastly greater number of secretors. I would not think the median OS can be to readily equated in the two groups with such disparity in numbers of patients in the groups.
What are the ranges for figures for OS in the two groups (what does the OS distribution graph look like for the two groups)?
As with the lack of full profile of chromosomal abnormalities to add to the picture, things may bot be quite so clear.
Thank you for your comments, finn and Edna.
Finn - I agree that it is hard to draw conclusions about the role of chromosomal abnormalities in the divergence in survival in the later period, given the limited information that is available about such abnormalities among the nonsecretory patients. For what it's worth, the Mayo researchers did say in their paper that they consider the distribution of chromosomal abnormalities among the nine nonsecretory patients to be "similar to secretory myeloma".
Edna - The divergence in overall survival between secretory and nonsecretory patients in the later period is quite clear, and it is statistically significant (p=0.03). Neither we nor the authors would have highlighted the divergence if this were not the case. Given the large number of secretory patients in the Mayo sample, the overall survival of those patients is quite precisely estimated. So it's really the estimate of the median survival of the nonsecretory patients that is subject to uncertainty, due to the comparatively small number of those patients. Still, as I mentioned, not only is there a clear divergence in the survival curves for the later period – a divergence not seen for the earlier period – the divergence is also statistically significant.
Although the authors do not discuss this in their paper, there seem to me to be two forces working in opposite directions when it comes to the survival of nonsecretory patients versus secretory patients.
The force working in favor of longer survival for nonsecretory patients is that, because their disease does not secrete protein into the blood, they are much less likely to have kidney damage than secretory patients.
On the other hand, nonsecretory disease is harder to monitor, making it harder to catch relapse and treat it quickly.
What may have happened in the later period is that the ability to monitor nonsecretory patients improved versus the earlier period. Imaging techniques, for example, are an important tool in the tracking of nonsecretory disease, and they have improved markedly in the last 15 years.
Improvements in monitoring would make the potential downside of nonsecretory disease less significant, allowing the relative upside of nonsecretory disease to become more apparent.
This is just a personal hypothesis for why the divergence may have occurred. The issue certainly requires more study.
It is true that there is a statistically significant difference in the calculated OS median between the groups, and if you look at the graph, the distributions look quite different and convincing. However, as Edna pointed out, due to small number of nonsecretory patients in the study, it is important to know the OS confidence interval in each group. The authors kindly provided this information and the 95% confidence interval for the secretory group is 5.1-5.6 years (median 5.4 years) and for the nonsecretory group 4.9 years-not reached (median was 8.3 years).
Thank you for the follow-up, Finn.
Yes, the 95 percent confidence intervals for the two estimates of median overall survival do overlap. However, as I'm sure you're well aware, this is taken into account in the p-value of 0.03 (3 percent) that I mentioned earlier, demonstrating the statistical significance of the difference in median survivals. That p-value says that, given the distribution of overall survival in the two groups, and given the size of the two groups, there is only a 3 percent chance that the median survivals are the same.
It's unfortunate, actually, that the authors chose to highlight the *median* survival of the two groups to demonstrate the relative improvement in survival for nonsecretory patients that has taken place. Because of the very long survival of the nonsecretory patients in the later period, and because the second period the authors studied is not that long ago, the median survival of the nonsecretory group is based on a limited number of observations. So the estimate of the median survival of nonsecretory patients in the second period is not very precise, which is reflected in the wide confidence interval for estimate.
It perhaps would have been better to look, for example, at differences in how many patients in each group survived at an earlier, frequently used, benchmark, such as 50 months. For both groups – but particularly for nonsecretory patients – estimates of the share of patient surviving up to that point are likely to be more precise than estimates of each group's median survival.
And, in fact, if you look at 50-month overall survival, you find that it was about 80 percent for nonsecretory patients, and 60 percent for the secretory group. For both groups, I suspect the confidence intervals for these estimates are probably small enough that they are unlikely to even come close to overlapping.
My point is that, as you mention, the overall survival "distributions look quite different and convincing". Overall survival is higher for the nonsecretory group at every time after diagnosis, and often substantially so. It therefore seems unlikely that, as data continues to be gathered for the two groups going forward, there will be anything but a continuation in the divergence in overall survival for these two groups. Wouldn't you agree?
This is clearly different from overall survival for the two groups for the earlier period, where the overall survival lines almost overlap at times for extended periods of time.
This is why its seems quite safe to conclude that there has been divergence in the overall survival of the two groups over time.
Since it may clarify some of the points made in the above comments, here are the overall survival curves for the two groups of myeloma patients for the second period discussed in the study (2001-2012):
Wow ... this is interesting and encouraging. I am a 100% non-secretory MM patient and of course this gives me great hope while, at the same time, everyone must understand the limits nonsecretors have. Yes, it is a big deal we do not produce the an M spike and therefor our kidneys are in a much safer zone, but we cannot be diagnosed any other way then a bone marrow biopsy, and now PET, but at diagnosis it was bone marrow. I had 6 doctors tell me I DID NOT have cancer and just a singular plasmacytoma because all of my tests were completely normal. Never been sick a day in my life, the only thing I had symptom wise was back pain. I got lucky; an oncologist insisted on a biopsy, or I might have been much further along before they found it.
We are not eligible for trials because we cannot be traced well enough, we have to have yearly (if not more often) bone marrow biopsies. So much is different. Hopefully PET can become more of a factor in monitoring nonsecretory, but every doctor says the same thing – you are doing great, we just have to be on it when it comes back ... if it does.
I see amazing things going on in the world of cancer and my hope is ... none of this matters. Maybe, soon, remission is just a shot away??
This study originates from patients at one institution and interesting. It would be useful to know if these findings would occur elsewhere outside of the US e.g. in Europe where tests such PET/CT imaging are not always available so readily and treatment programmes / protocols are more defined and less flexible than in the US.
Also it is generallly reported that 50% or so of MM patients develop renal disease in the course of their illness whether they had it before or not, so would the other 50% not then be like the non secretors in this study- i.e. have longer OS?
The 3% level of significance is not a very high level of significance, at the 5% level one might reject a null hypothesis. So yes the study shows something that needs further assessment. It also means that people with non secretory MM who present with maybe just backache might take longer to be diagnosed. If that is the case then are they not more at risk of fractures which can also affect OS?
Maike, yes I do think that there is a difference in OS between the two groups. As you said, the authors could have highlighted this by other means. Thanks again for writing this interesting article.
Thank you Maike. I do appreciate it so much, as the Beacon Staff makes issues so crisp and clear for non medical educated readers. The graphic works perfect for me. Have a wonderful Easter everybody. Hopefully we will not have use for our own resurrections for a long time to come.
I've developed into a nonsecretor within the last couple of years. I did look over comments above and maybe I missed it. Yes, I agree with Maike that we often get a break with our kidneys. But BMBs can be complicated way to measure. For example, I've had both illia (hips) radiated for pain. Last two BMBs show less than 1% plasma cells in my pelvis; radiation has destroyed a lot of the marrow, along with reduced cellularity (ability for marrow to produce white cells, red cells and platelets). Two years ago, mine was so low (around 16%; should be closer to 50%) that my specialist feared I'd developed MDS. PET scans and MRIs only way to know. Risky if insurance restricts how often they're allowed.
Now my point. I didn't see this mentioned: Nonsecretors don't qualify for 9 out of 10 clinical trails. If we did, might we live even longer? Restricts access to newer, experimental agents. Then again, if a late stage patient picks the wrong trial, that can hasten our demise. So maybe statistically another wash? But I'd still like it to be my choice!
Awesome exchange! Thanks, for taking the time, Maike! By the way, she's my editor and does a fab job; I'm not always easy to keep in line!
I was diagnosed in 2012 and since have become a nonsecretor. What if any are the implications of this study for patients like myself? or are the study results relevant only to patients who were always nonsecretors?