Understanding lab reports
My choice of topics to write about is often determined by the emails I get from our members. Corresponding with patients one-on-one is a vital feedback. For geeks like me it is all too easy to get side-tracked by some bit of “cool science” that has little impact on daily lives of our members. Your emails bring my attention back into focus. Keep ‘em coming folks!
I received a letter from “Martin” trying to sort out the meaning of flow cytometry results and pcr testing. With his permission, I attach a part of his email below:
I was diagnosed with CLL in May of XXXX. A FISH test revealed 11q deletion. Since December of the same year I have been a subject in a drug trial in Buffalo, NY involving REVLIMID. The protocol involves Flow Cytometry and PCR tests of both peripheral blood and bone marrow. Although such tests in June of last year proved negative, i.e. no detectable disease found, a second round of tests in October confirmed the negative pcr results of June, but the Flow Cytometry test showed a small cluster of unusual cells, possibly, but not necessarily, related to CLL. The researcher, Dr. Chanan-Khan, felt that it would be better to err on the side of caution and he has decided to continue with REVLIMID for a few months longer before ordering another set of tests.
Although I think I have a basic understanding of what the Flow Cytometry test does and what its results indicate, my understanding of the pcr test is extremely limited. Can you help? What exactly do the pcr as opposed to the Flow Cytometry test results indicate and what exactly does a negative pcr result (minimal detectable disease) indicate?
In addition to addressing flow cytometry and pcr testing that this member asked about, I thought I would throw in FISH testing as well and try to put these three very important tests into perspective for our members.
This is a very powerful tool that has a wide range of applications in hematology. But for our purposes the most important use of flow cytometry is in the confirmation of CLL diagnosis.
As you should know by now, mere increase in your white blood count does not (repeat, not) confirm CLL. Heck, WBC increases in totally cancer free people, if they happen to have an infection of some sort. That is exactly how things are supposed to happen, noting to get excited about, it is not an automatic indication of CLL.
But assuming the elevated WBC is indeed due to cancer, there are also several other lymphomas and leukemias out there that are “kissing cousins” of CLL that are also “lymphoproliferative” – fancy word meaning increased lymphocyte count. Some of these other cancers are much more aggressive than CLL (a good example is mantle cell lymphoma, MCL) and they would require different therapy strategy. Getting a rock solid diagnosis is the first and most important agenda item on newly diagnosed patients. That is where flow cytometry testing (sometimes abbreviated to just “flow“) comes into its own.
All cells have specific markers on their surface. These are classified into “Cluster Designations” or CD markers. Healthy normal B-cells have a characteristic CD marker profiles. Think of this as a fingerprint of sorts, if you will. Cancerous cells have whacked out CD marker profiles, different from healthy cells. By careful examination of the surface of a group of cancer cells it is possible to get a description of the specific CD marker profile for that group. A fancy jargon word for the CD marker profile is “phenotype”. Once pathologists have defined the phenotype of the cells they can (1) judge whether the cells are cancerous to begin with, and (2) decide which particular blood cancer it is.
Below is a table of the CD marker profiles of some of the more common leukemias and lymphomas. You will notice CLL and SLL share a common phenotype. That is why they are classified as the same thing, even though the physical symptoms and presentation can be quite different sometimes.
The CD markers to look for in CLL (or SLL) are CD5, CD19, CD20 and CD23. On healthy B-cells, CD5 marker is only rarely seen – it is much more common on T-cells and healthy mature T-cells express CD5 marker in great abundance. Seeing CD5 marker on B-cells is the first indication something is not quite right.
On typical CLL cells there are many thousands of copies of CD5, CD19 and CD23 markers per each cell, but the CD20 marker is present only sparingly – in other words, CD20 is expressed only weakly in CLL cells. This is an important distinguishing difference between CLL and say, follicular lymphoma or mantle cell lymphoma. These lymphomas have much higher number of CD20 markers per cell, one of the reasons why Rituxan (as an anti-CD20 monoclonal) works so much better for them.
Not only is it important to know what CD markers are seen on the cancer cells, it is also important to know which ones are absent – an important clue – similar to the dog that did not bark in a Sherlock Holmes mystery. For example, FMC-7 marker is usually not seen on CLL cells, or only very dimly so. CD23 is absent (most often) on MCL cells, but it is bright (strongly expressed) on CLL cells.
Putting it all together, flow cytometry looks at the surface of cells and can tell which CD markers are present, whether they are present in great numbers (“bright” or “strong” expression) or only in small numbers (“dim” or “weak” expression), rule out CD markers that are absent, and by this process help confirm (or rule out) specific B-cell cancers.
Your results are only as reliable as the testing lab where test is performed.
Getting a flow cytometry analysis done at a reputable lab is an essential part of nailing your diagnosis. Sometimes the overlap between the CD profiles between different cancers such as CLL and MCL can make it difficult. If for example there is slightly weaker expression of CD23, or stronger than expected expression of CD20, we are looking at a grey area and it is hard to make the call between CLL and MCL with any degree of confidence. Other tests are then needed to make the distinction clear. It may be necessary to make a trip to one of the expert centers to sort it all out.
Besides its use at confirming initial diagnosis, flow cytometry is also done if there is reason to suspect there may have been a transformation of the original CLL into something else (Richter’s transformation, a very aggressive lymphoma, is seen a small percentage of patients originally diagnosed with just garden variety CLL), or if there is reason to suspect there may be a second blood cancer of some sort (such as an incipient myeloid cancer). Small but non-trivial percentage of patients may have AML (or some other myeloid cancer) in addition to CLL, and it has been suggested that this is a potential risk in patients going through therapy that involves combination of alkylating agents (cyclophosphamide, chlorambucil) and purine analogs (fludarabine, pentostatin).
As we have been discussing, flow cytometry looks at the surface of cancer cells to see what kind of a beast we are looking at. What happens if the blood sample has only a very small number of cancer cells floating around in a sea of normal cells? This dilution effect makes it very hard for flow cytometry to detect enough cancer cells in the first place and therefore the results can be inconclusive.
In other words, if your white blood count (WBC) is very low, almost in normal range, either because you are newly diagnosed and there are just a few cancer cells floating around, or you have just been through therapy and most if not all of the cancer cells in your blood have been killed, flow cytometry may not give results worth a damn.
SLL patients who have most of their cancer cells hiding in swollen lymph nodes and only a few of them showing up in open blood circulation may need to have flow cytometry done on biopsy samples from their lymph nodes. Sometimes your doctor may ask for flow cytometry done on the stuff in your bone marrow, to get a better fix on what is hiding there. This means getting a bone marrow biopsy and the “aspirate” (sort of mushy gel-like stuff sucked out during the procedure – yeah, I know it sounds both gross and painful) sent for flow cytometry analysis.
I came across a couple of links that give a much more detailed background on the subject of flow cytometry. The first is a review article in “Clinical Chemistry” that describes all the different things one can do with this powerful tool. The second is an on-line tutorial that does a very good job of explaining how a flow cytometer works and the instrumentation used. The graphics are really cool. These references are for those among you who would like to learn more about this technology.
For starters, don’t get confused by “PCR” – an acronym for chemo-immunotherapy combo that uses pentostatin, cyclophosphamide and Rituxan that we just discussed in an earlier article on this site – and pcr testing which refers to very sensitive test that looks for minute traces of cancer cells in your body. I have plagiarized myself below – much of the material comes from an article I published on CLL Topics website back in 2003.
Not long ago, during the early heyday of Chlorambucil as the lead drug for CLL, if your overt symptoms subsided, and the lymph nodes subsided to the point where you no longer looked like a chipmunk, after a physical exam you were declared to be in “complete remission” by your doctors. There were no automated machines to do dirt cheap blood counts and I doubt anyone kept track of WBC except when something was clearly wrong. Later, the definition of a complete response was upgraded to mean “normalized” blood counts, no palpable lymph nodes or spleen, etc., but it was OK to have a few nodules in the bone marrow.
Flow cytometry was the next step, where peripheral blood and/or bone marrow aspirate could be examined for the tell-tale CD markers of CLL on the surface of cells. This was particularly effective as flow cytometry improved and was able to look for CLL cells defined by more than just one marker. Now testing labs routinely look for the classic CD5 and CD19 and CD20 and CD23 marker combination profile of CLL cells – the classic fingerprint of CLL we discussed above.
The next stage is where we are looking for the proverbial needle in a haystack to judge the level of CLL cells left behind after therapy. It is hard to find one needle in a large haystack. But if the needle was somehow encouraged to duplicate itself, so that there are a thousand copies made of each needle originally present in the haystack, it would be an easy matter to count all the needles, plus or minus a few missed in the count, since these omissions are too few to change the count by much. That is precisely what pcr (polymerase chain reaction) does.
Bean counting at the cellular level
Prior to pcr technology, our detection limits were no better than, say, 1 in a 1,000. In other words, if the blood or biopsy sample examined had 10,000 cells, we could detect the CLL only if there were more than 10 CLL cells present (10 out of 10,000 is the same as 1 out of 1,000). If you happened to have only 5 CLL cells in the 10,000 cell sample examined, you would get a clean report, no CLL cells detected. Aha! The report does not say there were no CLL cells, just that none were detected.
With pcr technology, a small and carefully chosen snippet of protein that is a good representation of your particular type of CLL cells is copied over and over again, amplifying the number of units of this marker in the sample. In other words, the needle in the haystack is multiplied many times over. You can see this makes it much easier to measure the level of residual disease. In fact, with pcr technology, our detection limits are now about 1 in 100,000 cells. Compared to the previous example, pcr technology has improved our detection ability by a hundred fold. People who got a clean blood report from older technology may or may not get a pass using the more sensitive pcr method. Bottom line, if a well conducted pcr test says you are negative, it means you have less than 1 CLL cell in 100,000 normal cells.
The debate continues on whether to use blood or bone marrow aspirate as the sample on which to base residual disease evaluation. Blood samples are much easier to get. But some researchers feel that in the case of CLL, the last traces of the disease are more likely to linger and hide in the bone marrow, and that is where one should look for it. Consensus is also less than 100% on what particular snippet should be chosen to amplify when using the pcr test. Details, details. One more reason why I would put greater emphasis on getting pcr testing done at a reputable lab or, better still, at one of the expert centers.
It is important to remember that a pcr negative response does not guarantee that there are absolutely no CLL cells in your body, none, zip, nada. It just means that if there are any at all, they are too few to be measured by even this our most sensitive testing procedure. A new technique may be invented next year, one that ups the ante on pcr technology. Then the new standard of complete remission will be changed once more.
The diagram above gives a good way of understanding the changing standards and how different drugs meet the challenges of getting “complete remissions”. As you can see, old fashioned drugs such as chlorambucil went only so far. Purine analogs such as fludarabine went further, at the nadir (bottom) the number of CLL cells left in the body are far fewer – about ten times fewer than with good old chlorambucil (“Leukeran”). Modern therapy combinations such as FCR go deeper than single agent fludarabine – as much as a hundred times fewer CLL cells are left over after patients are done with therapy.
“MRD” negative responses (no trace of minimum residual disease) using our most sensitive and sophisticated pcr testing methods are much to be desired. True, some of these MRD negative patients can relapse and become MRD positive, soon to be followed by full scale relapse. But a very significant percentage of patients can stay MRD negative for years. Some may bounce back and forth between MRD positive and negative states. The thing to remember is that really cleaning out the residual CLL population as much as we can is likely to give remissions that last a lot longer. Without pcr testing we would have no way of documenting the level of clean-out obtained, or keeping track of MRD status over time.
While Flow Cytometry looks at the surface of CLL cells and pcr testing counts the number of CLL cells left over in your body, FISH test looks deep inside the CLL cells, looking to see what makes your CLL tick. As we know by now, not everyone has the same type of CLL. Some lucky patients have very slow and smoldering version while others can have very aggressive variety. CLL Topics invented the “Bucket” classification to simplify matters, make it a bit easier for patients to figure out which bucket they belong to.
FISH test is perhaps the most important prognostic test to get done, both because of its ready accessibility, lower cost and dependability. The IgVH gene mutation test is harder to do, more expensive, and done only at a few testing labs. ZAP-70 testing is still a work in progress and right now I would not hang my hat on this indicator. That leaves FISH test as the most dependable prognostics test.
FISH looks deep inside the cell, right down inside the nucleus of the cell, looking for chromosomal defects in the cell’s DNA. These chromosomal defects are at the heart of every cancer. Chromosomal defects are what make cancer cells mutant cells that do not obey the normal rules followed by all other law-abiding healthy cells. The fact that you have been diagnosed with cancer means that there is one or more chromosomal defect(s) in your cancer cells. This is a fact of life. There are no ‘ifs’ or ‘buts’ about this.
Each type of cancer has its own typical set of chromosomal defects that are characteristic of that particular cancer. And that is true for CLL as well. We do not yet know all of the chromosomal defects that can lead to the development of CLL. We know some of the more common ones. As research progresses, we will no doubt discover other defects, learn how to test for them, learn their prognostic significance.
Right now, we know that the majority of CLL patients have defects in their 13th, 12th, 11th and/or 17th chromosomes. A given patient can have one or more of these defects or he may have a defect that we still know nothing about. More recently, defects in chromosome 6 have been identified in CLL patients. But this is still a relatively new finding and most commercial labs do not test for it.
The generic CLL FISH panel therefore often looks for only these four chromosomal defects: 13q, 12 trisomy, 11q and 17p. If none of the above four defects are detected, the lab reports this as a “normal” karyotype, but as you can see that is a not an accurate description of the actual state of affairs. It just means the patient in question has one or more chromosomal defects that we do not know about and therefore have not tested for. A more accurate description of a CLL patient who does not have any of these 4 defects would be “none of the above” or “unknown chromosomal defect or defects”. We lump all the CLL patients who have “unknown” into one group, and call them “normal”. Since your have the big “C”, it goes with the territory that your CLL cells have one or more chromosomal defects.
The table above is abstracted from one of the classic papers in CLL, published in the NEJM. It has withstood the test of time and I refer to it frequently. As you can see, patients with the most dangerous FISH abnormalities (17p or 11q deletions) are already at higher Rai stage when diagnosed, more likely to have swollen spleens and swollen abdominal nodes, more likely to have “B-symptoms” (frequent infections, unexplained weight loss, drenching night sweats, anemia, easy bruising because of low platelets etc) and more likely to require therapy soon after diagnosis. On the other hand lucky folks with the least dangerous FISH abnormality do better on all these fronts. Referring to our “Bucket” classification, 17p or 11q deletion puts patients in high risk Bucket C, while 13q deletion puts them in a more friendly low risk Bucket A category. 12 Trisomy and “Normal” FISH suggests an in-between “Bucket B”.
Another question that has been raised is the risk of clonal evolution. This refers to gradual addition of new defects to the one you may have had to begin with. Clonal evolution is a fact of life with CLL. Some patients have a more unstable variety of CLL; their CLL cells are more likely to pick up new defects as the disease progresses. For these patients it is often a slippery slope that seems to pick up more dangerous defects over time. Many of the standard chemotherapy drugs are also thought to accelerate the process of clonal evolution. Fludarabine, for example, has been cited for selecting for 17p deletion, the most dangerous of the FISH defects.
But even without the help of chemotherapy, clonal evolution can happen over time. CLL cells are genetically unstable, a classic feature of most cancer cells. Recently, it has been shown that clonal evolution happens more frequently in patients with the IgVH unmutated variety of CLL, and less frequently in patients with mutated IgVH. This is one of the reasons why mutated IgVH is an indicator of better prognosis, since it seems to give some degree of protection from clonal evolution to more dangerous chromosomal defects. Remember though, we are talking statistics here. A given individual patient with the good mutated IgVH can still have clonal evolution happen, just that it is not as likely as in the case of someone who has the unmutated IgVH variety. Clonal evolution is one reason to consider getting periodic FISH testing done, to see if you have become the ‘proud’ owner of a new chromosomal defect since the last time you looked.
How should we advice “Martin”?
First, congratulations are in order since Martin is still pcr negative. This confirms that his CLL is still below our most sensitve detection limits and he can therefore hope to get a long & durable remission on that front.
But what is the deal with that comment “Flow Cytometry test showed a small cluster of unusual cells, possibly, but not necessarily, related to CLL”? This sounds like flow cytometry identified a small collection of cells with a fingerprint of CD markers that was not the classic pattern one would expect from CLL.
It could be nothing, it could be something. What to do? For starters, I think it is a good idea to ask the researcher what exactly they think they identified. If I were to bet, I would expect this is MGUS (monoclonal gammopathy of unknown significance), since a significant portion of CLL patients have small colonies plasma cells that have wandered off of the healthy path. We discussed MGUS in a recent Topics Alert and you can get a lot more information about it there. Bottom line, what to do about this small colony of aberrant cells identified by flow cytometry depends on what exactly it is, what exactly is meant by “a small cluster”. Devil is in the details. It seems the Martin’s physician decided the best way to handle it is to continue with Revlimid therapy.