CLL Topics Banner: Therapies, Research and Patient Education for Chronic Lymphocytic Leukemia
CLL Topics Home Navigation Topics Alert Learning Tools About Us Feedback Feedback
Full Menu

CTL Therapy

chinese print

Combining CTL Therapy with a Bispecific Antibody

Date: August 8, 2003

by Chaya Venkat

Is This A Way To Spell C-U-R-E?

spiral watermark

T-cell Therapy with a Bispecific Antibody

Here is an ASH 2001 abstract that talks about both of my favorite subjects, CTL therapy and bispecific antibodies. Part of the problem with the CTL therapy that I have been discussing in previous articles is that each t-cell can kill at the most one b-cell. The actual kill ratio achieved is a lot less than that. So, we have to grow literally billions of CTLs to go in and do the mop-up job. That is also why CTL therapy is typically considered to be effective after more conventional therapies like chemo or monoclonals or combinations like RFC, when the vast majority of tumor cells have already been destroyed. The logic is that the externally trained CTLs can hope to do the job only when they are not outnumbered. The higher the ratio of CTLs compared to b-cells left over as "Minimal Residual Disease", the better the chance of full PCR negative status (dare I say "cure") after the mop-up CTL therapy. Clinicians typically hope for this ratio to be far better than 100 killer CTL cells for every 1 target cancer b-cell.

Now the new bispecific antibody spin, added on top of CTL approach: the experiment described below compared b-cell kill rates at a lowly ratio of 10: 1 CTL versus b-cell ratio. As expected, the kill rate of b-cells at this low ratio was only 20%. Now they added CD19XCD3 bispecific antibody to the mix. This has the effect of cross linking the CTL cells (via the CD3 hook) to the b-cell (via the CD19 hook). Forcing this kind of close proximity between the killer and the target increased the kill rate by more than 4 fold, to 81%.

To my mind, the odds are constantly improving on this approach, day by day: with better ways of growing the patient's own t-cells to be effective CTLs outside the body, better ways of using dendritic cells to be more effective antigen presenting cells and therefore better ways of indoctrinating the t-cells, better ways of culturing and growing the billions of copies needed, and now better ways of getting higher kill rates, more bang for the buck. All of this with minimal toxicity, no worries about graft-versus-host disease and the like since the patient's own cells are used.

Pretty exciting results, to my mind. So all you folks out there, who are approaching end of w&w and looking at first therapy decisions, whether you are looking at frontline Rituxan type approaches, or go the Full Monty for combos like RFC, keep an eye out for Autologous CTL therapy of the types mentioned in my past few articles as a mop-up procedure, after you have completed the necessary rounds of chemo or what ever. The seed t-cells etc will have to be harvested from your blood probably prior to going into chemotherapy, so you need to plan for it ahead of time. The problem is going to be finding locations where you have access to CTL and/or CTL/bispecific antibody type therapy options. I have a hunch these approaches are at least in
the planning stage at most of the CLL consortium centers, but we need to find out more.

So, here is some home work for you all. If you have contacts, if you presently see one of the CLL experts at Consortium centers, ask about CTL therapy. Ask about CD3XCD19 bispecific antibody therapy. Read up about it as much as you can before you ask, so that you can make the most sense of the answer. Last but not least, don't forget to write, and share your information with the rest of us. We are all in this together.

Above all, stay as healthy as you can, as long as you can. There is a dawn coming, I can sense it, and I want you all to be there for the new day.


[1546] Apoptosis in CLL-Cells Induced by Autologous NK T Cells Is Enhanced by a Recombinant Bispecific Tandem Diabody.

Martin Kornacker, Kipriyanov M. Sergey, Little Melvyn, Moldenhauer Gerhard, Edeltraud Weilguni, Markus Herbst, Manfred Hensel, Anthony D. Ho. 

Dep. of Internal Medicine V, University of Heidelberg, Heidelberg, Germany; Affimed Therapeutics, Ladenburg, Germany; Division of Molecular Immunology, German Cancer Research Center, Heidelberg, Germany.

Natural killer T cells (NK T cells) are defined by the coexpression of T- and NK-markers. Two types of NK T cells exist: One type is selected by CD1d, expresses a skewed T-cell receptor repertoire and is found primarily in thymus and liver. A second type is independent of CD1d, has an unbiased T-cell receptor repertoire and is found in spleen and bone marrow. A cell population similar to CD1d independent NK T cells can be expanded from peripheral blood mononuclear cells by the timed addition of g-IFN, IL-2 and OKT-3. In vitro culture leads to a 900 fold expansion of CD3+56+ NK T cells in three weeks. MHC independent cytolysis of  various syngeneic tumor cell lines could be demonstrated in a mouse model (Baker et al., Blood 97(10), 2001). Bispecific antibodies or their recombinant analogues crosslink effector cells with their targets and may enhance their cytotoxicity. We used MACS to separate CD3+ cells for expansion of NK T cell effectors and CD19+ targets from peripheral blood of CLL patients. Because primary CLL cells cannot be labelled sufficiently with 51Cr they are not suitable as targets in a standard 51Cr-release cytotoxicty assay. Therefore, we assessed cytotoxicity by measuring apoptosis in CLL cells. Targets were labelled with the red membrane dye PKH26, incubated with effector cells and analyzed for annexinV- FITC positivity by flow cytometry. After 4 hour incubation of NK T- with CLL cells at an effector to target ratio of 10 : 1, specific apoptosis in CLL cells was 20%. Programmed cell death in target cells could be increased by using a recombinant CD3 x CD19 bispecific tandem diabody generated by functional dimerization of a single chain molecule that contained four antibody variable domains (VH3-VL19-VH- 19-VL3) in an orientation preventing intramolecular pairing. Adding CD19xCD3 bispecific tandem diabody at a concentration of 1 mg/ml to the assay caused specific apoptosis in 81% of CLL cells. These results could be reproduced in primary CLL cells from 2 other patients. In conclusion, NK T cells induce apoptosis in primary autologous CLL cells without antigenic priming. Retargeting of NK T cells with a recombinant bispecific tandem diabody increased target cell death significantly. The fact that these ex vivo activated and expanded effector cells do not cause GVHD in a cross major transplantation model (Baker et al.) holds promise for their use in cellular therapy after allogeneic stem cell transplantation in CLL and lymphoma.

Keywords: CLL\ Antibody\ Killer cells

A Bispecific Antibody Combined with CTL Therapy May Have Potential for a Cure

Rituxan has become the most discussed mAB (monoclonal antibody) among CLL patients. Its low toxicity, tolerable infusion related side effects and ease of administration (the standard protocol is only four weeks long) have made it a popular frontline choice. But perhaps the most attractive consideration in choice of Rituxan frontline therapy is the thought that it does not burn any future therapy bridges. 

Unfortunately, there are serious limitations to Rituxan as single agent therapy in CLL. Let's look at what is less than wonderful, then we can understand why the new bispecific antibody technology is so encouraging.

  1. Not all CLL patients respond to Rituximab. Compared to standard chemotherapy protocols, the overall response rate and is low. Perhaps this has something to do with lower CD20 antigen density, or a substantial amount of "junk" soluble CD20 in the plasma acting as a sink for the Rituxan, wasting the precious drug. 
  2. Other possible reasons for the lower response rates could be insufficient ADCC activity. Some researchers believe that the Fc region of the mAB is not sufficiently antigenic. (Fancy words, but you know that they mean! Remember that Rituxan is "Y" shaped, with the tips of the arms of the "Y" attaching to CD20 markers on B-cells, the tail portion of the "Y" is the part that gets effector cells mobilized). This might suggest that ADCC in Rituximab therapy utilizes mostly NK-cells, and the more potent response of cytotoxic T-cells is just not there.
  3. Complement dependent cell-kill could be reduced due to complement inhibitory proteins such as CD55 and CD59, which are present at higher concentration on CLL cells, compared to, say, follicular lymphoma. Read all about it an excellent (and free) article in Blood Journal.
  4. While Rituximab has an excellent half-life, its bioavailability is less than desirable. (Again, just fancy words meaning Rituxan hangs around in the body long enough before you flush it down the toilet, but it does not get into all the nooks and crannies, like it should). Typically, patients with bulky lymph nodes, swollen spleen or heavily infiltrated bone marrow have poorer responses. Perhaps this is due to the bulky size and high molecular weight of the mAB; perhaps the relatively high avidity of the mAB to its designated target antigen has some bearing on this as well. (Just means the Rituxan molecule gloms on so fast to the first CD20 it sees that a whole bunch of the drug can be all glopped over the surface of the lymph node, and none of it letting go enough to gradually work its way into the interior. What we need is a sort of "Velcro effect", where the Rituxan can both tag and un-tag from the CD20, gradually seeping into the interior of the bulky lymph node, where it can do some good).
  5. Even among responders, the remissions seem to be significantly shorter than in other B-cell malignancies such as Follicular lymphoma, suggesting less efficient B-cell depletion, too many of the bad guys still around, and therefore quicker reconstitution of the clonal population.
  6. Repeat treatment with Rituximab is possible, but only roughly 50%-60% of patients who responded the first time do so the next time around. It appears the resistance is probably not due to loss of CD20 marker. So what gives? Selection process that leaves behind CLL cells with higher levels of complement inhibitory proteins after therapy may explain the observed resistance upon repeat treatment. The bad guys left behind after the first round are a tougher bunch to kill, and when they grow their numbers back up to where you need therapy, they are harder to kill the second time around.
  7. Extremely high overall response rates and even molecular (PCR negative) remissions have been obtained using combination therapies such as RFC and RF. However, these protocols have the usual toxicity and bone marrow suppression associated with the chemotherapy components. The time commitment is also substantial, (protocols are typically six rounds one month apart), especially if one adds in the time needed to recover from neutropenia and the like at the end of the protocol. Also, there is concern that if one does not have a good response to RFC or RF protocols, future treatment with Fludarabine based therapies may become a little iffy. 

I would like to propose a combination of two cutting-edge new technologies that have recently become available. In my opinion, this combination would be synergistic, it would retain all the advantages of modern immunotherapy associated with Rituxan, but would avoid most if not all of the problems cited above. Both of these new technologies have been through pre-clinical testing in animal models, and are just entering phase –1 clinical trials for CLL. Both have shown efficacy in phase –1 trials for other hematological malignancies. Both are currently in Phase -1 trials for CLL. Hold your breath just a little longer folks, the times they are changing fast! 

Micromet Technologies 

This German biotech company has developed a bispecific antibody technology with the catchy name of "BiTE" (Bispecific T Cell Engagers). We have discussed bispecific antibodies above and in the section on Monoclonal Antibodies, if you need a quick refresher. The link: Review of Monoclonal Antibodies in Development.

Unlike the very bulky and heavy "Y" shaped Rituxan molecule, Micromet's MT103 represent the minimalist version of a bispecific antibody. The antibody consists of just the tips of the "Y", strung together with a short peptide chain. So, instead of being a bulky "Y" shaped molecule, MT103 is a straight rod shaped molecule, with the business ends being at the two ends of the rod. And here is the interesting part, one end of the rod has a pincer matched just right to grab CD19 marker. The other end of the rod works for CD3. The shorthand for this structure is "bscCD3XCD19" (bispecific, single chain, targeting CD3 and CD19 markers. This stuff is not so hard to figure out, once you get the hang of it). 

Why target CD19? Like CD20 (targeted by Rituxan), CD19 is a marker expressed by B-cells. Unlike CD20, however, the CD19 marker is present on all developmental stages of B-cells and not just mature B cells. It is not, however, present on stem cells or plasma cells and is not expressed by T-cells, NK cells, stem cells, plasma cells or any other kind of cell line.  In terms of anti-CD19 therapy of the type we were discussing in my article (MT103 from Micromet), the antibody will target pre-B-cells as well as mature B-cells. In comparison to Rituxan therapy, where the target marker CD20 is expressed only on mature B-cells, the depletion of B-cell population by using anti-CD19 is likely to be more profound and deep. Thankfully, the stem-cells are not involved, otherwise CD19 would not be a safe or valid target for attack. And the advantage is that all mature B-cells express it, not just some of them. It (CD 19) is an essential feature of a B-cell being a B-cell, it cannot survive without it. Aha!! Problem number one above is solved. Unlike Rituxan therapy, where low levels of CD20 expression was a concern, no such problems here. And the action is similar to that of Rituxan, limited to only B-cells, since both CD20 and CD19 are expressed only by B-cells. No worries of across-the-board cell kill of all sorts of immune system cell lines, as in the case of Campath, for example. No worries about heavy duty immune suppression leaving you vulnerable to potential opportunistic infections. 

As we described above, when compared to Rituxan, MT103 is much smaller in size. That smaller size and its rod-like shape means it can get into places that "fat lady Rituxan" could not, such as the middle of those large lymph nodes strategically placed at the sides of your neck, guaranteed to make you look like an overweight chipmunk. 

One of the possible downsides is that MT103 does not hang around as long in the body, so we have to learn how to calculate the dosage and frequency of administration take that into account, get maximum bang for the buck. 

So far we have discussed the CD19 end of the rod. How about the other end of the rod, the one with the pincer tip for CD3 marker? Well, this marker is expressed by mature T-cells, and when this marker is tagged, the T-cell becomes a "CTL", a cytotoxic T-lymphocyte. No more worries about inefficient ADCC, and anergic T-cells sitting on the side lines like lazy bums, doing precious little. The targeted B-cell is hooked by the CD19 tip, and the killer T-cell is hooked by the CD3 tip. The two ends of the MT103 rod locks the target and effector cell together in a not-so-loving embrace, a match made in hell. (I rather liked the title of my previous article on this subject, "Fatal attraction of the cancer kind". P.C thought it was a bit too cheesy, but heck, what does he know). 

With ADCC going like gang busters, who cares about complement dependent pathways for cell-kill? CDC becomes less of an issue, as also the level of complement inhibitory proteins such as CD55 and CD59. This also reduces likelihood of developing resistance to multiple rounds of MT103, just in case you happen to need a mop up job a couple of years later. 

With very effective cell-kill locked into place, so to speak, the dosages required are much less than that of Rituxan, to the tune of 100,000 fold less. 

The manufacturers show that the effectiveness of the bscCD3XCD19 depends on the number of T-cells available, but that MT103 does its job so well and efficiently that it needs far fewer T-cells per B-cell that has to be killed. I tend to agree, but all the same, I would like a bit more work on that end of the equation. How about people in the other kind of watch and wait, who have been round the block a few more times than they care to remember? People with potentially depleted numbers or heavily compromised T-cells? Will the number of T-cells be enough to do the job in that case, and will the tagging ("ligation") of the CD3 marker on the T-cells be enough to get them out of their lethargy to become lean and mean killing machines? Me, I like to get extra insurance. This is where the second of the two technologies comes in. But, before I forget, here is the website address of Micromet — do browse around - Micromet's BiTE technology.

(Do read more about B-cells and how they grow and mature. In our Textbooks section you will find a wonderful on-line textbook by Charles A. Janeway, Paul Travers, Mark Walport, Mark Shlomchik. This is up-to-date, comprehensive, beautifully illustrated reference material. Here is the link: Immunobiology 6. Perhaps you will be tempted to click on it. Chapter 7 of Janeway has great stuff on lymphocytes. The particular section titled "Generation of lymphocytes in bone marrow and thymus" has information on CD19 expression. Do take the time to browse, it is well worth your time.)

Xcyte Therapies - Adoptive CTL Therapy

We have reviewed and discussed adoptive CTL therapy before, and we are all watching with interest the outcome of the Phase - 1 clinical trials underway at UCSD, for CLL patients. The manufacturer, Xcyte Therapies, has figured out how to collect T-cells from the patient's own body, then grow them into huge armies of T-cells, ready for re-infusion back into the patient at the correct time. There is a lot of cutting edge technology and smart science behind this breakthrough capability, and we have discussed some of it in T-cell Therapy at UCSD.

(Editor's Note: There used to be a presentation with pretty pictures on Xcyte's own website. Unfortunately, this website is no longer active and — as noted in T-cell Therapy at UCSD — the company itself is now out of business.)

The clinical trial under way now at UCSD (Dr. Kipps) is using this technology as a stand-alone approach. The idea is to see if the vastly larger number of willing and able T-cells can take care of the CLL cells, with no other assist. Possible. But I worry about two aspects of this approach: CLL cells are sneaky little rascals, they do not go about advertising the fact that they are caner cells. In fact, one of the reasons why they got to be a problem in the first place is that the T-cells have a hard time telling them apart form good and normal cells. I worry that this low profile - "low antigenicity" - of the CLL cells may make them escape cell-kill, even if there are a lot more killer T-cells patrolling the neighborhood. My second worry is common to all approaches where we goose the immune system to work at a higher pitch. What if the armies of goose stepping T-cells get out of control, start attacking innocent by-standers? Can we get into autoimmune disease problems, for example? 

Here is where I think combining Xcyte Therapy's armies of patient's own CTLs, and Micromet's bispecific MT103 pays off, big time. No more worries about late stage patients not having enough T-cells, ready to do battle. Heck, we just collect a few, then multiply them like rabbits till we have enough. No more worries if the T-cells are too tired and anergic to do their job, these are "Xcyted" T-cells, rejuvenated and grown away from the confusing propaganda cytokines of the CLL cells. No longer are they brainwashed little wusses, but disciplined armies ready to kill. No more worries about low profile CLL cells slinking around and avoiding detection. With one tip of the MT103 firmly locked on to all the CD19 markers on the CLL cell, it will have a harder time hiding; sort of like a guy with MT103 dreadlocks sticking out all over his head, attending a convention of bald men. And the other tip of the MT103, the one that tags CD3 markers on T-cells, that is ever so attractive to any T-cells in the neighborhood. 

If I am right and there is a high level of synergy between these two technologies working in tandem, then the amounts of the two components used together can be a lot less than each one used separately. I like it any time we can reduce the amount of a drug that needs to be used. Besides reducing chances of toxicity, there is also a little something called dollars per dose. 

How "Real" Is All This? Are We Still Talking Mice?

We are all done with pre-clinical work on cell lines and many unfortunate mice have been put through their paces. Micromet has two brand new papers out, one in Leukemia and the other in the Journal of Immunology. 

Micromet has just announced that they are initiating Phase -1 clinical trials in Europe, for CLL patients. They have been running Phase -1 trials for NHL patients since spring of last year. No published results yet, but the word is that the technology looks good, sufficient to get them a nice partnering deal with MedImmune. Micromet has just signed an agreement with this company to co-develop MT103 in North America, for treatment for B-cell tumors. That means CLL and NHL, watch out Rituxan. Pretty big market, lots of money to be made if things go right. The other thing I like is that Micromet has good patent position, they can protect their investment and therefore it is worth making the effort to nurture and feed this particular cash cow. As we all know, money makes the world go around, orphan drugs that do not have the ability to make money for anyone often get left behind. 

As for the second member of this tag team I would like to assemble, Xcyte Therapies is an American company based in Seattle. Again, these guys also have their intellectual property all nicely sown up, no trespassing signs all around the perimeter. I understand they are getting some kick-ass good results in their phase - 1 clinical trials using Xcyted patient specific T-cells for multiple myeloma. I am also impressed that our own favorite CLL Research Consortium is checking out technology for real live CLL patients in the UCSD clinical trial. Again, no confirmed or official reports of results in this trial, but I am getting good vibrations on this one too. 

Here is where I think we are: both technologies are ready and ripe for the plucking. Both have shown good response in Phase - 1 human trials, and both have the capability of working alone, but I think there is a beautiful synergy about them that is just begging to be explored. In fact, the Xcyte Therapies home page lists combination of their technology with monoclonal antibodies as one of the possible approaches.  How do you spell "CURE?" Here is one possible answer: "MT103 & Xcyte CTL".

Some Questions

Here is a question raised by a member on how the MT103 biscpecifc antibody might work in practice: "... you mention the MT103 locking onto the CD19 of the B-cell first and then attracting the T-cell. In all your extensive reading, have you found any indication that the MT103 would attach first to the CD19 and then to the CD3 cell? If a CD19 cell and a CD3 cell were next to the MT103 would the CD19 end of the antibody attach first or would the antibody just as likely attach to CD3 on the T-cell? (I understand that since there are way more CD19 cells in the body obviously they would be the first marker found by the MT103.) The reason I ask this is that if the antibody will attach to the T-cell first why not combined the "Xcyted" T-cells in vitro and then infuse them back into the patient? I would think that this technology would help to keep the MT103 around in the body as long as the T-cells and avoid the possible downside you mentioned: "One of the possible downsides is that MT103 does not hang around as long in the body" 

Excellent question! 

There is a great deal of fine-tuning required to make these monoclonals work just right, and one of the features that need to be optimized is the "Velcro effect" I discussed. This is way too simple an analogy, but it helps get the picture. In the case of bispecific antibodies, there is the further balancing act of how strongly the CD3 hook binds to the T-cells, versus how strongly the CD19 binds to the CLL cells, the so-called "avidity" of the hooks. If the CD3 hook binds too strongly and never lets go, pre-mixing the MT103 plus Xcyte T-cells outside the body before injection back into the patient means we now have a huge construct, (T-cells are quite large), much bigger than the Rituxan molecule ever was, and it makes the process of percolation into the deep corners of lymph nodes that much harder. That is why it is important for the CD3 and CD19 hooks to latch on and latch off as needed, act like Velcro. 

In fact, in the mice studies reported in the paper I cited, they infected the animals with human lymphoma cancer cells. In one of the experiments, they then mixed human T-cells plus the MT103 and injected the mixture into the animals. The idea is to prove human cancer cells can be killed by human T-cells plus MT103. 

Pharmacokinetics (how long the drug lasts in the body) is different for mice and men. Human T-cells do not last very long in mice, since they are attacked and destroyed by mice immune system. I am sure the researchers now have a better fix on how it works in humans, since they have been running the lymphoma clinical trials for about a year. 

Of course, the unspoken downside possiblity is where the MT103 simply tags a T-cell which, as a result of the antibody attack, promptly undergoes apoptosis without ever finding a B-cell to complete its intended task. To take this one step further, in the case of outnumbered armies, would an infusion of MT103 simply wind up clearing out all the CD3 positive cells in your body without making an appreciable dent in the tumor load — the negative outcome in which you keep the tumor and lose whatever small force of T-cells that you might have available?

Some truly mind-blowing science here, and I am just beginning to understand the simple basics of it. I do worry that my articles are almost comic strip versions of very complex issues. But heck, we  never said we were the experts, just the patients waiting to be cured, right? We need to learn just enough to spot a good trend when  we see it, and stay away from the real dogs.


Editor's Note: From press releases and other documents, it appears Micromet AG is still in business as of February 2007. NCT00274742 is the listing for its phase I safety trial of its bispecific antibody for NHL. It appears to have been recruiting since January 2006.




Disclaimer: The content of this website is intended for information only and is NOT meant to be medical advice. Please be sure to consult and follow the advice of your doctors on all medical matters.

Copyright Notice:

Copyright © 2002-2007 CLL Topics, Inc. All Rights Reserved.

All materials contained on this site are protected by United States copyright law and may not be reproduced, distributed, transmitted, displayed, published or broadcast without the prior written permission of CLL Topics, Inc. You may not alter or remove any trademark, copyright or other notice from copies of the content.

However, you may download and print material from exclusively for your personal, noncommercial use.


courtyard scene



Topic: CTL Therapy

up arrow