A Landmark Clinical Study Shows Improved CLL Survival with DiSC Assay Directed Chemotherapy1



Adapted from: Leukemia (2013) 27, 507–510; doi:10.1038/leu.2012.209

  1. NOTE: In this section and throughout this website, the terms DiSC Assay and TRAC Assay are used interchangeably.  The LRF CLL4 study protocol as originally published refers to the “DiSC assay” whereas data from the second randomization are reported using a “TRAC assay” designation.  The DiSC (differential staining cytotoxicity) Assay was developed by Weisenthal (1983).2-9  A throughput modification of the DiSC assay was later introduced by Andrew Bosanquet and designated TRAC (tumor response to anti-neoplastic compounds) Assay (1996).10  DiSC and TRAC are process variations of the same assay, in which the endpoint, cell survival/cell death, as determined by maintenance or loss of cell membrane integrity, remains identical. (methodology)  

  2. In this landmark study, the first-ever in oncology in which use of a laboratory test was held to the unprecedented standard of improving overall survival, there was a 2.5-fold greater mortality at one year in the group receiving standard, empiric therapy (Protocol arm) compared to the group receiving cytometric profiling-directed therapy (DiSC/TRAC Assay arm). (20% mortality in the Protocol arm versus 8% mortality in the DiSC/TRAC assay arm, p=0.04).

  3. The study is also historic in the respect that no other clinical trial has ever shown a survival advantage for chemotherapy alone in the setting of relapsed CLL.

  4. It is extremely important to note that, in order to show any overall survival advantage at all, the DiSC/TRAC Assay arm was required to overcome three monumental barriers imposed by the study design.  Two of these were largely unavoidable.  The third barrier, omission of an important study endpoint, would not have occurred if the the study were designed today.

Study Design Barrier Number 1:  Off Protocol Treatments

The design of this study made it virtually impossible for an overall survival advantage in either study arm to continue beyond one year.  This is because patients in both study arms who failed treatment (i.e. relapsed or did not respond) became eligible for salvage therapy with off-protocol drugs.  In both study arms, the salvage drugs were selected empirically - without DiSC/TRAC Assay results.  Off-protocol drugs used for salvage included new, targeted agents, such as Velcade© and Rituxan©, which were not available when the study began.       

Study Design Barrier Number 2: Omission of an Important Study Endpoint

The only clinical endpoint of this study was overall Survival (OS).  A more productive endpoint would have been Progression Free Survival (PFS).  A drawback of OS as it was used in this study is the confounding effect of salvage therapy with off-protocol drugs, as discussed in the paragraph above.  Today, an increasing number of new drug approvals are based upon PFS and not upon OS for this very reason.  At the time the study began, PFS was not as commonly used in clinical trials.  Click here to see the trend in the use of PFS from 1975 through 2009.

Study Design Barrier Number 3:  Duplication of Treatments in Both Study Arms

Unlike a drug efficacy study, in which patients are randomized to receive different treatments, many patients in both arms of this study - a study of treatment choice methods - received exactly the same treatments.  It is easy to understand how this could happen:  For CLL patients who fail first line chlorambucil (this includes most subjects in this study), fludarabine, known to be active in this disease setting, is likely to be the empirical treatment of choice.  Therefore, it is not surprising that many patients in the Protocol Arm of this study received the drug.  However, precisely because fludarabine is indeed so often active in CLL and because the DISC Assay is able to detect this activity prior to treatment, many patients in the DiSC/TRAC Assay arm also received fludarabine, based upon their assay results.  Therefore, the design of this study often produced a situation in which outcomes of relapsed CLL patients receiving empirically-selected, second line fludarabine were compared with outcomes of relapsed CLL patients receiving DiSC/TRAC assay-selected, second line fludarabine.  Obviously, this made it even harder to prove a survival advantage in the DiSC assay arm.  

Predictive Accuracy Overcomes Study Design Barriers

  1. Collectively, the barriers outlined above might easily have produced an inconclusive study result, especially since the statistical power of the study did not necessarily lend itself to revealing modest survival differences. 

  2. And yet, in this prospective, randomized study, a clear, statistically-significant overall survival advantage (2.5-fold fewer deaths, p=0.04) was seen at one year among patients whose chemotherapy drugs were selected on the basis of DiSC/TRAC assay results. 

  3. Further, the survival advantage in the DiSC/TRAC Assay arm continued until precisely the point in the study at which off-protocol salvage drugs were introduced in an uncontrolled fashion.  Beyond this point, the possibility of achieving a meaningful survival advantage in either arm of the study was eliminated.   

Study Design

  1. 1.In the LRF CLL4 study (1999-2004), 777 CLL patients were randomized to treatment with fludarabine versus chlorambucil versus fludarabine plus cyclophosphamide.  DiSC/TRAC assays were performed upon blood or bone marrow specimens obtained from 544 of the 777 patients.  However, test results were was NOT used to select therapy for any patient in the first randomization in this study.  Physicians were blinded to DiSC/TRAC assay results.  Result: No statistically significant difference in OS was observed in any of the three, protocol-mandated (i.e. not assay-guided) study arms.  However, patients receiving drugs prospectively found active for them in the DiSC/TRAC assay achieved a six-fold improvement in OS versus patients receiving drugs prospectively found inactive against their own CLL cells.

  2. 2.At disease progression, patients’ tumor cells were re-tested in the DiSC/TRAC assay for sensitivity or resistance to the following agents:

  3. chlorambucil

  4. cyclophosphamide

  5. methyl-prednisolone

  6. prednisone vincristine

  7. doxorubicin

  8. mitoxantrone

  9. cladribine

  10. fludarabine

  11. pentostatin

  12. 3.Patients underwent second randomization (2000-2008) to one of two study arms:

  13. Patients in the DiSC/TRAC Assay arm received personalized treatments which were selected for them on the basis of their DiSC/TRAC assay results.  

  14. Patients in the Protocol arm were treated with drugs selected without benefit of DiSC/TRAC assay results. 

  15. 4.Patients in both study arms were restricted to treatments with the drugs listed above.

  16. 5.Patient mortality was reported at 3, 6, and 12 months.  Thereafter, it was reported only annually.

  17. 6.Upon further disease progression (i.e. first progression following the second randomization), patients in both study arms became eligible to receive agents which were not available at the time the study began.  Because these new agents were not available when the study began, they were not tested in the DiSC/TRAC assay.  Therefore, at second progression, all patients received drugs which were selected by the empirical method.  

Discussion

Many of the agents used in this study for salvage following disease progression are known to yield clinical benefit in some percentage of unselected patients.  Examples of these agents include:

  1. Rituximab (now an extremely important drug in CLL treatment)

  2. Bendamustine

  3. Lenalidomide

  4. Bortezomib

  5. and others

Since, upon disease progression, DiSC/TRAC Assay patients and Protocol patients received off-protocol drugs, selected empirically, it was not possible to show a long term survival advantage in either arm of the study.  However, prior to the introduction of the statistically-confounding variable of off-protocol drugs, the DiSC/TRAC Assay was shown to be an effective tool that produced superior survival at one year compared with empirical, physician’s choice drug regimens.  Arguably, the DiSC/TRAC assay succeeded to the full extent that was possible within the context of the study design.  

The following, hypothetical example (which almost certainly occurred more than once in the course of this study) is offered to facilitate discussion:  Let us assume that a patient in the empirical (Protocol) arm relapsed at six months and achieved a salvage remission through treatment with Rituxan©, or other new drug, selected empirically.  Meanwhile, a patient in the DiSC/TRAC assay arm continued in remission on study protocol drugs.  Owing to the study design, the relapse event in the empirical arm was not captured as part of the study data.  As a consequence several questions remain unanswered:

  1. If the new, effective salvage drugs had not been available to patients in both study arms would the overall survival advantage seen in the DISC/TRAC assay arm at one year have been greater?  Would the OS advantage in the DiSC/TRAC assay arm have continued beyond one year?  If so, by how long and at what magnitude?

  2. If the salvage drugs had been available for testing in the DiSC assay and if DiSC assay results had been used to match each patient with the salvage drug or drugs most likely to provide benefit, what would have been the effect upon OS in that cohort?  Moreover, what if the DiSC assay had been used to guide therapy upon subsequent relapses?   

  3. If Progression-free survival had been captured in the study data, would one of the study arms have emerged as superior using the PFS endpoint?  If so, how much PFS advantage and for how many months or years??

  4. If DiSC assay results had been used to guide first line chemotherapy rather than, as was the case in this study, being reserved for first relapse, how would that have effected PFS and OS?

Unfortunately, the data produced by this study do not shed light on these additional questions.  However, based upon these results, in which DiSC/TRAC Assay use improved the one year survival rate, taken in context with results of dozens of other uniformly positive prospective and retrospective DiSC Assay studies, it is reasonable to speculate that CLL survival could be further improved through DiSC Assay-guided drug selection.

Other Important Findings of this Study

In addition to the unprecedented finding of improved CLL survival with DiSC/TRAC Assay-directed chemotherapy, several other findings of the LRF CLL4 study are highly worthy of note:

  1. There was a significant correlation between in vitro (DiSC/TRAC Assay) and in vivo sensitivity to the regimen used in patient treatment, which translated into longer survival for the patients treated with drug(s) showing DiSC Assay sensitivity.  Link to Kaplan Meier Graph.

  2. Cross resistance was not observed in vitro between fludarabine and pentostatin, suggesting that pentostatin could be a salvage drug in fludarabine-resistant patients.

  3. The DiSC/TRAC Assay was an significant predictor of survival at diagnosis, independent of 17p and 11q deletion and IGHC gene mutation status.

  4. This is the first randomized clinical trial in which a prognostic marker used in oncology was validated under “real world” laboratory conditions.  Validation of other oncology markers (e.g. ER, PR, Her2, KRAS, EGFR, etc.) has occurred under highly artificial conditions.  Click here for a discussion on that topic.

Authors’ Conclusions

  1. “In summary, the [DiSC] TRAC assay may be of value in practice, particularly in untreated patients who are unfit for intensive regimens and for whom treatment with chlorambucil or fludarabine is planned. The use of these agents in patients with in vitro drug resistance predicts nonresponse. In the relapse setting, short-term survival may be  improved, while further studies including new agents (for example, bendamustine, lenalidomide) in the panel of drugs tested may give additional information on the clinical value of testing.  At diagnosis, the [DiSC] TRAC assay provided prognostic information additional to other known factors.”


References

  1. 1.Matutes E, Bosanquet AG, Wade R, et al, The use of individualized tumor response testing in treatment selection: second randomization results from the LRF CLL4 trial and the predictive value of the test at trial entry. Leukemia (2013) 27, 507–510; doi:10.1038/leu.2012.209.  Link to full text.

  2. 2.Weisenthal LM, Marsden JA, Dill PL, et al. A novel dye exclusion method for testing in vitro chemosensitivity of human tumors. Cancer Res 1983; 43:749–757.

  3. 3.Weisenthal LM, Shoemaker RH, Marsden JA, et al. In vitro chemosensitivity assay based on the concept of total tumor cell kill. Recent Results Cancer Res 1984; 94:161–173.

  4. 4.Weisenthal LM, Lippman ME. Clonogenic and nonclonogenic in vitro chemosensitivity assays. Cancer Treat Rep 1985; 69:615–632.

  5. 5.Weisenthal LM, Dill PL, Finklestein JZ, Duarte TE, Baker JA, Moran EM. Laboratory detection of primary and acquired drug resistance in human lymphatic neoplasms. Cancer Treat Rep. 1986 Nov;70(11):1283-95. [Note: This is a seminal paper. Link to full text.) 

  6. 6.Weisenthal LM. Antineoplastic drug screening belongs in the laboratory, not in the clinic (editorial). J Natl Cancer Inst 1992; 84:466–469.

  7. 7.Weisenthal LM. Cell culture assays for hematologic neoplasms based on the concept of total tumor cell kill. In: Kaspers GJL, Pieters R, Twentyman PR, et al., eds. Drug Resistance in Leukemia and Lymphoma: The Clinical Value of Laboratory Studies. Chur, Switzerland: Harwood Academic Publishers, 1993:415–432.

  8. 8.Bosanquet AG, Burlton AR, Bell PB, Harris AL. Ex vivo cytotoxic drug evaluation by DiSC assay to expedite identification of clinical targets: results with 8-chloro-cAMP. Br J Cancer. 1997;76:511–518. [PMC free article] [PubMed]

  9. 9.Bosanquet AG, Johnson SA, Richards SM. Prognosis for fludarabine therapy of chronic lymphocytic leukaemia based on ex vivo drug response by DiSC assay. Br J Haematol. 1999;106:71–77. [PubMed]

  10. 10.Bosanquet AG, Richards SM, Wade R, Else M, Matutes E, Dyer MJS, et al. Drug cross-resistance and therapy-induced resistance in chronic lymphocytic leukaemia by an enhanced method of individualised tumour response testing. Br J Haematol. 2009;146:384–395. [PubMed]

  11. 11.Bosanquet AG, Durant J. Ex vivo drug sensitivity test by TRAC assay (Tumour Response to Anti-neoplastic Compounds assay): a fourth-generation test based on the DiSC assay. Br J Haematol 2005, 129(suppl 1):61.


  12. Back to Healthcare Professionals