EditorialPublication Exclusive

Liquid biopsies offer ‘transformative’ potential for detection of aggressive lymphomas

This issue of HemOnc Today includes an article that describes a study from the NCI that investigated the use of circulating tumor DNA as a tumor marker for patients with aggressive lymphoma.

The publication of this study in The Lancet Oncology coincided with the online publication of a similar study in Blood that described the potential use of circulating tumor DNA (ctDNA) during and after treatment in patients with diffuse large B-cell lymphoma (DLBCL).

Although there are some differences between these two studies, the conclusions are broadly similar:

  • Immunoglobulin gene sequences are highly specific clonotypic markers that can be detected in the plasma of most patients with DLBCL;
  • Their levels correlate with tumor burden;
  • Increasing levels in the plasma predict clinical relapse with varying lead times; and
  • This noninvasive technique has the potential to replace routine imaging and other clinical tests in the posttreatment surveillance setting.
  • A wake-up call

    John Sweetenham, MD, PhD

    John Sweetenham

    Reading these papers was something of a wake-up call for me.

    As a physician with a primary interest in lymphoid malignancies, I have largely missed out on the burgeoning literature on circulating tumor cells and ctDNA. This seemed to be predominantly the domain of solid tumor oncology and not high on my list of priority reading.

    Clearly, that needs to change. I need to get caught up on the literature on ctDNA to understand its potential and its limitations because it looks as though this technology could have a major impact on how we manage our patients.

    The application of ctDNA may go well beyond its use as a tumor marker for response and relapse and extend to a “liquid biopsy” method for monitoring tumor resistance and genetic evolution during the course of the disease. Further, it’s likely that measurement of ctDNA will have clinical utility across many tumor types and may eventually replace some of the methods we use for measurement of minimal residual disease in hematologic malignancies.

    I don’t pretend to understand the methodologic and technical challenges associated with detection of ctDNA. I do understand, however, that these issues are substantial, partly because ctDNA is present in the plasma at very low levels and is typically composed of small fragments of DNA, on a background of relatively large amounts of non-tumor circulating DNA. On the plus side, the tumor DNA carries the same genetic mutations that exist in the parent tumor, and as specific mutations arise in the tumor during therapy, these are detected in the ctDNA.

    The current literature suggests that ctDNA is detectable in a range of different malignancies. For example, one recent study detected ctDNA in more than 75% of patients with advanced pancreatic, ovarian, colorectal, bladder, breast, gastroesophageal, hepatocellular, and head and neck cancers, as well as melanoma. Further, ctDNA was detectable in 50% to 75% of patients with limited-stage colorectal, gastroesophageal, pancreatic and breast cancers. By contrast, the rate of detection of ctDNA was substantially lower in cancers of the kidney, prostate, thyroid or brain.

    The studies in aggressive lymphoma report detection rates for clonotypic ctDNA in excess of 80%, and it was found in patients with advanced and limited-stage disease.

    Transformative potential

    The potential clinical utility of this test could be transformative.

    The early studies in lymphoma have demonstrated a relatively high detection rate of ctDNA and its correlation, at least in one study, with tumor burden detected by imaging. Levels of ctDNA have been shown to closely follow response and relapse and — in many cases — increasing levels of ctDNA have coincided with or preceded clinical relapse (in some cases, with a lead time of several months).

    Whether close monitoring of response and early detection of resistance or relapse during treatment will be beneficial in lymphomas remains to be seen. The impact of early detection of relapse will only be as good as the second-line treatments that are available.

    Where this technique could have a major impact is in posttreatment surveillance. Several recent studies have questioned the utility of routine surveillance imaging in aggressive lymphomas. Exposure to harmful radiation and high false-positive rates — especially with functional imaging — shift the risk–benefit balance away from routine surveillance. A noninvasive approach such as ctDNA could reset that balance and provide a reliable test in the posttreatment setting.

    For example, early data in colorectal and breast cancers suggest that initial levels of ctDNA are prognostic for OS. Studies also have shown that when primary diagnostic biopsies are not available, ctDNA can be used for detection of KRAS mutations with a sensitivity of more than 80%. Further, detection of KRAS and NRAS mutations in the ctDNA of patients with colorectal cancer has been shown to correlate with resistance to EGFR blockade, suggesting that this “liquid biopsy” technique could be used to direct therapy and monitor resistance. Emerging somatic mutations could be detected, which may help clinicians determine their choice of next therapy.

    Of course, despite the apparent promise of this technique, there are many unknowns and it likely will be several years before liquid biopsies are routine. The applicability of this technique is currently limited to research laboratories at academic institutions, and its utility in the real world of community-based oncology is unknown. Current data suggest that for lymphoma, fresh frozen tissue is the preferred source for determining the tumor-specific clonotypic sequence, although the investigators report that when sufficient tissue was available, they were able to get a sequence from formalin-fixed, paraffin-embedded tissue in most patients.

    In their real-world validation, a usable sequence was found in slightly more than 70% of patients. Although there are diseases in which monitoring of minimal residual disease has had a major impact on treatment and outcome — chronic myeloid leukemia and acute lymphoblastic leukemia are the most notable examples — whether the same would be true across the spectrum of solid and hematologic malignancies is unknown and will require prospective evaluation.

    Whether the additional lead time provided by ctDNA detection of relapse will translate into improved outcomes for patients is not clear, but if the preliminary data for surveillance in lymphoma are confirmed in later studies, ctDNA certainly could replace or complement some of our existing surveillance tests, and patients will be able to avoid the anxiety, discomfort and cost of unnecessary biopsies or other tests that are the consequence of false-positive results.

    As a recent convert to the concept of liquid biopsies, I am excited by the potential for this technique and look forward to new data on its applicability across multiple tumors and in multiple health care settings. Translating the currently available tests into an affordable, reliable clinical investigation will be challenging and likely will take several years, but the prospect of a reliable blood test for following some cancer types — including aggressive lymphomas — is starting to look better.

    References:

    Bettegowda C, et al. Sci Transl Med. 2014;doi:10.1126/scitranslmed.3007094.

    Dawson SJ, et al. N Engl J Med. 2013;doi:10.1056/NEJMoa1213261.

    Kurtz DM, et al. Blood. 2015;pii:blood-2015-03-635169.

    Murtaza M, et al. Nature. 2013;doi:10.1038/nature12065.

    Roschewski M, et al. Lancet Oncol. 2015:doi:10.1016/S1470-2045(15)70106-3.

    For more information:

    John Sweetenham, MD, is HemOnc Today’s Chief Medical Editor for Hematology. He also is senior director of clinical affairs and executive medical director at Huntsman Cancer Institute at the University of Utah. He can be reached at john.sweetenham@hci.utah.edu.

    Disclosure: Sweetenham reports no relevant financial disclosures.

    This issue of HemOnc Today includes an article that describes a study from the NCI that investigated the use of circulating tumor DNA as a tumor marker for patients with aggressive lymphoma.

    The publication of this study in The Lancet Oncology coincided with the online publication of a similar study in Blood that described the potential use of circulating tumor DNA (ctDNA) during and after treatment in patients with diffuse large B-cell lymphoma (DLBCL).

    Although there are some differences between these two studies, the conclusions are broadly similar:

  • Immunoglobulin gene sequences are highly specific clonotypic markers that can be detected in the plasma of most patients with DLBCL;
  • Their levels correlate with tumor burden;
  • Increasing levels in the plasma predict clinical relapse with varying lead times; and
  • This noninvasive technique has the potential to replace routine imaging and other clinical tests in the posttreatment surveillance setting.
  • A wake-up call

    John Sweetenham, MD, PhD

    John Sweetenham

    Reading these papers was something of a wake-up call for me.

    As a physician with a primary interest in lymphoid malignancies, I have largely missed out on the burgeoning literature on circulating tumor cells and ctDNA. This seemed to be predominantly the domain of solid tumor oncology and not high on my list of priority reading.

    Clearly, that needs to change. I need to get caught up on the literature on ctDNA to understand its potential and its limitations because it looks as though this technology could have a major impact on how we manage our patients.

    The application of ctDNA may go well beyond its use as a tumor marker for response and relapse and extend to a “liquid biopsy” method for monitoring tumor resistance and genetic evolution during the course of the disease. Further, it’s likely that measurement of ctDNA will have clinical utility across many tumor types and may eventually replace some of the methods we use for measurement of minimal residual disease in hematologic malignancies.

    I don’t pretend to understand the methodologic and technical challenges associated with detection of ctDNA. I do understand, however, that these issues are substantial, partly because ctDNA is present in the plasma at very low levels and is typically composed of small fragments of DNA, on a background of relatively large amounts of non-tumor circulating DNA. On the plus side, the tumor DNA carries the same genetic mutations that exist in the parent tumor, and as specific mutations arise in the tumor during therapy, these are detected in the ctDNA.

    The current literature suggests that ctDNA is detectable in a range of different malignancies. For example, one recent study detected ctDNA in more than 75% of patients with advanced pancreatic, ovarian, colorectal, bladder, breast, gastroesophageal, hepatocellular, and head and neck cancers, as well as melanoma. Further, ctDNA was detectable in 50% to 75% of patients with limited-stage colorectal, gastroesophageal, pancreatic and breast cancers. By contrast, the rate of detection of ctDNA was substantially lower in cancers of the kidney, prostate, thyroid or brain.

    The studies in aggressive lymphoma report detection rates for clonotypic ctDNA in excess of 80%, and it was found in patients with advanced and limited-stage disease.

    Transformative potential

    The potential clinical utility of this test could be transformative.

    The early studies in lymphoma have demonstrated a relatively high detection rate of ctDNA and its correlation, at least in one study, with tumor burden detected by imaging. Levels of ctDNA have been shown to closely follow response and relapse and — in many cases — increasing levels of ctDNA have coincided with or preceded clinical relapse (in some cases, with a lead time of several months).

    PAGE BREAK

    Whether close monitoring of response and early detection of resistance or relapse during treatment will be beneficial in lymphomas remains to be seen. The impact of early detection of relapse will only be as good as the second-line treatments that are available.

    Where this technique could have a major impact is in posttreatment surveillance. Several recent studies have questioned the utility of routine surveillance imaging in aggressive lymphomas. Exposure to harmful radiation and high false-positive rates — especially with functional imaging — shift the risk–benefit balance away from routine surveillance. A noninvasive approach such as ctDNA could reset that balance and provide a reliable test in the posttreatment setting.

    For example, early data in colorectal and breast cancers suggest that initial levels of ctDNA are prognostic for OS. Studies also have shown that when primary diagnostic biopsies are not available, ctDNA can be used for detection of KRAS mutations with a sensitivity of more than 80%. Further, detection of KRAS and NRAS mutations in the ctDNA of patients with colorectal cancer has been shown to correlate with resistance to EGFR blockade, suggesting that this “liquid biopsy” technique could be used to direct therapy and monitor resistance. Emerging somatic mutations could be detected, which may help clinicians determine their choice of next therapy.

    Of course, despite the apparent promise of this technique, there are many unknowns and it likely will be several years before liquid biopsies are routine. The applicability of this technique is currently limited to research laboratories at academic institutions, and its utility in the real world of community-based oncology is unknown. Current data suggest that for lymphoma, fresh frozen tissue is the preferred source for determining the tumor-specific clonotypic sequence, although the investigators report that when sufficient tissue was available, they were able to get a sequence from formalin-fixed, paraffin-embedded tissue in most patients.

    In their real-world validation, a usable sequence was found in slightly more than 70% of patients. Although there are diseases in which monitoring of minimal residual disease has had a major impact on treatment and outcome — chronic myeloid leukemia and acute lymphoblastic leukemia are the most notable examples — whether the same would be true across the spectrum of solid and hematologic malignancies is unknown and will require prospective evaluation.

    Whether the additional lead time provided by ctDNA detection of relapse will translate into improved outcomes for patients is not clear, but if the preliminary data for surveillance in lymphoma are confirmed in later studies, ctDNA certainly could replace or complement some of our existing surveillance tests, and patients will be able to avoid the anxiety, discomfort and cost of unnecessary biopsies or other tests that are the consequence of false-positive results.

    As a recent convert to the concept of liquid biopsies, I am excited by the potential for this technique and look forward to new data on its applicability across multiple tumors and in multiple health care settings. Translating the currently available tests into an affordable, reliable clinical investigation will be challenging and likely will take several years, but the prospect of a reliable blood test for following some cancer types — including aggressive lymphomas — is starting to look better.

    References:

    Bettegowda C, et al. Sci Transl Med. 2014;doi:10.1126/scitranslmed.3007094.

    Dawson SJ, et al. N Engl J Med. 2013;doi:10.1056/NEJMoa1213261.

    Kurtz DM, et al. Blood. 2015;pii:blood-2015-03-635169.

    Murtaza M, et al. Nature. 2013;doi:10.1038/nature12065.

    Roschewski M, et al. Lancet Oncol. 2015:doi:10.1016/S1470-2045(15)70106-3.

    For more information:

    John Sweetenham, MD, is HemOnc Today’s Chief Medical Editor for Hematology. He also is senior director of clinical affairs and executive medical director at Huntsman Cancer Institute at the University of Utah. He can be reached at john.sweetenham@hci.utah.edu.

    Disclosure: Sweetenham reports no relevant financial disclosures.