Taking the shortcut to frontline treatment

From Dr. Schmitt and the Radich lab, Clinical Research Division

September 17, 2018 • By - AS Kuhlmann

Often, failure of cancer treatment can be attributed to mutations arising in critical genes. These so-called resistance-mutations can accumulate as a result of genetic alterations favoring their accumulation in cancer cells and/or be selected by treatments. Despite the great efficiency of recent pan-target therapies or pan-inhibitors that generally overcome any resistance mutation, some cases of resistance persist.

For Drs. Michael Schmitt (University of Washington) and Jerry Radich (Clinical Research Division), identifying these mutations and understanding the mechanisms leading to their emergence is a critical step to prioritizing treatment options and improving outcomes. In a recent study published in the journal Clinical Cancer Research, the researchers, in collaboration with Dr. Justin Pritchard from Pennsylvania State University, investigated the emergence of resistance mutations over the course of sequential treatments.

The group studied chronic phase chronic myeloid leukemia (CP-CML) and Philadelphia positive acute lymphoblastic leukemia (Ph+ ALL), two diseases with different sensitivities to pan-inhibitors. CP-CML and Ph+ ALL patients present a translocation between 2 chromosomes, the so-called the Philadelphia chromosome, resulting in fusion of the BCR and ABL genes and creating the tyrosine kinase-encoding BCR-ABL oncogene. However, while few mutations are observed in the ABL1 gene in CML patients, mutations are frequent in Ph+ ALL patients and associated with tyrosine kinase inhibitors (TKI) treatment failure. To understand this discrepancy, the ABL1 gene was selected for mutation analyses from peripheral blood samples. Duplex sequencing, a highly accurate sequencing method was used because of its ability to detect low-frequency mutations in heterogeneous populations by sequencing both strands of DNA (see Figure).

Duplex sequencing is a highly sensitive and specific sequencing method, analyzing both strands of genomic DNA. As such, true mutations are identified when found on both strands, in contrast to other methods such as next-generation sequencing, which generates a significant amount of errors and false positives. Figure provided by Dr. Michael Schmitt.

According to Dr. Schmitt, “the relatively low burden of tumor-initiating cells in CP-CML may explain the unusual success of molecularly targeted therapy in this disease. However, other factors may also play a role. For example, Ph+ ALL may be characterized by a higher degree of genomic instability, which could result in an increased propensity to accumulate mutations”. Indeed, CP-CML patients presented with very low mutation frequency similar to healthy patients, averaging 3.7x10^-7 unique mutations per base pair of the sequenced ABL1 gene. In the case of Ph+ ALL patients, many mutations were identified relative to the CML or healthy patients that were most likely selected by prior treatments. Indeed, 93% of Ph+ ALL patients harbored several mutations associated with resistance to TKI treatments, which were not detected at baseline. Dr. Schmitt explained that “because resistance mutations arise in single cells, they are not expected to become directly detectable unless drug therapy results in selective growth of the mutant cells. We are currently analyzing a larger set of treatment-naive patients to further establish whether pre-existing mutations can be found in some settings. We are also exploring whether the likelihood of developing resistance mutations is influenced by the extent of mutational heterogeneity at the time of diagnosis”.

CP-CML is characterized by more differentiated cells, as such, mutated tumor initiating cells are rarely detected, however, this fraction is more predominant in Ph+ ALL. Additionally, computational simulations in the ABL1 gene determined the likelihood of pre-existing mutations versus mutations acquired during prior sequential treatments in patients. More resistance mutations were detected in leukemia initiating cells in Ph+ ALL than in CML patients. Mutations were less likely to occur de novo than following sequential selection by treatments. Two out of 13 treatment refractory Ph+ ALL patients were identified with three mutations inferred as stepwise apparition based on treatment history, confirming the risk presented by sequential TKI treatments administration. Overall, these data support the notion that resistance mutations in the ABL1 gene accumulate as a consequence of prior sequential treatments.

As an option, the use of pan-inhibitors could benefit from use as a frontline therapy. “I look forward to seeing whether our results can be translated into improved outcomes in cancer patients. In particular, the study suggests that patient outcomes might be improved if pan-target therapies (i.e. therapies that can overcome any single resistance mutation) are used in the front-line setting. This concept is actively being tested in the clinic with ponatinib, which is a BCR-ABL pan-inhibitor. In addition, pan-target therapies are actively being developed against several other drug targets,” concluded Dr. Schmitt.

Fred Hutch/UW Cancer Consortium faculty members Drs. Lawrence Loeb and Jerry Radich contributed to this research.

Funding for this study was provided by the National Institutes of Health.

Schmitt MW, Pritchard JR, Leighow SM, Aminov BI, Beppu L, Kim DS, Hodgson JG, Rivera VM, Loeb LA, Radich J. 2018. Single-molecule sequencing reveals patterns of pre-existing drug resistance that suggest treatment strategies in Philadelphia-positive leukemias. Clinical Cancer Research. [Epub ahead of print].