Q: Some workers in the field of precision oncology are growing despondent because of observed limitations of therapeutic effectiveness. What do you believe are now the best approaches to consider in order to move towards potential cures?
A: The pillars of precision oncology are genomically-targeted and immune-targeted therapies. Of interest, the fields of immunotherapy and genomics, often considered to be separate, are actually married to each other. Inhibiting checkpoints exploited by the tumor to shield itself from the immune machinery can reactivate the immune system. But, once reactivated, the immune system must be able to differentiate tumor cells from normal elements. The mutanome produces neo-antigens that permit this differentiation—and the more genomically aberrant the cancer, the more likely that the reactivated immune system will eradicate it. In contrast, with genomically-targeted therapies, the fewer the aberrations, the less likely that resistance will occur.
Yet, in treating patients, we are using old clinical trial paradigms rather than adjusting to the realities unveiled by genomic interrogation.
For instance, we administer genomically targeted therapies, mainly as monotherapy, to patients with heavily-pretreated, advanced metastatic disease—ie., the equivalent of heavily-pretreated chronic myelogenous leukemia (CML) blast crisis (where the response rate to imatinib is vanishingly low) (see Table). These patients have highly complex molecular portfolios and we should not expect that genomically targeted monotherapies will be curative.
We should not be then wringing our hands and talking about the limitations of our therapy–but rather the limitations of our approach.
The real eye opener is that so many patients with such late-stage disease respond at all. The response rate of metastatic solid tumors to genomically targeted agents is actually much higher than that for imatinib in late-stage CML (see Table). Yet, CML is considered the poster child for successful genomically-targeted therapy, with the median survival increasing from about 4-5 years to a near normal life expectancy.
The key ingredients for transforming CML included:
- Identifying the driver (Bcr-Abl)
- Identifying matched targeted therapy (imatinib)
- Moving from late-stage disease (blast crisis—the biologic equivalent of metastatic solid tumors) to newly diagnosed disease.
In solid tumors, as in CML, the cancer evolves with time and becomes more genomically complex. If we therefore want to use genomically targeted treatments to cure more patients with solid tumors, we need to fundamentally change our approach:
- Move to newly diagnosed disease
- Use customized combinations rather than monotherapy or random combinations for metastatic disease, which inevitably harbors complicated molecular alterations that differ from patient to patient.
Intriguingly, for patients whose tumors have highly chaotic genomes, immunotherapy is most effective. Indeed, in diseases such as melanoma, characterized by high tumor mutational burden, the long-term disease-free survival with immunotherapy may now be about 50%–an incredible improvement over the ~18% two-year survival seen with traditional chemotherapy.
In summary, we should be elated with the results yielded to date by precision oncology. If we, however, want to leap from improvements to cures, we need to stop relying on age-old clinical trial designs and adjust our strategy to the biology of each patient’s cancer.
Table: Precision Treatments and Response Rates
|Cancer diagnoses||Precision Treatments||Biomarkers||Response rates (%)|
|CML (newly diagnosed)
CML (heavily-pretreated blast crisis)
|Colorectal cancer||Pembrolizumab||Mismatch repair deficiency||~70-80%|
|Gastrointestinal stromal tumors||Imatinib||KIT||~50-80%|
|Hodgkin lymphoma (refractory)||Nivolumab
|Non-small cell lung cancer||Crizotinib
Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.