Molecular Targeting in Cancer Therapy: Can We Fulfill the Promise?

Some years the presentations at the American Society of Clinical Oncology (ASCO) Annual Meeting represent modest advances in treatment, while other years dramatic breakthroughs are presented. I recall the thrill, at ASCO 2005, of being in the audience during the presentation of the combined results of the adjuvant breast cancer trials that showed a dramatic improvement with the addition of trastuzumab (Herceptin).[1] The audience was ecstatic, there was a prolonged standing ovation, and the presenter was likened to “a rock star” by the moderator. The excitement such presentations bring us gives hope that we take back with us to our offices and research centers.

Additional Promising Data in Non–Small-Cell Lung Cancer

In this paper, authors Joseph, Wu, and Muggia present Part II of a two-part review entitled “Targeted Therapy: Its Status and Promise in Selected Solid Tumors.” This part focuses on targeted therapy in breast, lung, colon, and ovarian cancer. Overall, the review is very thorough. However, because the authors provide a view from 30,000 feet, of necessity they omit some of the finer points in the development of the drugs involved and their use in these diseases.

To their list of targeted therapies for non–small-cell lung cancer (NSCLC), I would add one recent discovery and expand on another. Shaw et al[2] recently reported the results of using crizotinib to treat lung cancer patients who harbor the ROS1 mutation. The ROS1 mutation present is present in 1% to 2 % of NSCLC patients. Of 14 evaluable patients treated with crizotinib, 8 responded (57%) and 4 others had stable disease. There is also cell culture evidence[3] that the Hsp90 inhibitor ganetespib may be active in these patients. Joseph et al mentioned anti–Programmed Death (PD)-1 therapy as a promising approach. Topalian[4] reported that 9 of 25 NSCLC patients with PD-1 present (36%) achieved an objective response when treated with an anti–PD-1 antibody-compared with none of the 17 patients who had PD-1–negative tumors.

Molecular Tumor Characterization and Its Clinical Relevance-or Lack Thereof

What are some of the future needs of this area of oncology? With the advent of whole-genome sequencing, along with expression profiles and epigenetic mapping of tumors, it is becoming difficult for the clinician to keep track of all the information these new tools generate. Currently, tumors are classified by their organ of origin and cell type; if a molecular test is performed, it is added to the traditional pathology report as an addendum. Recent discoveries, however, have shown that tumors of different origins may share similarly altered biology and may benefit from similar treatments. The TNM staging system is no longer enough. There is a need for a clinical framework that could be used to classify relevant mutations and aberrant pathways so as to help clinicians. Such a framework would combine traditional pathology with the new information in a standardized and comprehensive fashion. It should be expandable to allow for the incorporation of new discoveries.

A plan is needed for the day-coming soon-when we routinely receive whole-genome sequencing of our patients’ tumors. How will we view this information, what will we look for, and how will we store the massive amounts of data each sequence will generate?

Some have questioned whether there will be any benefit to these efforts. As tempting as it may seem to have the entire genome of a patient’s tumor available, we may be better off dealing with a subset of mutations that we understand clinically and hav the ability to treat. Whole-genome sequencing may be best reserved for the research setting. For now, these are all unanswered questions, but they will be upon clinicians soon.

Issues Related to Tumor Heterogeneity

There also needs to be an acknowledgement and better understanding of tumor heterogeneity at the clinical level. Both traditional chemotherapy and targeted therapy fail at least in part because tumors contain subpopulations of cells that are not sensitive to the therapy used. This insensitivity may be due to pre-existing resistant subpopulations, or it may arise as a result of continued accumulation of mutations and selective pressure from the therapy administered. How we measure this insensitivity and monitor it during treatment is a vital consideration. When do we switch from one targeted therapy to another? Can we monitor the molecular target of the therapy for an escape signal or will we need to rely on traditional radiologic evidence of progression? Is it best to combine targeted therapies?

Why do tumors have such a propensity for finding pathways around blockades? Is this intrinsic to normal metabolism or only to malignant cells? What is the mechanism driving the development of these “escape hatches”? Can that be identified and blocked?

Clinical Trials in Targeted Therapy

A new approach to clinical research also seems necessary for targeted therapy. Individual tumors contain hundreds of mutations. Some of the mutations occur in only 1% to 2% of patients with a particular tumor type. How should a clinical trial be structured so that it will accrue enough patients? Can and should molecular endpoints be adopted, rather than using radiographic response? Some have suggested an approach that would entail obtaining tissue from a large number of patients, maintaining the specimens in a tissue bank, then testing for a particular target when an agent became available. Patients who had the target could then be evaluated for possible study enrollment.

Conclusion

A broad overview of the field, such as the one Joseph et al have provided in this review, makes us realize that we are in the infancy of our ability to understand and apply molecular targeting and genomics. Although a few targeted therapies are dramatic in their effectiveness, most have a modest impact. Many of the targeted agent studies have led to negative results. What has looked like a good idea has often been undone by our not understanding the intricacies and pathway redundancies inherent in tumor biology. Work remains in the development of a clinically useful tumor classification system that includes molecular characterization of tumors, in our understanding of the implications of tumor heterogeneity, and in the development of more relevant and efficient clinical trials. Nonetheless, there is great excitement that a new era in the treatment of cancer is beginning.

Financial Disclosure:The author has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.

References:

REFERENCES

1. Romond E, Perez E, Bryant J, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med. 2005;353;16: 1673-84.

2. Shaw A, Camidge D, Engelman J, et al. Clinical activity of crizotinib in advanced non-small cell lung cancer (NSCLC) harboring ROS1 gene rearrangement. J Clin Oncol. ASCO 2012 Meeting Abstracts;30(suppl): Abstract 7508.

3. Proia D, Acquaviva J, Jiang Q, et al. Preclinical activity of the Hsp90 inhibitor, ganetespib, in ALK-and ROS1-driven cancers. J Clin Oncol. ASCO 2012 Meeting Abstracts;30(suppl): Abstract 3090.

4. Topalian S, Hodi S, Brahmer J, et al. Safety, activity, and immune correlates of Anti-PD-1 antibody in cancer. N Engl J Med. 2012;366: 2443-54.