It is difficult to overstate how much our ability to study the fundamental molecular profiles within individual tumors has revolutionized our basic understanding of the biology of malignant disease.1-3 Further, this information has clearly identified both the heterogeneity of possible driver mutations and pathways within the cancers of individual patients as well as the large number of possible resistance mechanisms available to permit individual cancers to overcome both traditional cytotoxic and the more “targeted” antineoplastic agents.4
However, as interesting as this information may be to cancer biologists, what is far more relevant to cancer patients, their families, and treating oncologists is how such data can be increasingly utilized to favorably mpact clinical outcomes.5-7
The term “precision medicine” has been used to describe the impact of knowledge of the unique molecular changes present in individual tumors to devise a more precise therapeutic strategy for that particular patient.8 The “precision” will hopefully result in both improved efficacy (greater “on-target” effect) and reduced toxicity (less “off-target” effect).
It is relevant to acknowledge here that this process of “precision medicine” is certainly not a new concept. For example, for more than 40 years physicians have used knowledge of the presence (or absence) of estrogen/progesterone receptors in breast cancers to determine if hormonal manipulation (eg, removal of the ovaries, administration of tamoxifen, etc) is an appropriate management strategy for that specific individual.
However, what has changed over the past 10+ years is the number of unique settings where such data are of clinical relevance and the rather remarkable acceleration in the development of such knowledge in additional malignancies.
Consider for a moment the fact that it was less than 2 years ago when the first solid reports appeared identifying the fact that a class of anti-neoplastic agents known as BRAF inhibitors could produce rather spectacular responses in patients with metastatic melanoma whose cancers possessed particular mutations in BRAF.7 However, despite this very short time interval, a recently reported phase III trial has now revealed the clinicalsuperiority (prolongation of time-to-disease progression) associated with combining a BRAF inhibitor with another “targeted” agent shown to influence a common mechanism of resistance to such inhibition compared with treatment with the BRAF inhibitor alone.9
So, within a remarkably limited period of time we have gone from a theorized role for the presence of a BRAF mutation in the progression of malignant melanoma10 to the demonstration of the favorable impact associated with administering an inhibitor directed at that mutation7 to a strategy that effectively (at least for a clinically meaningful period of time) interferes with a cancer’s ability to resist the effects of such inhibition of its growth.9
The Impact of the Revolution in Genomic Sequencing Technology
Perhaps the single most important factor in propelling this spectacular new era forward has been the stunning advances in the technology associated with sequencing.11,12 The combined activities of industry and academia have led to a profound reduction in both the time required and costs associated with generating these vitally important data.
Consider the fact that the landmark report of the first completely sequenced human genome less than 2 decades ago required a number of years to accomplish at a cost of several billion dollars. It is estimated that this process can currently be accomplished in a matter of days at a cost of several thousand dollars.12 And both the time required and cost of conducting the sequencing for individual tumor specimens (and corresponding non-malignant tissue in the same individual) will almost certainly decrease in the future.
Viewing the Map Versus Breaking the Code for Individual Cancers
At this point it is essential to acknowledge the fundamental difference between possessing specific sequencing data on individual tumors or even highly detailed knowledge of the differences present within such tumors compared with the corresponding genome of the non-malignant tissue in that individual patient versus actually knowing how to interpret the clinical implications of the findings or how to use the information to develop the most effective therapeutic strategy.
Another possible way to highlight this critically important distinction is to compare obtaining the sequencing data itself (as truly eloquent as this may be) with the simple discovery of a previously hidden treasure map or a secret code. And, as interesting as such events may be, it is only after the map/ code has been correctly interpreted and the treasure found or the code broken that the map/code is of genuine value to its owner(s), just as the successful interrogation of the highly complex genomic data and the discovery of clinical relevance will be required before one can conclude there is true value to the patient associated with this effort.
The Multiple Unanswered Questions
How long it will take to discover the clinical relevance of these molecular data in particular clinical settings (eg, triplenegative breast cancer, squamous cell lung cancer) or in individual patients in such settings (eg, PI3 kinase mutation in ovarian cancer) and to develop effective therapeutics based on the abnormalities identified remains a critical unknown.
It is clear that it will be increasingly possible to obtain molecular profiles on individual tumors in a highly costeffective manner. Further, with time the essential informatics support will surely be able to analyze/interpret possible clinical implications of unique findings. However, a number of highly relevant questions will need to be addressed to permit the most effective use of this spectacular new technology.
For example, if a molecular abnormality is identified in an individual patient’s cancer that has been associated with the favorable effect of a particular commercially available anti-neoplastic agent but in a different tumor type, is it essential for a clinical trial to be conducted and the results reported before it is possible for patients to be treated with that agent in the new tumor site?
Should third-party payers be encouraged or perhaps even required to pay for such treatment based on the presence of a very specific molecular profile that identifies a potential driver mutation in a tumor where formal regulatory approval for this indication has not occurred? Now add to this query the fact that only a very small percentage of patients with this particular tumor type will possess the abnormality (eg, <2%-5%), making it very difficult (if not truly impossible) to conduct a comparative phase III or even a non-randomized phase II study.
In fact, it is highly likely that the scenario posed above will be very common once whole genome sequencing becomes rather standard in the management of advanced or recurrent cancers. How should the 1% of patients found to have “Mutation X” in “Cancer Y” be managed if the documented presence of “Mutation X” in “Cancer Z” has been shown to be favorably impacted if “Drug A” (US Food and Drug Administration— approved for “Cancer Z”) is administered?
A Novel Clinical Cancer Research Paradigm: “N of 1”
In the opinion of this commentator, the outline of a possible solution to this complex individual patient and societal dilemma can be provided by proposing a paradigm-changing clinical research strategy, called by some “N of 1.”
This strategy requires the outcome data generated following the patient described above (eg, 1% of patients with “Mutation X” in “Cancer Y” treated with “Drug A”) to be collected (with any information identifying the individual completely removed) in a database to be joined with such data obtained from similar patients (again, among that 1% of patients with “Mutation X” in “Cancer Y” treated with “Drug A”) to evaluate the efficacy (or perhaps even unique toxicity) associated with the use of this specific agent in the particular setting.
By combining a number of “N of 1” experiences it should be possible to obtain a reasonable view of the relative or absolute benefits associated with the use of this agent. Depending on the particular setting, the relevant end point could be the objective response rate or progression-free survival (compared with a recognized historical population) in the first 20 or 30 “N of 1” patient experiences.
An additional and quite novel end point to consider would be to use the patient as her/his own control, with the time to disease progression on the current regimen (selected at least in part based on molecular testing) being compared with the time to disease progression on the individual patient’s prior treatment regimen.
Based on a rational view of biological systems there is no reason to believe that the rate of disease progression in an individual tumor should improve/ decrease over time in the absence of a favorable effect of a biologically and clinically active anti-neoplastic program. Thus, if the time to disease progression is longer following the current regimen, compared with the prior regimen, it is reasonable to conclude this outcome must have been due (at least in part) to the effects of the current treatment.13 This novel approach to evaluating efficacy of molecularly based antineoplastic therapy has been employed in at least one previously reported highly provocative clinical trial.14
Further, the value of the “N of 1” strategy may be substantially enhanced by a proposal that the overall result of a particular experience will be declared to be “positive” only if a genuinely major favorable outcome is documented.
Thus, in this research scenario, a 10% to 15% objective response rate with responses lasting a median of 2 to 3 months will unlikely be considered an outcome that would justify “further research,” as is often stated in the conclusions of manuscripts describing the results of phase II drug trials in the cancer arena. Rather, in this concept, considering both the likely toxicity and anticipated costs of these treatment strategies, one will be looking for solid evidence of more meaningful clinical activity (eg, minimum 30%-40% objective response rate persisting a median of >5-6 months).
Future advances in anti-neoplastic drug therapy will be increasingly based on knowledge of the relevant driver mutations/pathways in individual cancers. Technological advances will soon make it possible to rapidly, cost-effectively, and routinely examine the molecular profiles of the tumors in individual cancer patients.
However, a new clinical research paradigm will be required to evaluate the effectiveness of therapy when small subsets of patients are being managed with a novel strategy. Finally, serious and unavoidable cost considerations will soon mandate a far higher threshold for new/novel anti-neoplastic strategy to be considered to have produced an outcome of genuine clinical value.
Author Affliation: From Cancer Treatment Centers of America, Philadelphia, PA.
Funding Source: None.
Author Disclosure: Dr Markman reports that he has received consultancies for advisory boards from Foundation Medicine.
Authorship Information: Concept and design; analysis and interpretation of data; and drafting of the manuscript.
Author correspondence: Maurie Markman, MD, Senior Vice President for Clinical Affairs, Eastern Regional Medical Center, Cancer Treatment Centers of America, 1331 E Wyoming Ave, Philadelphia, PA 19124. E-mail: email@example.com. DeVita VT Jr, Rosenberg SA. Two hundred years of cancer research. N Engl J Med. 2012;366(23): 2207-2214.
2. Patel JP, Gonen M, Figueroa ME, et al. Prognostic relevance of integrated genetic profiling in acute myeloid leukemia. N Engl J Med. 2012;366 (12):1079-1089.
3. McDermott U, Downing JR, Stratton MR. Genomics and the continuum of cancer care. N Engl J Med. 2011;364(4):340-350.
4. Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366(10):883-892.
5. Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med. 2001;344(14):1031-1037.
6. Demetri GD, von Mehren M, Blanke CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl
J Med. 2002;347(7):472-480.
7. Sosman JA, Kim KB, Schuchter L, et al. Survival in BRAF V600-Mutant advanced melanoma treated with vemurafenib. N Engl J Med. 2012; 366(8):707-714.
8. Mirnezami R, Nicholson J, Darzi A. Preparing for precision medicine. N Engl J Med. 2012;366 (6):489-491.
9. Flaherty KT, Infante JR, Daud A, et al. Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. N Engl J Med. In press.
10. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417(6892):949-954.
11. Pasche B, Absher D. Whole-genome sequencing: a step closer to personalized medicine. JAMA. 2011;305(15):1596-1597.
12. Tran B, Dancey JE, Kamel-Reid S, et al. Cancer genomics: technology, discovery, and translation. J Clin Oncol. 2012;30(6):647-660.
13. Markman M, Markman J, Webster K, et al. Duration of response to second-line, platinumbased chemotherapy for ovarian cancer: implications for patient management and clinical trial design. J Clin Oncol. 2004;22(15):3120-3125.
14. Von Hoff DD, Stephenson JJ, Rosen P, et al. Pilot study using molecular profiling of patient’s tumors to find potential targets and select treatments for their refractory cancers. J Clin Oncol. 2012;28(33):4877-4883.