Editorial

Delivering on the promise of precision oncology

Promises are like babies — easy to make, hard to deliver.

– Author unknown

The concept of genome-driven oncology became a reality almost 20 years ago with the introduction of imatinib, targeted to the ABL kinase.

John Sweetenham, MD, FRCP, FACP,
John Sweetenham

Even though this drug established a new paradigm for cancer treatment, progress in development of other targeted therapies was, at best, incremental for several years afterward. Although other targets were recognized, the field of so-called “precision oncology” advanced slowly until the emergence of next-generation sequencing (NGS) and the ability to match treatments to genomic alterations.

This approach was initially confined to a limited number of academic institutions, but the rapid expansion of vendors offering genomic testing, clinical decision support tools, clinical trial-matching platforms and, perhaps most importantly, new targeted drugs has resulted in widespread adoption of this concept across community oncology, as well as academic oncology centers.

Real-world impact

A brief walk around the exhibit hall at any major oncology meeting will provide ample evidence that tumor sequencing and treatment matching has become a major industry and that the investments in these platforms are increasing at a rate that seems to challenge the rise in the number of available targeted therapies.

Despite these changes, incorporating molecular profiling, molecular tumor boards and genomically matched treatments into regular clinic workflows remains a work in progress for many cancer centers and oncology practices. There are many reasons for this, including “genomic literacy” among oncologists; insurance coverage for NGS testing; availability of matched FDA-approved therapies or clinical trials; and the integration of test ordering, result reporting and clinical decision support into electronic health records.

As a result, despite early reports of exceptional response rates, prolonged PFS and cost savings with genomically matched treatments, the uptake into routine practice is slow and the overall present-day impact of this new approach is unclear.

Studies have shed some light on the current real-world impact of precision oncology from different perspectives.

In the first of these, investigators at Oregon Health & Science University have used publicly available data to explore the potential benefit of genome-directed therapy for patients with advanced or metastatic cancers. This population-level, observational study has substantial limitations, but some of the conclusions suggest that the current reality of genome-driven oncology is falling a long way short of expectations.

The authors identified 31 drugs with FDA approval from 2006 to early 2018 — which they defined as genome directed — and estimated the number of patients with advanced cancer eligible for therapy using mortality statistics from the American Cancer Society.

Researchers calculated that, in 2006, of almost 565,000 patients with advanced cancer, just under 29,000 (5.09%) were eligible for a genome-directed therapy, increasing to 8.33% (n = 50,811 of 609,640) in 2018.

Based on reported response rates to the 31 drugs in the analysis, researchers estimated that, in 2006, 0.7% of patients would have experienced a response, compared with 4.9% in 2018. Using published response data, the authors estimated that the median response rate for patients treated with a genome-informed drug was 54%, with a median response duration of 29.5 months.

So, despite impressive response and PFS rates for patients with advanced malignancy who can be paired with a targeted agent, the number of such patients is low.

Several other studies have reached a similar conclusion.

Researchers from Sanford Cancer Center reported an analysis of the impact of molecular profiling in a community-based program. In this much smaller, observational study, 120 patients with various tumor types enrolled onto a prospective study of the impact of genomically driven therapy.

Of the original 120 patients, 109 had successful tumor profiling, of whom 63 (58%) had an alteration considered actionable by the local molecular tumor board. Of those who completed molecular profiling, about 14% enrolled in a genomically matched clinical trial, and a further 21% were matched with an FDA-approved treatment or were treated off-label.

For the treated patients — on or off study — the overall response rate was only 12.8%, with no complete responses. For the patients achieving partial response or stable disease, 4-month PFS was 25.6%.

The PFS and OS curves showed no difference for those receiving therapy on a matched trial compared with those treated with an FDA-approved agent.

On the one hand, this is a disappointing result, given the overall relatively poor response and survival rates observed and the relative small percentage of patients whose disease course was impacted by tumor profiling. On the positive side, it demonstrates the feasibility of implementing a robust tumor profiling, molecular tumor board and decision support/matching process within a health system.

Long road ahead

The precision oncology space is very crowded, with vendors offering varying depths and breadths of tumor profiling and platforms for decision support, collaborative groups undertaking large-scale tumor profiling efforts such as ORIEN and GENIE, and treatment-matching trials such as the NCI-MATCH study and ASCO’s TAPUR study.

Although these efforts continue, many centers and practices are working to get up to speed with the current state of the art and ensure that their patients have the best possible access to targeted treatments with the highest potential for benefit.

This is a challenging process. I listed the barriers to implementation earlier and, in addition to the educational component to raise the level of genomic literacy for us as oncologists, there are many operational hurdles to overcome, including the challenge of complete EHR integration.

Up to now, those organizations that have been able to develop their own in-house NGS platforms appear to have had the most success in terms of implementation of their precision oncology programs. Even in these examples, actionable mutations seem to be present in only around 20% to 30% of patients with advanced or metastatic cancers, and the true current clinical benefit remains unclear.

There is no question that precision oncology will impact multiple areas of cancer care in the future, across the continuum from prevention and early detection, through choice of therapy and the management of short- and long-term toxicities.

The focus on precision medicine at this year’s ASCO Annual Meeting is evidence of the potential impact of genomic oncology for our present and future patients.

That said, it’s important to recognize that there is still a long road ahead to deliver on the promise of precision oncology. As our understanding grows, better sequencing platforms emerge, decision support tools become more integrated and user friendly, and new targets and drugs are identified, genome-driven therapies will for sure become part of our routine workflow. They may also produce major shifts from our current organ-based approach to cancer treatment.

Demonstrating value for this approach also will be essential.

Although some initial studies suggest that the precision approach delivers high value, this has really only been studied in the context of distinct episodes of care rather than across the full course of treatment for a specific disease.

Another important component of the value proposition will be the cost of targeted drugs. Despite the hope that generics and biosimilars will help to bring down the price of targeted therapies, there is little evidence to support this. A report on the cost of generic imatinib suggests that the cost savings may only be around 10%. After almost 20 years, we have seen an overall upward trend in the cost of the paradigm-shifting drug that changed the landscape of cancer care and opened the era of precision oncology.

How long will it take to deliver on the promise if this trend continues?

References:

Cole AL and Dusetzina SB. Health Aff (Millwood). 2018;doi:10.1377/hlthaff.2017.1684.

Marquart J, et al. JAMA Oncol. 2018;doi:10.1001/jamaoncol.2018.1660.

Nadauld LD, et al. Health Aff (Millwood). 2017;doi:10.1377/hlthaff.2017.1575.

Powell SF, et al. JCO Precis Oncol. 2018;doi:10.1200/PO.17.00220.

For more information:

John Sweetenham, MD, FRCP, FACP, is HemOnc Today’s Chief Medical Editor for Hematology. He also is senior director of clinical affairs and executive medical director of 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.

Promises are like babies — easy to make, hard to deliver.

– Author unknown

The concept of genome-driven oncology became a reality almost 20 years ago with the introduction of imatinib, targeted to the ABL kinase.

John Sweetenham, MD, FRCP, FACP,
John Sweetenham

Even though this drug established a new paradigm for cancer treatment, progress in development of other targeted therapies was, at best, incremental for several years afterward. Although other targets were recognized, the field of so-called “precision oncology” advanced slowly until the emergence of next-generation sequencing (NGS) and the ability to match treatments to genomic alterations.

This approach was initially confined to a limited number of academic institutions, but the rapid expansion of vendors offering genomic testing, clinical decision support tools, clinical trial-matching platforms and, perhaps most importantly, new targeted drugs has resulted in widespread adoption of this concept across community oncology, as well as academic oncology centers.

Real-world impact

A brief walk around the exhibit hall at any major oncology meeting will provide ample evidence that tumor sequencing and treatment matching has become a major industry and that the investments in these platforms are increasing at a rate that seems to challenge the rise in the number of available targeted therapies.

Despite these changes, incorporating molecular profiling, molecular tumor boards and genomically matched treatments into regular clinic workflows remains a work in progress for many cancer centers and oncology practices. There are many reasons for this, including “genomic literacy” among oncologists; insurance coverage for NGS testing; availability of matched FDA-approved therapies or clinical trials; and the integration of test ordering, result reporting and clinical decision support into electronic health records.

As a result, despite early reports of exceptional response rates, prolonged PFS and cost savings with genomically matched treatments, the uptake into routine practice is slow and the overall present-day impact of this new approach is unclear.

Studies have shed some light on the current real-world impact of precision oncology from different perspectives.

In the first of these, investigators at Oregon Health & Science University have used publicly available data to explore the potential benefit of genome-directed therapy for patients with advanced or metastatic cancers. This population-level, observational study has substantial limitations, but some of the conclusions suggest that the current reality of genome-driven oncology is falling a long way short of expectations.

The authors identified 31 drugs with FDA approval from 2006 to early 2018 — which they defined as genome directed — and estimated the number of patients with advanced cancer eligible for therapy using mortality statistics from the American Cancer Society.

PAGE BREAK

Researchers calculated that, in 2006, of almost 565,000 patients with advanced cancer, just under 29,000 (5.09%) were eligible for a genome-directed therapy, increasing to 8.33% (n = 50,811 of 609,640) in 2018.

Based on reported response rates to the 31 drugs in the analysis, researchers estimated that, in 2006, 0.7% of patients would have experienced a response, compared with 4.9% in 2018. Using published response data, the authors estimated that the median response rate for patients treated with a genome-informed drug was 54%, with a median response duration of 29.5 months.

So, despite impressive response and PFS rates for patients with advanced malignancy who can be paired with a targeted agent, the number of such patients is low.

Several other studies have reached a similar conclusion.

Researchers from Sanford Cancer Center reported an analysis of the impact of molecular profiling in a community-based program. In this much smaller, observational study, 120 patients with various tumor types enrolled onto a prospective study of the impact of genomically driven therapy.

Of the original 120 patients, 109 had successful tumor profiling, of whom 63 (58%) had an alteration considered actionable by the local molecular tumor board. Of those who completed molecular profiling, about 14% enrolled in a genomically matched clinical trial, and a further 21% were matched with an FDA-approved treatment or were treated off-label.

For the treated patients — on or off study — the overall response rate was only 12.8%, with no complete responses. For the patients achieving partial response or stable disease, 4-month PFS was 25.6%.

The PFS and OS curves showed no difference for those receiving therapy on a matched trial compared with those treated with an FDA-approved agent.

On the one hand, this is a disappointing result, given the overall relatively poor response and survival rates observed and the relative small percentage of patients whose disease course was impacted by tumor profiling. On the positive side, it demonstrates the feasibility of implementing a robust tumor profiling, molecular tumor board and decision support/matching process within a health system.

Long road ahead

The precision oncology space is very crowded, with vendors offering varying depths and breadths of tumor profiling and platforms for decision support, collaborative groups undertaking large-scale tumor profiling efforts such as ORIEN and GENIE, and treatment-matching trials such as the NCI-MATCH study and ASCO’s TAPUR study.

Although these efforts continue, many centers and practices are working to get up to speed with the current state of the art and ensure that their patients have the best possible access to targeted treatments with the highest potential for benefit.

PAGE BREAK

This is a challenging process. I listed the barriers to implementation earlier and, in addition to the educational component to raise the level of genomic literacy for us as oncologists, there are many operational hurdles to overcome, including the challenge of complete EHR integration.

Up to now, those organizations that have been able to develop their own in-house NGS platforms appear to have had the most success in terms of implementation of their precision oncology programs. Even in these examples, actionable mutations seem to be present in only around 20% to 30% of patients with advanced or metastatic cancers, and the true current clinical benefit remains unclear.

There is no question that precision oncology will impact multiple areas of cancer care in the future, across the continuum from prevention and early detection, through choice of therapy and the management of short- and long-term toxicities.

The focus on precision medicine at this year’s ASCO Annual Meeting is evidence of the potential impact of genomic oncology for our present and future patients.

That said, it’s important to recognize that there is still a long road ahead to deliver on the promise of precision oncology. As our understanding grows, better sequencing platforms emerge, decision support tools become more integrated and user friendly, and new targets and drugs are identified, genome-driven therapies will for sure become part of our routine workflow. They may also produce major shifts from our current organ-based approach to cancer treatment.

Demonstrating value for this approach also will be essential.

Although some initial studies suggest that the precision approach delivers high value, this has really only been studied in the context of distinct episodes of care rather than across the full course of treatment for a specific disease.

Another important component of the value proposition will be the cost of targeted drugs. Despite the hope that generics and biosimilars will help to bring down the price of targeted therapies, there is little evidence to support this. A report on the cost of generic imatinib suggests that the cost savings may only be around 10%. After almost 20 years, we have seen an overall upward trend in the cost of the paradigm-shifting drug that changed the landscape of cancer care and opened the era of precision oncology.

How long will it take to deliver on the promise if this trend continues?

References:

Cole AL and Dusetzina SB. Health Aff (Millwood). 2018;doi:10.1377/hlthaff.2017.1684.

Marquart J, et al. JAMA Oncol. 2018;doi:10.1001/jamaoncol.2018.1660.

Nadauld LD, et al. Health Aff (Millwood). 2017;doi:10.1377/hlthaff.2017.1575.

Powell SF, et al. JCO Precis Oncol. 2018;doi:10.1200/PO.17.00220.

For more information:

John Sweetenham, MD, FRCP, FACP, is HemOnc Today’s Chief Medical Editor for Hematology. He also is senior director of clinical affairs and executive medical director of 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.