publication date: Jan. 8, 2020

Guest Editorial

Immunotherapy, precision medicine in lung cancer drive sharp decline in cancer mortality overall

Official headshot for Otis Brawley, MD, MACP, FASCO, FACE, American Cancer Society Chief Medical and Scientific Officer

Otis W. Brawley, MD

Bloomberg Distinguished Professor of Oncology and Epidemiology,

Johns Hopkins University


As I look through just-published tables of age-adjusted cancer mortality, I recognize an unprecedented development:

Immunotherapy is showing such a dramatic impact in the treatment of locally advanced and advanced non-small cell lung cancer that this effect elevates the statistics for all lung cancer and—this I find astonishing—you can even see its effect in age-adjusted cancer mortality overall.

I am a cautious observer. I resist the common oncologic groupthink that declares any small advance a tremendous breakthrough. In cancer, dribs, drabs and fluky observations have often triggered dancing in the streets, but this is none of the above. It’s big, real, undeniable, and it’s an honor to write these words:

Today, 49 years after the signing of the National Cancer Act, we look at the 2017 cancer data and see validation of its small-c catholic approach to cancer. Rigorous research, visionary drug regulation, and relentless public health measures have brought about tangible change.

In the latest data, published by the American Cancer Society in the annual edition of Cancer Facts and Figures, the 2017 age-adjusted cancer mortality overall stands at 152.4 per 100,000, which is 29% lower than the age-adjusted death rate in 1991 (215.1 per 100,000).

In the aggregate, the 2017 rate is 2.2% lower than the 2016 rate. This is the largest year-to-year drop since beginning of the death rate decline. I feel comfortable making the prediction that the trend in lung cancer will continue as immunotherapy moves into the adjuvant and neoadjuvant realm.

Lung cancer has always been the biggest driver in the 26-year trend of declining age-adjusted cancer mortality. However, the decline we are seeing between 2016 and 2017 is unprecedented:

Lung cancer age-adjusted death rate for men dropped by 5%. For women, it dropped by 4%.

This was a dominant driver in the 2.2% decline in overall age-adjusted death rate.

Of the nearly 300,000 Americans dying of cancer in 1991, 34% died of lung cancer. Now fast-forward to the latest numbers: of the slightly more than 600,000 who will die of cancer this year, about 25% will die of lung cancer.

Cancer Facts and Figures uses data collected by the U.S. Centers for Disease Control and Prevention and the National Center for Health Statistics. The incidence and mortality data are age-adjusted to a standard population, in this case the year 2000 U.S. population, to remove the effect of the aging. Rates per 100,000 are used to remove the fact that the size of the population has more than doubled. Also, rates better reflect individual risk of cancer death.

What’s driving the decreases in age-adjusted lung cancer mortality?

It’s always the same trifecta: prevention, detection and treatment. Let’s look at them one-by-one.

It’s undeniable that tobacco control dating back to the 1960s is a major element in the decline in lung cancer deaths. But tobacco control gives you small, steady improvements. Dramatic results have to come from elsewhere. 

What about screening? Early detection?

The National Lung Screening Trial has demonstrated that low-dose CT screening has the potential to reduce risk of death by 20%. Alas, dissemination of effective screening programs has proven logistically difficult, and studies suggest that fewer than 2% of eligible current and former smokers are getting screened. So, I can’t see how screening can account for so dramatic a drop.

This leaves one possible factor: treatment.

As a medical oncologist for more than three decades, I’ve seen limited advancement in the treatment of lung cancer. In frustration, we started to refer to it as a “recalcitrant tumor.” 

Advances in treatment of locally advanced and advanced NSCLC come from several disciplines. There have been improvements in imaging and staging, especially through use of positron emission tomography scanning, and improvements in radiation therapy, especially through use of intensity-modulated radiation therapy and respiratory gated therapy.

And—finally—there are drugs. Here is a list of drug-based therapies approved for NSCLC. As you glance at these drugs, which have been approved over the past two decades and consider endpoints for their approval, it would be reasonable to spar about the fine points of whether a particular drug deserved to be approved for a particular indication based on a particular endpoint.

drug tables download web

As a card-carrying skeptic and a past member of the FDA Oncologic Drugs Advisory Committee, I eagerly join such debates. However, as I look at the aggregate data for this formerly recalcitrant cancer, I see validation of FDA’s regulatory strategy that has evolved over these two decades.

Yes, randomized controlled trials remain the gold standard in medicine. And, of course, showing an increase in overall survival should be a goal we must pursue whenever possible.

Alas, the facts on the ground show that RCTs aren’t always possible and overall survival can’t always be assessed. This can happen because some diseases have long natural histories, sometimes prolonged by several lines of effective therapies.

RCTs in such indications would have to go on for a decade or longer, likely past the point where the scientific questions being tested could be expected to be rendered moot. Also, diseases that afflict small populations defy randomization. Indeed, how would you randomize a trial in an ultra-rare sarcoma?

And equipoise can be lost, because doctors and patients become convinced by earlier data.

The increasing use of precision drugs and reliance on molecular markers complicates the assessment of drugs and treatments. It reduces the number of patients available to go into large prospective randomized trials and leads to the use of bucket trials such as NCI-MATCH or ASCO TAPUR trials.

These trials are composed of multiple concurrent small phase II studies.

It’s increasingly difficult to argue with the notion that there will be fewer drug approvals based on progression-free survival, time to progression and overall survival endpoints. The interpretation of clinical benefit is becoming more complicated, too, and more approvals will be due to patient-reported outcomes, response rate and an understanding of cancer biology.

FDA initiated the Accelerated Approval Program in 1992 to allow faster approval of drugs for serious conditions that fill an unmet medical need based on endpoints that are “reasonably likely reasonably likely to predict a real clinical benefit.”

As we look at the data for NSCLC, we can argue that the careful and rigorous use of validated intermediate endpoints in clinical trials has itself been validated. A number of these drugs received accelerated approval based on intermediate endpoints—mostly response rates, but also progression-free survival. Regular approval in the frontline follows after studies demonstrate overall survival.

However, progression-free survival and response rates are often used as a basis for regular approval in second- and third-line therapies as well as targeted therapies.

While checkpoint inhibitors appear to drive much of the change in lung cancer, other therapies are also making contributions. Thus, a table that lists all approved NSCLC can’t show what actually happens in the clinic, where oncologists are finding ways to use these drugs sequentially and in novel combinations.

While NSCLC data shows dramatic impact on population-level statistics, there are other dramatic successes, notably CML, an indication in which patients now have normal life expectancy, thanks to the development and dissemination of the precision medicine agent imatinib, and later, dasatinib, omacetaine, ponatinib and bosutinib.

The first of these drugs, imatinib was given AA for CML in 2001 based on response rate and regular approval in 2003 again based on response rate. Table 2 lists the CML drugs, types of approval granted, and endpoints on which approval was based.

There is also exciting progress in the treatment of metastatic melanoma with the approval of dabrafenib, trametinib, pembrolizumab, nivolumab, ipilimumab, talimogene laherparepvec, encorafenib and binimetinib.

The future of cancer therapy is ever-more multidisciplinary. It includes advanced imaging, surgery, radiotherapy, chemotherapy and immunotherapy with more complicated assessment of endpoints.

FDA regulation, too, must become increasingly multidisciplinary. The movement toward an oncology center that brings together all aspects of regulation of cancer-related products into one unified administrative entity would be a step in the right direction.

As America overcomes a great scientific challenge by developing and approving better therapies for cancer, a great new social challenge appears on the horizon.

The ACS paper that demonstrates the latest decrease in age-adjusted mortality also explores the decline in mortality from melanoma, focusing in part on populations less likely to have insurance, and therefore access to new expensive treatments.

The study found that poor populations didn’t experience these declines to the extent of those who are insured.

This is a damning statement about disparities in American society and our health care system. Newer therapies save lives—if you can write hefty enough a check.

The author is the Bloomberg Distinguished Professor of Oncology and Epidemiology at Johns Hopkins University. He is a former chief medical and scientific officer and executive vice president of the American Cancer Society.

Copyright (c) 2020 The Cancer Letter Inc.