University of California San Francisco | About UCSF | UCSF Benioff Children's Hospital San Francisco
Search Site | Find a Doctor

Interview with Dr. David Jablons: New Lung Cancer Test Can Guide Treatment, Improve Outcomes

Audio Interview

Listen to a Patient Power interview with Dr. David Jablons, chief of general thoracic surgery at UCSF Medical Center and program leader of the Thoracic Oncology Program at the UCSF Helen Diller Family Comprehensive Cancer Center. Jablons speaks about a new lung cancer test designed to help guide treatment and improve survival rates.

Interview Transcript

Andrew Schorr:

Researchers at UCSF Medical Center in San Francisco are advancing a new lung cancer test to help guide treatment for people with this very serious illness. The test is designed to help improve the odds of survival. Coming up we'll hear from a leading lung cancer expert, Dr. David Jablons, who helped develop this test. It's all next on Patient Power.

Hello and welcome to Patient Power sponsored by UCSF Medical Center. I'm Andrew Schorr.

Lung cancer is our biggest cancer killer. In the United States, well over 200,000 people are diagnosed each year, and the death rate, unfortunately, is very high. We need to know how to stage patients better to know who needs treatment. If they're able to have surgery, for example, who needs chemotherapy? How do we test for that?

Well, at UCSF they've done some very promising research on this. And to help us understand this, we are joined by a leader in the field — Dr. David Jablons. He's chief of general thoracic surgery at UCSF. He's also their director of the Thoracic Oncology Program and the Thoracic Oncology Lab.

Dr. Jablons, welcome to Patient Power.

Dr. Jablons:

Hi, Andrew. Thanks so much for having me on your show.

Andrew Schorr:

All right. I know that you've really been devoting your career to helping understand the immunology of cancer, in this case lung cancer. Help us understand what you've done with molecular testing and what the significance could be for lung cancer.

The Challenge of Catching and Treating Lung Cancer

Dr. Jablons:

Sure. As you said in your introduction, lung cancer remains the dominant cancer killer not just in the United States but worldwide, claiming over a million deaths. In fact, it's predicted there will be approximately 1.2 or more million deaths a year around the world.

And the tragedy of lung cancer is twofold. One, for the most part, it is found in locally advanced to late stage, stage 3 or 4 typically. And then when we're fortunate to find it in early stage, which we're doing more and more now because of finally the adoption of a screening protocol, in this case, low dose spiral CT screening — even though we might find a two- or three-centimeter tumor and the patient undergoes a successful operation where lymph nodes are sampled and they are thought to have a complete surgical resection and these patients are staged at Stage I disease — even in that setting, the prognosis, unlike other solid organ cancers such as breast cancer, colon cancer, where stage I tumors from those diseases have a 90 percent survival or better, in lung cancer, it's just not quite so good.

And the reason is not completely clear, except to say that the tumors tend to be biologically a little bit more revved up, they tend to spread sooner, and we don't have as good a handle on the molecular profile, so to speak, of these tumors up until now to gauge their "personalities," if you will, their good or bad biology. And that's really what this technology that we've been working on is all about and that we recently published.

Basically, the idea is when patients have an early stage cancer resected, even stage I disease, their survival is described as not 90 or 95 percent. It's closer to 60 to 65, maybe 70 percent, which means 30 to 40 to almost 50 percent of these patients, especially if they have a larger tumor or if they have a lymph node within the lung specimen that's been removed involved — 30 to 50 percent of these patients, their survival is only that good, 50 percent.

And what that means and implies is that even though they've been imaged carefully and aggressively, so to speak, preoperatively with appropriate CAT scans, PET scans, brain MRIs etc., there is microscopic occult disease that just cannot be imaged in any way down to a cellular level.

New Genetic Test to Guide Early Stage Treatment

But what we have been working on and what is finally available now for patients is a way to analyze their removed or resected tumor. And just use a simple standard sample, that arrives in pathology that's been preserved in formalin and is imbedded in paraffin, where a section of that tumor can be analyzed for a gene profile that can give you a risk stratification based on the innate biology of the tumor, so to speak, to determine how aggressive that tumor is and as a mirror, so to speak, of the presumed incidence and correlation with occult or hidden microscopic metastatic disease.

And in particular what the technology was geared up to do, it was designed to improve the ability to prognosticate. In other words, to identify patients at high risk for recurrence and death after complete resection for early stage I and stage II.

And as you alluded to, in the United States, the crisis of non-small cell lung cancer really is not abetting at all. Over 220,000 Americans will be diagnosed with lung cancers. I mentioned a majority are in advanced stage. While we're making good progress in the advanced stage setting with understanding the molecular pathways, some of the receptors and the signaling and the "targeted" therapies and the use of specific designed drugs that can block those targets, whether they're epidermal growth factor, ALK mutations, whether we use anti-angiogenesis agents like bevacizumab — all these things are making a difference in extending lives with good quality of life.

But in the early stage setting, the critical thing is to find patients who are really not cured with surgery, and frankly those who are. And then once we have identified this subset of patients in a reliable and robust way, then apply what we know and have proven in over a decade of combined research in centers of excellence around the country and around the world.

May Increase Cure Rate

What we know is curative additional treatment, as you alluded to, which is adjuvant or postoperative chemotherapy — and in this case, good old fashioned cisplatin, doublet chemotherapy — that has been shown in early stage patients to not just prolong survival but actually increase significantly, between 5 to 15 percent, the chance of cure. The point of the story here is in the early stage setting we are really trying to get these patients cured, once and for all. In the advanced stage setting, as good as and as exciting as the progress is, sadly, we still don't really cure that many patients even though we can extend their survival.

And what's the impact and the magnitude? Well, it's significant because the lung cancer epidemic, pandemic really, is so out of control. In the United States, if you look at the stage I and stage II patients it's essentially 30 to 50,000 patients being diagnosed on a yearly basis. And if you think about it, what we can do now and what this data and test has shown, is that in the early stage patients, we can do a 14-gene analysis that allows us to identify patients that get assigned to a low or high or intermediate risk of long-term, five year survival after a complete resection.

This technology has been designed to address the dominant form of non-small cell lung cancer. Non-small cell lung cancer is the dominant form of lung cancer, compared to, say, small cell, meaning 80 percent at this point or more, 85 percent of all lung cancer diagnosed is the non-small cell type. And of the non-small cell type the majority now, at least 68 to almost 70 percent in the United States, is what's called an adenocarcinoma variant, when the pathologist looks under the microscope.

And so this technology was designed to identify in the dominant cancer form, the adenocarcinoma and its subsets, or the nonsquamous form of non-small cell lung cancer, it was designed to identify and risk stratify patients who after complete surgical resection had either a good prognosis with a low risk score, meaning their survival was closer to 80 percent at five years, versus those patients who sadly or unfortunately had a high risk score where their prognosis was at five years more like 40, 44 percent.

If you think about it, it's a remarkable separation of survival, 30 percent, based on a 14-gene assay. And remember that all of these patients — and this is why it's of interest not only to, of course, all the patients listening out there who found incidentally or through screening or for whatever reason, identified a small, early stage cancer and had a successful operation — unfortunately all those patients start off with the great hope that they're cured. But if you look at their biology, about 40 percent, 45 percent, almost half, which is exactly, if you look at the numbers a 60 percent survival at five years, that 40 to 45 percent of patients can be now identified by looking at their tumor for a molecular risk stratification where their prognosis would be 40 percent, not 75 or 80 percent.

The reason that's important is not to say, well, OK, this is problematic, which it is. It's important because we have proven curative beneficial treatment already that we use on a routine basis that we know is safe. It's chemotherapy, it has its side effects, but when you identify patients who are at high risk, the great hope is that these patients can then have standard NCCN [National Comprehensive Cancer Network] guideline-approved indications for patients whose tumors are identified as high risk by sort of the standard histologic, morphologic, meaning size, extent of disease margin involvement, lymph node, lymphovascular invasion, all these kind of standard criteria, the best we could do short of this sort of renaissance, if you will, an advent of a molecular enlightened era.

Which amazingly and interestingly enough, has already existed for the last five to six, eight years really, in breast cancer using Oncotype DX assay for breast cancer to identify those patients — a little bit of the reverse of this — who have good biology and probably do not need adjuvant chemotherapy.

The same is really true for this test, although the real problem in lung cancer is to find the bad biology patients and then do something that we know will improve and save lives, and that is to give adjuvant chemotherapy.

We haven't done the prospective trials to prove that these high risk patients then benefit as well, or more we predict, from chemotherapy. But there is very good data out there to suggest that patients who have high risk criteria by any molecular profile actually do better when given postoperative chemotherapy. So there's very good reasons for optimism.

Large Study Confirms Treatment Approach

There's very exciting confirmation of this study, just to finally sum up here, that other centers around the country and around the world have been looking at molecular profiles in lung cancer, in early stage lung cancer. But none of them have ever had significant — and by this way I mean very, very large, independent, blinded under strict CLIA (clinical laboratory improvement ammendments), meaning laboratory standards that are held to the strictest criteria. No studies have ever been done to the rigor of the studies we did.

We did a combined project with our colleagues and friends in the Kaiser Permanente Northern California system where we looked at almost 500 stage I patients by their criteria in a blinded independent fashion. We looked at the 500 samples in patients that we helped develop the initial test from UCSF. It was then partnered with a company that was started in the Bay Area, Pinpoint Genomics, which subsequently was bought and merged into Life Technologies, which is very exciting because this is allowing this test to get out and get to people around the world.

Then it was also done in concert with a consortium that we've been very active with over the last 20 years in China, believe it or not, where we have trained and worked with our colleagues along the major centers of excellence on the eastern seaboard of China, developing collaborations in a very effective inner group mechanism, so we could analyze a thousand patients, of which about half, or 500, similar to the Kaiser cohort, were early stage I patients out of China.

At the end of the day, in a blinded, independent CLIA fashion as robust as you could ever possibly divine, the dice were rolled, the test was run. And at the end of the day, this 14-gene assay was able on two continents from almost 2,000 patients — it's really the largest molecular study ever done in lung cancer, or maybe in any cancer for that matter, and certainly the largest in early stage lung cancer — that was able to stratify a low risk population that had a 75 percent to 80 percent survival versus a high risk population that had a 40 to 44 percent, with a T value that was amazingly significant.

Fourteen genes, two continents, same answer.

Andrew Schorr:

Wow. I know how excited you are, Doctor. So I just want to go over one point so we understand for patients. And that is, unfortunately, if somebody is diagnosed with lung cancer, that's of course such a serious diagnosis. Maybe with some of these spiral CT results, some people will be tested earlier and the cancer will be identified earlier, but the problem, if I've got it, before the molecular test was it was still maybe not sure whether even the people who looked to be on the road to cure from surgery really had cancer hiding somewhere in their body, and where more aggressive postoperative care could make a big difference.

And your molecular test that you've helped work on, that's been validated, as you said, could spell longer survival for these people who had those cancer cells hiding that otherwise would have led to their demise.

Dr. Jablons:

Well, that's exactly right. And what the molecular test does is it's a prognostic test, but it gives us refined ability, and your physician, surgeon and oncologist who you would consult with, it helps to get a more accurate gauge of a patient's chance for long-term survival and cure beyond standard histopathologic staging that is the standard of care.

And there is no doubt — I mean, everyone listening would agree that it's time to come into the 21st century. You can get online on your iPhone and at the speed of light look up a billion different things. We can clone your genome in a day. You can travel at seven miles up and 600 miles an hour, you know what I'm saying? Why are we using 16th century staining under a microscope? It's still effective, and it's important — very, very important. However, it needs additional adjuncts.

And those adjuncts are this renaissance that's going on right now in our understanding of biology. And especially cancer, where the need is so great to understand the molecular underpinnings of the disease — one, for therapeutic intervention and advanced stage, but, two, and apropos of our discussion today, for molecular prognostication and then eventually also for predicting response to therapy. But that is exactly right.

And the good news for people listening is that if you had your operation done three months ago or six months ago or possibly even a year ago — although the test and the validation studies where these patients did not have therapy based on the test but we looked at samples five years or later that have been in the bank — we can identify this signature and that signature can still be meaningful. And, if this were my patient, and we've been using the test now for over a year and a half or more, if you had a non-squamous cell lung cancer, early stage cancer resected, I would want to get the test done and then talk to your surgeon, internist, pulmonologist, and/or even oncologist if they're in the mix, about do they think there's a role for chemotherapy.

Because we know when the chemotherapy is given early it can make a difference. And how early that is — three to six months, a year, it's hard to know. But when it's given after patients recur, while it has benefit for sure, for amelioration of symptoms, for slowing down disease progression, unfortunately, once patients recur and get chemotherapy it's never curative. So the time to do it is up front.

Andrew Schorr:

Right. So, again, certainly people who are headed for surgery for early stage lung cancer want to check to make sure this molecular test can be available for them, and if, as Dr. Jablons just said, you had surgery not so long ago there are samples of your tumor that are available that could still be tested to guide whether you should start chemotherapy.

Dr. David Jablons, this is an example of the trend towards precision medicine, and I know you at UCSF have been a leader in this. I want to congratulate you on the work. I know you're part of a big team, as you said, on two continents.Thank you for your devotion to patients in this very serious cancer and making a real difference. We wish you well in your continued research.

Dr. Jablons:

Thank you so much, Andrew, for the talk.

Andrew Schorr:

All right. This is what we do on Patient Power is connect you with leading experts. Dr. Jablons is truly a leader in this field in understanding the biology, the immunology of cancer, and having each patient get the treatment that's right for them.

Thank you for joining us on Patient Power, and thanks to UCSF. I'm Andrew Schorr. Remember, knowledge can be the best medicine of all.

Recorded November 2012

 

Reviewed by health care specialists at UCSF Medical Center.

This information is for educational purposes only and is not intended to replace the advice of your doctor or health care provider. We encourage you to discuss with your doctor any questions or concerns you may have.

Related Information

UCSF Clinics & Centers

Helen Diller Family Comprehensive Cancer Center

Thoracic Surgery and Oncology
1600 Divisadero St., Fourth Floor
San Francisco, CA 94143-1705
Phone: (415) 885-3882
Fax: (415) 353-9525

Condition Information

Patient Experiences