Lung Cancer
Kenneth Albert, MD
Lung cancer is not the most common cancer in either men or women. Prostate cancer
is certainly number one in men and breast cancer is number one in women. It cuts lung
cancer at approximately half; so you can see breast cancer is much more common than
lung cancer. However, if you look at the deaths from cancer, lung cancer, it is by far the
biggest killer in both men and women.
If you look at the changing incidence of lung cancer in men, the United Kingdom is
actually decreasing quite substantially from the earlier 1960s and '70s. In the United
States, it actually leveled off in the 1980s, and it has started to decline in the last
couple of years, and in France it is still rising. So it really just depends on where you
are and on your habits. In women, the United States is still going up. It hasn't showed
any sign of leveling off. The United Kingdom has already leveled off and started to
decline a little bit. In France, it is has always been low.
Types of lung cancer. There are two major types: non-small cell and small cell.
Approximately 75-80% of the tumors that we see are non-small cell, while only 20-25%
are small cell itself. If we look at the small cell subtypes, previously the vast majority of
those particularly with the disease found in males was squamous cell cancer, but more
recently the majority of them are adenocarcinomas, and 40% of all lung cancers are
adenocarcinomas. So this has really shot up and changed the nature of the disease as
well. Squamous cell is only 17% of all lung cancers.
In smokers, particularly males, squamous cell is a fairly predominant tumor type. But if
we look at nonsmokers, adenocarcinoma is by far the most common subtype. The same
is true of females, even to a greater extent. In smokers, adenocarcinoma is more
common in females.
LUNG CANCER Page 1 1
Staging. The T-stage is based on the size, the location, the amount of atelectasis and a
few other things. The N-stage is fairly simple. N1 is intrapulmonary lymph nodes, N2 is
outside and N3 is outside the chest and M1 is self-explanatory.
Stage 1 and 2 are localized to the lung itself. Stage 3A is more advanced with
metastases in general or chest wall invasion and 3B is your unresectable tumors. T3N0
is tumors that involve the parietal pleura either on the chest wall, the diaphragm or the
mediastinum, now it is considered to be stage 2 as well since the survival is actually
better. In these two groups up here, TI and T2, each one has been divided into A and
B. Now, there is 1A, 1B, 2A, 2B and then T3 can be observed in 2B as well.
Stage1 tumors have to be less than 3 cm and it can't involve a major lobar bronchus
and create lobar atelectasis or consolidation, and it can't have any lymph node
involvement either in the lung or outside. T2, they are just bigger lesions. Either greater
than 3 cm or those that involve the visceral pleura and that cause lobar atelectasis.
Segmental atelectasis is the T1 finding.
Stage 2 is the same two groups except for there is intrapulmonary lymph node
involvement so that the criteria are the same. This is 2A, this is 2B but there is
intrapulmonary lymph nodes either along the segmental bronchi or either the lobar
bronchi that are involved. Also, again, 2B consists of T3, which is tumors that are
involved in the parietal pleura or 2 cm with carina.
3A is locally advanced. T3N0 again goes to the 2B now but T3N1 and 2, T3 involves
basically anything that has parietal pleura. if you see something with complete collapse
of the lung, by definition they have to have a T3 disease.
The other component of a 3A is N2 disease and that means now lymph nodes up in this
area that are involved. So you can kind of get a feeling from the number of different
subgroups in this phase that there are a lot of different patient populations in here. N2
disease is a lot different from T3 disease so implications in terms of survival and
treatment are also different.
LUNG CANCER Page 2 2
3B again is something that is unresectable in general which includes N3 disease which
is either contralateral, if you have a tumor over here, contralateral lymph node
involvement on the opposite side or a scalene lymph nodes in this particular area or
you have a T4 lesion which is either a tumor growing into a major structure like an
aorta, esophagus, left atrium or some major unresectable organ or a malignant pleural
effusion.
One of the things surgeons keep in mind is the spread of tumors along lymph node
chains. If you have an upper lobe lesione and up along the peritracheal nodes here so
the subparietal nodes are oftentimes spared or the left side is spared. If you have a
middle lobe or lower lobe lesion, it tends to drain not down here actually but to the
subparietal lymph nodes and then up along the right side. On the left side actually, the
lower lobe tends to drain to the subparietal nodes and then switches over to the right
side as well so you get more contralateral involvement of the left lower lobe. The left
upper lobe tends to drain along this direction. Just like the right upper lobe it also goes
to the AP window which is on the left side.
So when you see a report where someone on mediastinoscopy biospies a mediastinal
lymph node, it is not accurate enough. It doesn't tell you enough information to really be
able to treat the patient as well as they can be treated.
Stage 1, clinically the survival is the best - 50-55%. Stage 2 drops down to 30% or less.
Stage 3A is 15 or less and Stage 3B and 4 are dismal. So if you have an accurately
staged 1 patient, either T1 or T2, survival is 60-75% or even 80% for small lesion so
that is fairly reasonable. But you the T1N0s are 75-80%. That is a pretty reasonable
survival but that means even when we start getting T2, no lymph node involvement, this
is a larger tumor or visceral pleural involvement it drops down in the 50s so you are
really just flipping a coin to see whether patients survive and then when you drop down
to stage 2, it goes down below the 30s. Generally 52 is kind of high, but it is 30-40% in
most stage 2 tumors. The 3A it drops from the 30s down to as low as 15% depending
upon which 3A we are talking about. Then 3B and 4 are dismal - less than 10%.
LUNG CANCER Page 3 3
Patients unfortunately present mostly with advanced 3B or 4 disease half the time and
then half the remaining patients would be the early resectable disease, stage 1 or 2 or
locally advanced, which is potentially resectable, but may not be good to resect
necessarily.
Distance spread for lung cancer as everybody knows is brain, bone, liver adrenals,
lymph nodes as well as the lungs. Those are the most common sites. But the problem
from the surgical standpoint is that no matter what you do from a surgical standpoint,
two-thirds or more of the patients fail distantly. So no matter what you do, except for the
small subgroup of patients with very small primary lesions that we find early
incidentally, the vast majority of the patients are bound to recur.
Adjuvant therapy was instituted because of the high risk of recurrence. The Lung
Cancer Study Group as well as the Veterans' Group and other groups studied
postoperative chemotherapy, postoperative radiation, postoperative chemotherapy and
radiation and then more recently looking at preoperative therapy, either chemotherapy
or chemoradiotherapy.
Postoperative adjuvant radiation therapy and has a benefit for stages 2 to 3A from
postoperative treatment which increased local control from around 80-85% up to 97%
or better. But the problem is it doesn't translate into increased survival because of the
fact that it is not a local problem that kills the patient. It is the systemic metastases.
Postoperative chemotherapy was looked at extensively and people always bring up the
point that this is CAP chemotherapy and most of the postoperative trials were done with
chemotherapies that are not used any longer because they are inferior and so nobody
really knows now whether postoperative chemotherapy would work or not. There are
some people that are starting to look at that again. But this is one study, again by the
Lung Cancer Study Group looking at Phase 2 and 3 patients, the immunotherapy
control arm which is basically no therapy and did not see significant overall survival
advantage although the disease-free survival was slightly prolonged.
LUNG CANCER Page 4 4
The dismal result of all of these trials led people to try and think of new ways to give
therapy and one of the newer ways was neoadjuvant therapy, also called preoperative
therapy induction therapy. That is giving some kind of therapy prior to definitive local
control measures. Local control measures can either be radiation or surgery but some
of the therapy given before that. So in terms of surgery, neoadjuvant therapy that has
been looked at includes both chemotherapy, chemoradiotherapy and radiation therapy
alone.
There have been a number of preoperative radiation trials looking at preoperative
radiation by itself prior to surgical resection and those haven't shown any benefit. In
fact, even in Pancoast's tumors, the classic treatment that was developed back in the
'50s and '60s was to treat these patients with preoperative radiation therapy then
resection. That is still done in most places now. But the more that we view their data in
terms of the radiation therapy and in looking at intraoperative administration,
preoperative or postoperative, patients did the same no matter what they did so that the
combination of chemotherapy and radiation is important but the order is not necessarily
important. So there are some areas like in St. Louis where they don't get preoperative
therapy, just resect the tumor and give them postoperative therapy and it seems to work
out the same. So preoperative radiation has never really been shown to be definitively
better than the surgery alone with postoperative adjuvant if needed.
So what I want to do is look at some of the data from stage 3A disease and what I mean
by 3A, there are a lot of different types of 3As. They used to be T3N0 which is now 2B
but I am going to talk about them anyway because all of the trials have included those
patients. You can see survival for those patients with T3N0 is 40-50%. Usually closer
to 50% so that is why they have been changed recently to 2B because their survival
really doesn't fit into 3A. Then N2 disease is not even a single disease. An N2 can be a
micrometastatic deposit in one lymph node or it can be bulky disease in multiple station
lymph nodes and their survival is dramatically different as well.
N2 positive patients. If you look at just a single station, that is one lymph node area on
that map that has been involved with tumor and the rest of them were all biopsied and
negative, which gives you about a 35% survival with surgery alone. If you look at a
LUNG CANCER Page 5 5
single station lymph node excluding the subcarinal area which tends to be a somewhat
bilateral station and more difficult and poor prognostically input implications is actually
a better survival if you can eliminate those patients and look at a single station that is
not subcarinal. Up to 45%. Then if you look at patients with more than one area
involved, they go down to 9%. If you again exclude the subcarinal area, it goes up to
22% so you can see that depending on which group you look at you can have
anywhere from less than 10% survival up to 45% survival.
In terms of preoperative or induction therapy, there are a number of advantages that
have been proposed. One of them is that you treat the patient then when primary tumor
size is as small as it can be, malignant metastases are also presumably small and they
would be several months later. You potentially might decrease surgical seating which
may or may not happen. The disadvantages that have been worried about have been
the increased morbidity and mortality from the surgical procedure, which actually has
not turned out to be the case, and delayed primary tumor control. That, as you can see
from the other data, is not the biggest worry. The metastatic disease is what kills the
patient so that has turned out not to be the case.
There are other factors in all of this. When the Lung Cancer Study Group evaluated
their adjuvant trials, one of the primary things and possible reasons for failure was that
the patients didn't tolerate the therapy very well. After undergoing a major operation
and trying to recover from that and then getting chemotherapy. Also patients
psychologically weren't ready to take more chemotherapy because their tumor was
already out and they would give up easily and say, "I don't want anymore
chemotherapy because my tumor is gone. It may not be there anymore and I don't want
anymore." So, preoperative therapies tend to get a higher dose intensity of
chemotherapy than postoperative treatment so that may be one of the key reasons why
the preoperative seemed to be work better.
The two randomized trials looking at preoperative induction chemotherapy prior to
surgical resection in 3A patients. A trial from Barcelona that was published in the New
England Journal a few years ago but these were both small trials. They were both
designed to have 120 patients in them. They were both stopped after 60 patients by
LUNG CANCER Page 6 6
independent review boards that thought that the results were so compelling that it was
not ethical to continue the trials so that each had only 60 patients, here 30 on each
arm, this is 32 and 28. This arm is surgery alone and in Barcelona everybody got
radiation so it's surgery plus radiation versus chemotherapy plus surgery plus radiation.
At the M.D. Anderson they got surgery and chemotherapy prior to surgery. Some of
these patients also got radiation if they had close margins or residual disease in their
lymph nodes.
The chemotherapy was used, and actually again the greatest chemotherapy, in
Barcelona they used cyclophosphamide and cisplatin and in M.D. Anderson they used
cyclophosphamide which was pretty worthless, cyclophosphamide and cisplatin. So
these aren't the kinds of modern regimens that everybody used but still the results are
kind of interesting. The third trial at NIH that never really reached the pinnacle of
significance because it only had about 27 patients in it. It showed the same trend as
these two studies. The multiple number of phase 2 trials that were done also showed
the same kind of results in a nonrandomized fashion.
But if you look at the results here in surgery alone, 85% of resectable in Barcelona, the
median survival time is only 8 months and 10 months in M.D. Anderson. These were
criticized a little bit saying, "Well this is not what we would expect because surgical
time is usually longer than that". But this data has been looked at in terms of the Lung
Cancer Study Group, they included thousands of patients and it is actually consistent
with those controls for this stage of disease.
If you look at chemotherapy prior to therapy, the median survival time is at 26 months
for Barcelona and actually it is up to approximately 60 months or 5 years in M.D.
Anderson. So that is pretty compelling and you would think that this would be a
reasonable thing to do although people now have discounted a lot of this because
some of the data that was collected regarding K-RAS expression in these patients and
both of these groups went back and looked at the K-RAS expression and it was not
balanced between the two groups. So now people just throw out all of the results
because of that which I think is a little bit unfair because that was one of the goals of
the trial was to better the K-RAS.
LUNG CANCER Page 7 7
This is just to show you so you can see it. This survival curve is from the M.D.
Anderson trial and you can see this is surgery alone and you do get a few survivors. It
levels off and this is with chemotherapy prior to surgery. Again, it is leveling out and
there is definitely a big difference between the two curves. It is not one of those curves
where they come down and then they eventually meet down here with the same
survival. It levels off and is definitely different. Then you look at the Barcelona and it is
hard to see but it is the same thing basically. Surgery alone comes down here and then
surgery plus chemotherapy levels off here with a definite distinct difference in survival.
So it looks like it works.
The issue of primary control is one that we brought up. Whether surgery is necessary
in a lot of these patients because they have high dose radiation therapy along with
therapy producing, at least in some patients or advanced patients, similar results. So
the question still is open whether surgery is needed or whether you can do the same
thing with radiotherapy. The only data that we know of at least in earlier stage disease
was small lesions. We know that surgery is far superior to the radiation therapy but in
more advanced lesions like the 3A and 3B, radiotherapy has been fairly equivalent to
surgery but that is because the systemic disease is not controlled. So if you start to
control system disease, then it may end up eventually being proven that surgery has
always been a better local control modality than radiotherapy.
This is a trial that is ongoing but it is looking at the question whether
chemoradiotherapy is better than chemotherapy alone in a preoperative setting. So
they are randomizing patients to receive again unusual chemotherapy, mitomycin,
vizicin and cisplatin and actually get a different chemotherapy cisplatin to this group
and they get radiation therapy to this group as well and do surgery and then follow it up
with either radiation or chemotherapy. So it is kind of an unusual trial but some of the
data in terms of the resectable and the disease free survival so far has been different in
favor of the radiotherapy arm. But this is very early on and the endpoints they looked at
are really endpoints of local control and not necessarily systemic control. So
resectability response ranks a little bit higher with that we would expect with
radiotherapy so there is really no definitive evidence right now that radiotherapy in a
preoperative setting adds anything to chemotherapy. I think it detracts from the
LUNG CANCER Page 8 8
induction therapy because what you really want to know is how well the systemic
disease is going to respond to the chemotherapy and not whether a local tumor can
respond to radiotherapy because that is what really determines survival. So patients
who respond dramatically to chemotherapy are the ones that we can be more
aggressive on in terms of resecting patients and the radiotherapy kind of confounds
that by covering up possible therapy failures.
What I had mentioned before is there are some special considerations about bronchial
obstruction. The fact that if you have an obstructed bronchus and potential infection
beyond that, if you treat patients like that preoperatively, there were a number of deaths
when that first started being done in Toronto. It was responsible for the major part of
the mortality of this kind of approach until it was realized they shouldn't do that. So if
you do have a bronchial obstruction, it is best to either relieve it either with surgery or
radiation or something prior to giving chemotherapy or at least know what the problem
is and cover it with antibiotic.
The major thing is to avoid a lot of hydration in patients perioperatively because the
chemotherapy patients for some reason are more susceptible to having postoperative
noncardiogenic pulmonary edema which is catastrophic in patients who are undergoing
pneumonectomies or big resections. Having had chemotherapy makes them more
susceptible to that so we need to be aware of that and keep their fluids down, quite dry.
Now, just some data regarding 3B which is classically considered unresectable. These
there is a trial Suave 8805 which looked at actually 3A and 3B disease and treated it
with chemotherapy, radiotherapy and there is an induction protocol. The reason this is
important is because this is what is being used now in trials with 3A disease - a similar
regimen. Then they were reevaluated and patients that seemed to have a good
response and were in good shape underwent surgical resection and this was the
results from that trial. The interesting and surprising thing about this trial was that
patients with 3B disease in green had similar survivals as those with 3A with this
induction and treatment. All of these patients underwent more aggressive surgery
where they had to do bilateral lymph node resections and that kind of thing but it kind of
gives you an idea that if you can give a good and early induction chemotherapy
LUNG CANCER Page 9 9
regimen and control systemic metastases and potentially some of the disease that was
thought to be unresectable previously may actually be resectable.
You can't apply this widely. You have to select the patients very carefully and also
stage them very accurately and staging always, at least for most of these trials,
includes mediastinoscopy and PET scans are now being used more widely to try and
also stage patients. But we have also found that a lot of false negatives and false
positives PET scans so that still hasn't been proven. So I think mediastinoscopy is still
required in all of these patients to really know what the risk of recurrence is and how it
should be treated.
M1 disease. Patients with M1 disease occasionally are surgical candidates. It is not
very common but patients particularly with solitary brain metastases and solitary
adrenal metastases have been shown to have a more favorable prognosis at least in
terms of resection candidates than in other patients with more widely metastatic
disease.
These are theories looking at solitary brain mets and survival after surgical resection of
the brain met and the primary lung cancer. You can see the five year survival varied
anywhere from 9% to 34%. So, depending on again the prognosis and the type of
disease that is a reasonable survival. It is more like 3A survival - patients with 3A
disease. So people have used this to go ahead and do primary resections in the
presence of solitary resectable brain metastasis. Again, what we are doing now is to
apply the same approach to 3A disease to this category as well and that is these
patients have about the same survival as the patients with 3A disease, again because
of systemic recurrence. Two-thirds or more of these patients are going to have a
systemic recurrence versus any problem in primary brain site or primary lung site. So
we are now actually taking patients with solitary brain mets, resecting the brain
metastasis, giving them chemotherapy again in an induction setting and then going
back and resecting the primary lung lesion. Again in the same way of trying to improve
survival.
LUNG CANCER Page 10 10
The Memorial data with multiple metastases. They had 231 patients with different
numbers of resections including up to three or four resections of solitary brain
metastases over time. They found that the poor prognostic factors were multiple
metastases, incomplete resection, male gender, infratentorial location was an important
one and the presence of other systemic metastases obviously as well as advanced
age.
This looks at some of the breakdown of the 182 patients with a single resection and
their disease free survival was only five months and their median survival was 14.6
months which is a little bit better than we would expect otherwise. I will show you in a
minute some radiation controlled studies. Actually, patients had two and three
resections and their survival was as long as 42 months so it shows that some patients
who do have limited numbers of metastases can be controlled with surgery alone.
Again, this is not patients who received chemotherapy, so that, potentially, these
patients could have a higher survival if they had chemotherapy as well.
The five year survival is 12.5% which is a little bit less than a lot of the other series but
it shows that it is better than the general systemic metastasis. Two-thirds died again
with systemic disease which again includes chemotherapy.
The RCOG study, 8528 looked at just brain irradiation along with the lung primary
without surgery. You can see that their survival at one year was only 20 to 33%. The
early stage lung cancer. This is kind of group where people have been treated for years
with surgery alone, whether it is stage 1 or stage 2. Anything with disease in the lung
has been treated surgically. As you remember from the survival curve that I showed
earlier, the disease survival can be as low as 30-40% in the stage 2 subgroup so this is
not a really outstanding survival with surgery alone. Again, the standard approach has
been surgery, definitive therapy, radiation therapy if patients refused surgery or
seemed inoperable. The Lung Cancer Study Group has shown with a randomized trial
that the standard surgery should be a lobectomy or pneumonectomy, which is far better
than limited resection including segmentectomies or wedge incisions. The local
recurrence rate goes from about 6% with a lobectomy to as high as 38% with a wedge
LUNG CANCER Page 11 11
resection. The survival is inferior with those lesser resections so most surgeons now
feel lobectomy is the least you should be doing.
Memorial's data from stage 2 disease and shows over 39% five year survival so this
group does not have great survival. We have started thinking that we shouldn't accept
this as acceptable to surgery alone and apply other therapies. What we looked at more
recently is new adjuvant therapies in high risk early stage lung cancer which includes
all of these early lung cancer stages except T1N0. So, if there is a small peripheral
primary that is easily resectable, we still do that. But anything more than that, we
actually now are giving chemotherapy prior to surgical resection. These are the number
of centers that are all participating. This is a multicenter study that we are doing now in
the phase 2 setting. Giving carboplatin and Taxol prior to resection is what all these
places that are participating are giving.
The schema of the trial is to give them two cycles of carboplatin and Taxol and then, if
they have progression of the disease when they go off treatment, if they respond or it is
a stable disease, which everybody has, they go on to surgical resection and then three
postoperative cycles of chemotherapy.
So in the future, we are looking at this poor prognostic early stage disease looking at
getting chemotherapy to T2 and N1 patients as well as T3 disease with complete chest
wall involvement. We limit it to solitary brain or adrenal metastases but Memorial is
actually giving chemotherapy in a preoperative setting to patients who have a solitary
metastases anywhere that is resectable. Then there is 3B and M1 disease that don't
really have protocols but they eventually show some improvement.
A special type of laser bronchoscope can isolate early lesions and we are looking into
screening for lung cancer patients who are at high risk because of a previous lung
cancer or are high risk for other reasons. One of the arms of the screening is doing
monoclonal antibody screening of sputa looks very promising to identify patients with
early changes in the mucosa prior to any overt cancer.
LUNG CANCER Page 12 12
Then another one is looking at this light bronchoscopy which is a type of blue light laser
which will make neoplastic changes or pre-malignant lesions become autofluorescent.
This you can see is the white light bronchoscopy for these areas. You can't really see
much abnormal in these areas but here you can see an extensive brown area and over
here there is more of a brown area here which are potential lesions that might develop
into tumors. So these can be hopefully related early and avoid big resections.
The other part of this treatment would be to use photodynamic therapy to ablate these
mucosal changes before they become a tumor, but it is still experimental. You may
have heard the Japanese reports of using photodynamic therapy for early lung cancer
and these are the kinds of patients that they were attempting to use it in. Problems
occur with patients with solid masses in their lung and other things, people that are
pulmonary cripples who can't have surgery and they want to know about photodynamic
therapy. It is really not something that has been able to treat solid lesions. The best
thing to do is to find any lesion, which in the United States are rare to find this early
lesion and treat it with photodynamic therapy. But it is something that might in the
future, photodynamic therapy is very limited and used rarely to treat patients who have
had high dose radiation and have recurrence in their airway and have a tumor growing
in their airway that is not treatable by any other means, which is rare.
This is Natan trial which is ongoing now which is looking at N2 disease which is again,
there are several questions regarding how best to treat patients with metastases but
they take what you might consider inoperable N2 disease with bulky nodes and definite
N2 involvement to be treated with radiotherapy which used to be the primary treatment
modality. Well, now it is chemoradiotherapy and that is one reason why the trial is
failing recently knowing how randomized it is. The other group is chemotherapy
followed by surgery followed by chemotherapy. So this was to look at this new way of
induction chemotherapy versus the old standard. This started a number of years ago
but may actually be changed to a different trial because they can't really finish it.
The Interview trial from NCI which is now looking at the question of surgery, as a
modality. This randomizes the chemoradiotherapy or local modality therapy in the
definitive option and then the other arm is chemoradiotherapy prior to surgery. This is
LUNG CANCER Page 13 13
slowly accumulating patients. The problem is that people don't like the idea of
randomizing surgery or no surgery so it has gotten a number of patients accrued but it
is not accruing very quickly.
In conclusion, lung cancer basically is systemic disease and it wouldn't surprise me at
all if in three to five years we are treating everybody with chemotherapy up front and
then resecting patients who have residual disease or radiation, depending on what
turns out in those trials. Surgery is the most effective form of local control. I would
guess that I would want chemotherapy up front followed by the most effective local
control which is surgery, if possible. Surgery is often contraindicated in higher stage
disease and selected patients we know have an increased survival. Again, surgery is
something that is to be applied very selectively and only in those patients that are
looked at very carefully and thoroughly prior to undergoing more advanced resections.
LUNG CANCER Page 14 14