TUMORS PREDOMINANTLY OF THE CENTRAL BRONCHI

In this chapter, we use the term central airway lesions to refer
to tumors and premalignant conditions predominantly arising
proximal to the terminal bronchiolar and alveolar epithelium.
Surprisingly, little information is available on the cells from
which central airway carcinomas arise. It might be expected
that since the high risk posed by tobacco exposure has been
known for many years, the earliest changes in the airway lining
cells would by now be well known and the subject of interventional
trials. That this is not the case is because of the
macroscopic invisibility of early carcinoma and its precursor
lesions, to the inaccessibility of the lower airway epithelium to
direct inspection and serial sampling and to the great extent
of the lower airway epithelial surfaces. It has nevertheless been
known for many years that, in the central airways, a distinct
series of changes of variable severity may be seen in the airways
of smokers. Typically these lesions, described in detail later in
the chapter, have been regarded as precursors of squamous carcinoma
but there is evidence that they may also represent precursors
of other histological types of central airway carcinoma,
including undifferentiated large and small cell lung carcinomas
(SCLC).
Central Airway Precursor Lesions Lung carcinoma,
like tumors in other organs, is thought to arise from a stepwise
series of molecular and cellular alterations in precursor
cells. The first studies to demonstrate significant histological
changes in the lower airways of the human population were the
autopsy analyses of Auerbach and colleagues2 performed over
50 years ago. These studies consisted of serial cross sectioning
of tracheobronchial tissue removed at autopsy from 1522 adult
smokers and control nonsmoking patients without invasive carcinoma.
Nearly 42,000 bronchial cross sections were evaluated
and epithelial lesions consisting of atypical cells were found in
93% of sections from current smokers but in only 1.2% of sections
from individuals who were never-smokers. Cellular abnormalities
in smokers’ lungs were often multifocal and were
independent of age, sex, place of residence, and the presence
of pneumonia. A significant number of individuals who were
former smokers in this study also had cellular abnormalities,
whereas they were rare in individuals who never smoked. 3
Sputum Cytology and Risk Assessment Although
these studies documented the prevalence of cellular lesions
in the lower airways at a single time point in smokers who
did not have carcinoma, prospective confirmation that these
lesions represent precursor lesions for carcinoma has been
problematic. A major difficulty is a lack of any creditable intervention
strategy that can be offered to individuals who are
found to have dysplasia. In other organs, preinvasive lesions
can be excised and this has been a successful strategy in cervix
and colon where screening has proven successful in reducing
the incidence and mortality of invasive carcinoma. 4,5
The most analogous studies in the lung are the lung cancer
screening trials of the 1970s and 1980s. In these studies, the
effectiveness of sputum cytology as a screening tool was evaluated
in 21,000 subjects at Sloan-Kettering Medical Center
and the Mayo Clinic. Spontaneous sputa were evaluated by
conventional cytological criteria for the presence of carcinoma.
Cytology proved to be an insensitive tool for detection of lung
cancer. Only 41% of subjects were sputum positive among
those who had carcinoma at an initial ( prevalence ) screen and
17% with carcinoma in subsequent ( incidence ) screenings. In
the Mayo Clinic project, no effect on mortality could be demonstrated
by chest radiograph or sputum cytology performed
quarterly for 6 years. The only positive trend resulting from
these interventions was a slightly improved median survival
for screen-detected stage I carcinomas in comparison to those
detected outside the screened group, a difference that was not
statistically significant. 6
The failure of sputum and radiograph screening to reduce
mortality from lung cancer has largely been attributed to overdiagnosis
bias . 7,8 Overdiagnosis bias refers to the detection of
tumors through screening that, even if undetected, would not
have affected mortality. It implies either that death may occur
from other causes before any mortality effect from screened tumors
can occur, or that the detected tumors were indolent and
would not have affected survival. The lack of impact of screening
on overall mortality has lead to the proposal that small
screen-detected tumors are biologically distinct from advanced
tumors that are responsible for the high mortality in lung cancer.
Small screen-detected tumor are postulated to have lower
growth rates and different epidemiological associations than
those of advanced lethal tumors and, moreover, may not be
precursors of advanced lesions, 9 as is widely believed. The
morphological and biological features of screen-detected central
airway tumors have been analyzed by Mayo pathologists. 10
Only 86% of the screened tumors were unequivocally invasive
with 14% regarded as preinvasive by one or more of the
pathologists on the reviewing panel. Inclusion of preinvasive
lesions in the outcome analysis may have contributed to overdiagnosis
bias in the Mayo Lung Project and provides some
biological basis for the outcome of the trial.
One crucial difference between the lung cancer screening
trials and screening practices for colon and cervix is that
the latter target preinvasive lesions rather than fully developed
invasive carcinomas. A similar approach has not been fully
explored in lung cancer. The significance of less than fully malignant
cells in the sputum (Fig. 22.1) has recently been evaluated.
11,12 These studies have shown that cytologic atypia is a
marker for increased lung cancer risk. The association of sputum
atypia with lung cancer increases with the severity of the
atypia, with short interval between sputum collection and with
squamous tumor histology. Higher-grade cytological changes
thus seem to arise from late events in the central airways.
Whether sputum atypia can be a useful indicator for the presence
of premalignant dysplasia in the central airways, whether
the presence, grade, and extent of dysplasia predicts invasive
carcinoma, and whether meaningful interruption or delay in
neoplastic progression is possible when premalignant changes
are identified are questions currently being investigated using
new methods for visualizing and directly sampling bronchial
mucosa for histological assessment.