Tobacco smoking is by far the main contributor to lung cancer. Cigarette smoke contains at least 73 known carcinogens, including benzo["a"]pyrene, NNK, 1,3-butadiene and a radioactive isotope of polonium, polonium-210. Across the developed world, 90% of lung cancer deaths in men during the year 2000 were attributed to smoking (70% for women). Smoking accounts for about 85% of lung cancer cases.

Passive smoking—the inhalation of smoke from another's smoking—is a cause of lung cancer in nonsmokers. A passive smoker can be defined as someone living or working with a smoker. Studies from the US, Europe and the UK have consistently shown a significantly increased risk among those exposed to passive smoke. Those who live with someone who smokes have a 20–30% increase in risk while those who work in an environment with secondhand smoke have a 16–19% increase in risk. Investigations of sidestream smoke suggest it is more dangerous than direct smoke. Passive smoking causes about 3,400 deaths from lung cancer each year in the USA.

Marijuana smoke contains many of the same carcinogens as those in tobacco smoke. However, the effect of smoking cannabis on lung cancer risk is not clear. A 2013 review did not find an increased risk from light to moderate use. A 2014 review found that smoking cannabis doubled the risk of lung cancer.

In most series, LCLC's comprise between 5% and 10% of all lung cancers.

According to the Nurses' Health Study, the risk of large cell lung carcinoma increases with a previous history of tobacco smoking, with a previous smoking duration of 30 to 40 years giving a relative risk of approximately 2.3 compared to never-smokers, and a duration of more than 40 years giving a relative risk of approximately 3.6.

Another study concluded that cigarette smoking is the predominant cause of large cell lung cancer. It estimated that the odds ratio associated with smoking two or more packs/day for current smokers is 37.0 in men and 72.9 in women.

Taken as a class, long-term survival rates in BAC tend to be higher than those of other forms of NSCLC. BAC generally carries a better prognosis than other forms of NSCLC, which can be partially attributed to localized presentation of the disease. Though other factors might play a role. Prognosis of BAC depends upon the histological subtype and extent at presentation but are generally same as other NSCLC.

Recent research has made it clear that nonmucinous and mucinous BACs are very different types of lung cancer. Mucinous BAC is much more likely to present with multiple unilateral tumors and/or in a unilateral or bilateral pneumonic form than nonmucinous BAC. The overall prognosis for patients with mucinous BAC is significantly worse than patients with nonmucinous BAC.

Although data are scarce, some studies suggest that survival rates are even lower in the mixed mucinous/non-mucinous variant than in the monophasic forms.

In non-mucinous BAC, neither Clara cell nor Type II pneumocyte differentiation appears to affect survival or prognosis.

This type of cancer occurs most often in Caucasians between 60 and 80 years of age, and its rate of incidence is about twice as high in males as in females. There are roughly 1,500 new cases of MCC diagnosed each year in the United States, as compared to around 60,000 new cases of melanoma and over 1 million new cases of nonmelanoma skin cancer. MCC is sometimes mistaken for other histological types of cancer, including basal cell carcinoma, squamous cell carcinoma, malignant melanoma, lymphoma, and small cell carcinoma, or as a benign cyst. Researchers believe that exposure to sunlight or ultraviolet light (such as in a tanning bed) may increase the risk of developing this disease. Similar to melanoma, the incidence of MCC in the US is increasing rapidly.

Immunosuppression can profoundly increase the odds of developing Merkel-cell carcinoma. Merkel-cell carcinoma occurs 30 times more often in people with chronic lymphocytic leukemia and 13.4 times more often in people with advanced HIV as compared to the general population; solid organ transplant recipients have a 10-fold increased risk compared to the general population.

The true incidence, prevalence, and mortality of GCCL is generally unknown due to a lack of accurate cancer data on a national level. It is known to be a very rare tumor variant in all populations examined, however. In an American study of a database of over 60,000 lung cancers, GCCL comprised between 0.3% and 0.4% of primary pulmonary malignancies, with an age-adjusted incidence rate of about 3 new cases per million persons per year. With approximately 220,000 total lung cancers diagnosed in the US each year, the proportion suggests that approximately 660 and 880 new cases are diagnosed in Americans annually.

However, in a more recent series of 4,212 consecutive lung cancer cases, only one (0.024%) lesion was determined to be a "pure" giant-cell carcinoma after complete sectioning of all available tumor tissue. While some evidence suggests GCCL may have been considerably more common several decades ago, with one series identifying 3.4% of all lung carcinomas as giant-cell malignancies, it is possible that this number reflect

Most published case series and reports on giant cell-containing lung cancers show that they are diagnosed much more frequently in men than they are in women, with some studies showing extremely high male-to-female ratios (12:1 or more). In a study of over 150,000 lung cancer victims in the US, however, the gender ratio was just over 2:1, with women actually having a higher relative proportion of giant-cell cancers (0.4%) than men (0.3%).

Giant-cell carcinomas have been reported to be diagnosed in a significantly younger population than all non-small-cell carcinomas considered as a group. Like nearly all lung carcinomas, however, GCCs are exceedingly rare in very young people: in the US SEER program, only 2 cases were recorded to occur in persons younger than 30 years of age between 1983 and 1987. The average age at diagnosis of these tumors has been estimated at 60 years.

The vast majority of individuals with GCCL are heavy smokers.

Although the definitions of "central" and "peripheral" can vary between studies, GCCL are consistently diagnosed much more frequently in the lung periphery. In a review of literature compiled by Kallenburg and co-workers, less than 30% of GCCLs arose in the hilum or other parts of the "central" pulmonary tree.

A significant predilection for genesis of GCCL in the upper lobes of victims has also been postulated.

When BAC recurs after surgery, the recurrences are local in about three-quarters of cases, a rate higher than other forms of NSCLC, which tends to recur distantly.

15% of lung cancers in the US are of this type. Small cell lung cancer occurs almost exclusively in smokers; most commonly in heavy smokers and rarely in non-smokers.

LCC is, in effect, a "diagnosis of exclusion", in that the tumor cells lack light microscopic characteristics that would classify the neoplasm as a small-cell carcinoma, squamous-cell carcinoma, adenocarcinoma, or other more specific histologic type of lung cancer.

LCC is differentiated from small-cell lung carcinoma (SCLC) primarily by the larger size of the anaplastic cells, a higher cytoplasmic-to-nuclear size ratio, and a lack of "salt-and-pepper" chromatin.

Outdoor air pollutants, especially chemicals released from the burning of fossil fuels, increase the risk of lung cancer. Fine particulates (PM) and sulfate aerosols, which may be released in traffic exhaust fumes, are associated with slightly increased risk. For nitrogen dioxide, an incremental increase of 10 parts per billion increases the risk of lung cancer by 14%. Outdoor air pollution is estimated to account for 1–2% of lung cancers.

Tentative evidence supports an increased risk of lung cancer from indoor air pollution related to the burning of wood, charcoal, dung or crop residue for cooking and heating. Women who are exposed to indoor coal smoke have about twice the risk and a number of the by-products of burning biomass are known or suspected carcinogens. This risk affects about 2.4 billion people globally, and is believed to account for 1.5% of lung cancer deaths.

Reliable comprehensive incidence statistics for c-SCLC are unavailable. In the literature, the frequency with which the c-SCLC variant is diagnosed largely depends on the size of tumor samples, tending to be higher in series where large surgical resection specimens are examined, and lower when diagnoses are based on small cytology and/or biopsy samples. Tatematsu "et al." reported 15 cases of c-SCLC (12%) in their series of 122 consecutive SCLC patients, but only 20 resection specimens were examined. In contrast, Nicholson "et al." found 28 c-SCLC (28%) in a series of 100 consecutive resected SCLC cases. It appears likely, then, that the c-SCLC variant comprises 25% to 30% of all SCLC cases.

As the incidence of SCLC has declined somewhat in the U.S. in recent decades, it is likely that c-SCLC has also decreased in incidence. Nevertheless, small cell carcinomas (including the c-SCLC variant) still comprise 15–20% of all lung cancers, with c-SCLC probably accounting for 4–6%. With 220,000 cases of newly diagnosed lung cancer in the U.S. each year, it can be estimated that between 8,800 and 13,200 of these are c-SCLC.

In a study of 408 consecutive patients with SCLC, Quoix and colleagues found that presentation as a solitary pulmonary nodule (SPN) is particularly indicative of a c-SCLC — about 2/3 of their SPN's were pathologically confirmed to be c-SCLC's containing a large cell carcinoma component.

Squamous cell carcinoma (SCC) of the lung is more common in men than in women. It is closely correlated with a history of tobacco smoking, more so than most other types of lung cancer. According to the Nurses' Health Study, the relative risk of SCC is approximately 5.5, both among those with a previous duration of smoking of 1 to 20 years, and those with 20 to 30 years, compared to never-smokers. The relative risk increases to approximately 16 with a previous smoking duration of 30 to 40 years, and approximately 22 with more than 40 years.

Giant-cell lung cancers have long been considered to be exceptionally aggressive malignancies that grow very rapidly and have a very poor prognosis.

Many small series have suggested that the prognosis of lung tumors with giant cells is worse than that of most other forms of non-small-cell lung cancer (NSCLC), including squamous cell carcinoma, and spindle cell carcinoma.

The overall five-year survival rate in GCCL varies between studies but is generally considered to be very low. The (US) Armed Forces Institute of Pathology has reported a figure of 10%, and in a study examining over 150,000 lung cancer cases, a figure of 11.8% was given. However, in the latter report the 11.8% figure was based on data that included spindle cell carcinoma, a variant which is generally considered to have a less dismal prognosis than GCCL. Therefore, the likely survival of "pure" GCCL is probably lower than the stated figure.

In the large 1995 database review by Travis and colleagues, giant-cell carcinoma has the third-worst prognosis among 18 histological forms of lung cancer. (Only small-cell carcinoma and large-cell carcinoma had shorter average survival.)

Most GCCL have already grown and invaded locally and/or regionally, and/or have already metastasized distantly, and are inoperable, at the time of diagnosis.

EMECL is extremely rare, with only a handful of cases reported in the literature.

In the lung, two salivary gland-like carcinomas, mucoepidermoid carcinoma and adenoid cystic carcinoma, while extremely uncommon, occur far more often than does EMECL.

The prognosis of EMECL is relatively good, and considerably better than most other forms of NSCLC. The skull and dura are possible sites for metastasis from pulmonary EMC. The MIB-1 index is a predictive marker of malignant potential.

MCACL has a much more favorable prognosis than most other forms of adenocarcinoma and most other NSCLC's. Cases have been documented of continued growth of these lesions over a period of 10 years without symptoms or metastasis. The overall mortality rate appears to be somewhere in the vicinity of 18% to 27%, depending on the criteria that are used to define this entity.

All in all, small-cell carcinoma is very responsive to chemotherapy and radiotherapy, and in particular, regimens based on platinum-containing agents. However, most people with the disease relapse, and median survival remains low.

In "limited-stage" disease, median survival with treatment is 14–20 months, and about 20% of patients with limited-stage small-cell lung carcinoma live 5 years or longer. Because of its predisposition for early metastasis, the prognosis of SCLC is poor, with only 10% to 15% of patients surviving 3 years.

The prognosis is far more grim in "extensive-stage" small-cell lung carcinoma; with treatment, median survival is 8–13 months; only 1–5% of patients with extensive-stage small-cell lung carcinoma treated with chemotherapy live 5 years or longer.

The prognosis of patients with FA as a whole is considered to be better than that of most other forms of non-small cell carcinoma, including biphasic pulmonary blastoma.

Adenocarcinoma of the lung is currently the most common type of lung cancer in "never smokers" (lifelong non-smokers). Adenocarcinomas account for approximately 40% of lung cancers. Historically, adenocarcinoma was more often seen peripherally in the lungs than small cell lung cancer and squamous cell lung cancer, both of which tended to be more often centrally located. However, recent studies suggest that the "ratio of centrally-to-peripherally occurring" lesions may be converging toward unity for both adenocarcinoma and squamous cell carcinoma.

In terms of case numbers, the estimated 8,800 to 13,200 c-SCLC cases occurring annually in the U.S. makes this disease roughly comparable in incidence to Hodgkin's Disease (8,500), testicular cancer (8,400), cervical cancer (11,300), and cancers of the larynx (12,300). However, these four "better-known" cancers all have exceptionally high (85%-95%) cure rates. In contrast, less than 10% of c-SCLC patients will be cured, and thus the number of annual cases of c-SCLC is a reasonable approximation of the annual number of deaths. Therefore, given the significant incidence and mortality attributable to this malignancy, (see "Prognosis and survival") it is arguably critical to better understand these aggressive lesions so specific strategies for their management can be rationally designed.

However, as patients with tumors containing mixtures of histological subtypes are usually excluded from clinical trials, the properties of multiphasic tumors like c-SCLC are much less well understood than those of monophasic tumors. C-SCLC contains both SCLC and NSCLC by definition, and since patients with SCLC and NSCLC are usually treated differently, the lack of good data on c-SCLC means there is little evidence available with which to form consensus about whether c-SCLC should be treated like SCLC, NSCLC, or uniquely.

Lung cancers have been historically classified using two major paradigms. Histological classification systems group lung cancers according to the appearance of the cells and surrounding tissues when they are viewed under a microscope. Clinical classification systems divide lung cancers into groups based on medical criteria, particularly their response to different treatment regimens.

Before the mid-1900s, lung cancer was considered to be a single disease entity, with all forms treated similarly. In the 1960s, small cell lung carcinoma (SCLC) was recognized as a unique form of lung cancer, based both on its appearance (histology) and its clinical properties, including much greater susceptibility to chemotherapy and radiation, more rapid growth rate, and its propensity to metastasize widely early on in its course. Since then, most oncologists have based patient treatment decisions on a dichotomous division of lung cancers into SCLC and non-small cell lung carcinomas (NSCLC), with the former being treated primarily with chemoradiation, and the latter with surgery.

An explosion of new knowledge, accumulated mainly over the last 20 years, has proved that lung cancers should be considered an extremely heterogeneous family of neoplasms with widely varying genetic, biological, and clinical characteristics, particularly their responsiveness to the large number of newer treatment protocols. Well over 50 different histological variants are now recognized under the 2004 revision of the World Health Organization ("WHO-2004") typing system, currently the most widely used lung cancer classification scheme. Recent studies have shown beyond doubt that the old clinical classification paradigm of "SCLC vs. NSCLC" is now obsolete, and that correct "subclassification" of lung cancer cases is necessary to assure that lung cancer patients receive optimum management.

Approximately 98% of lung cancers are carcinoma, which are tumors composed of cells with epithelial characteristics. LCLC's are one of 8 major groups of lung carcinomas recognized in WHO-2004:

- Squamous cell carcinoma

- Small cell carcinoma

- Adenocarcinoma

- Large cell carcinoma

- Adenosquamous carcinoma

- Sarcomatoid carcinoma

- Carcinoid tumor

- Salivary gland-like carcinoma

Although reliable and comprehensive incidence statistics are nonexistent, LCLC-RP is a rare tumor, with only a few hundred cases described in the scientific literature to date. LCLC's made up about 10% of lung cancers in most historical series, equating to approximately 22,000 cases per year in the U.S. Of these LCLC cases, it is estimated that about 1% will eventually develop the rhabdoid phenotype during tumor evolution and progression. In one large series of 902 surgically resected lung cancers, only 3 cases (0.3%) were diagnosed as LCLC-RP. In another highly selected series of large-cell lung carcinoma cases, only 4 of 45 tumors (9%) were diagnosed as the rhabdoid phenotype using the 10% criterion, but another 10 (22%) had at least some rhabdoid cell formation. It appears likely, therefore, that LCLC-RP probably comprises between 0.1% and 1.0% of all lung malignancies.

Similar to nearly all variants of lung carcinoma, large cell lung carcinoma with rhabdoid phenotype appears to be highly related to tobacco smoking. It also appears to be significantly more common in males than in females.

Accurate incidence statistics on MCACL are unavailable. It is a very rare tumor, with only a few dozen cases reported in the literature to date.

In the few cases described in the literature to date, the male-to-female ratio is approximately unity, and right lung lesions occurred twice as commonly as left lung lesions. Approximately 2/3 of cases have been associated with tobacco smoking. Cases have been reported in patients as young as 29.

Adenosquamous lung carcinoma (AdSqLC) is a biphasic malignant tumor arising from lung tissue that is composed of at least 10% by volume each of squamous cell carcinoma (SqCC) and adenocarcinoma (AdC) cells.

Smoking is the most important risk factor for laryngeal cancer. Death from laryngeal cancer is 20 times more likely for heaviest smokers than for nonsmokers. Heavy chronic consumption of alcohol, particularly alcoholic spirits, is also significant. When combined, these two factors appear to have a synergistic effect.

Some other quoted risk factors are likely, in part, to be related to prolonged alcohol and tobacco consumption. These include low socioeconomic status, male sex, and age greater than 55 years.

People with a history of head and neck cancer are known to be at higher risk (about 25%) of developing a second cancer of the head, neck, or lung. This is mainly because in a significant proportion of these patients, the aerodigestive tract and lung epithelium have been exposed chronically to the carcinogenic effects of alcohol and tobacco. In this situation, a field change effect may occur, where the epithelial tissues start to become diffusely dysplastic with a reduced threshold for malignant change. This risk may be reduced by quitting alcohol and tobacco.

Squamous-cell carcinoma (SCC) of the lung is a type of non-small-cell lung carcinoma and is more common in men than in women. It is closely correlated with a history of tobacco smoking, more so than most other types of lung cancer. According to the Nurses' Health Study, the relative risk of SCC is approximately 5.5, both among those with a previous duration of smoking of 1 to 20 years, and those with 20 to 30 years, compared to never-smokers. The relative risk increases to approximately 16 with a previous smoking duration of 30 to 40 years, and approximately 22 with more than 40 years.