Ken’s Pathology Guide

Adenocarcinoma of the Lung

Epidemiology:

Risk factors:

Smoking (75% in smokers)
High-dose ionizing radiation (uranium)
Asbestos (particularly in smokers)
Radon

Most common type of lung cancer in women and nonsmokers

Although 75% of adenocarcinomas are found in smokers

Common sites:

Peripheral location in the lung
Involve the pleura commonly

Gross features:

Gray-yellow
Cavitation is rare
Broncholoalveolar carcinoma (now AIS and invasive mucinous carcinoma):

Almost always in periphery of lung
Mucinous type is more likely to be multicentric and may grossly show mucin
Often multiple diffuse, soft, gray nodules coalescing to form a pneumonia-like consolidation
May have a mucinous, gray translucence

Histologic features:

glandular differentiation
mucin production (80%)
most show mixed patterns (formerly called “mixed subtype”)
classification of invasive ADC by predominant pattern:

lepidic (formerly mixed subtype with nonmucinous BAC)
acinar
papillary
solid (poor prognosis)

poorly differentiated
If 100% solid, need to see intracellular mucin

should be present within at least 5 cells in at least 2 high-power fields
mucicarmine or PASD positive

micropapillary (poor prognosis)

variants:

invasive mucinous adenocarcinoma (formerly mucinous BAC)
colloid adenocarcinoma
fetal adenocarcinoma
enteric adenocarcinoma

types / growth patterns:

acinar
often bronchioloalveolar pattern of spread at the periphery of the tumor
papillary type shows papillary architecture similar to other papillary carcinomas

>25% papillary architecture is an unfavorable prognostic factor

mixed

Adenocarcinoma in situ (AIS) (<= 3 cm by definition) (formerly bronchioloalveolar BAC type):

pure bronchioloalveolar growth pattern with no evidence of stromal, vascular, or pleural invasion
lepidic pattern – growth along preexisting structures with preservation of alveolar architecture
often grows around the edge of a focal scar

smaller scar is better prognosis

usually non-mucinous (rarely mucinous)
mucinous subtype (rare – most are now classified as invasive mucinous adenocarcinoma):

tend to be multicentric (?now invasive mucinous adenocarcinoma)
tall columnar cells
abundant apical cytoplasmic mucin
small basally oriented nuclei
airspaces often filled with mucin

non-mucinous subtype:

tend to be solitary
Clara cells or tye II pneumocytes
cuboidal cells with a hobnail or saw-toothed appearance often
nuclear inclusions (50% of cases)

PAS positive
Surfactant apoprotein positive immunohistochemically
EM – network of 40nm branching microtubules thought to arise from inner nuclear membrane

Mixed mucinous and non-mucinous type
note that metastatic adenocarcinomas to the lung can mimic this pattern, particularly mucinous subtype

minimally invasive adenocarcinoma (MIA) (<= 3 cm by definition)

lepidic pattern predominantly, with <= 5mm invasion
non-mucinous predomninantly (rarely mucinous)

invasive mucinous adenocarcinoma (formerly mucinous BAC)

goblet or columnar cells
abundant intracellular mucin

note that mucin production does not make the diagnosis – needs to be goblet or columnar cells

patterns may be seen similar to non-mucinous adenocarcinoma, but not solid:

lepidic
acinar
papillary
micropapillary

criteria to differentiate from mucinous AIS or mucinous MIA (only need one of these criteria to call invasive mucinous adenocarcinoma):

size > 3cm
invasion > 0.5cm
multiple nodules

or lack of a circumscribed border with military spread into adjacent lung parenchyma

strong tendency to multicentricity involving multiple lobes and bilateral lungs (?aerogenous spread)
strong association with KRAS mutations (~76%)

mixed mucinous and non-mucinous adenocarcinoma

needs at least 10 % of each component

fetal adenocarcinoma:

tubules composed of glycogen-rich, nonciliated cells resembling fetal lung tubues
subnuclear vacuoles are common and characteristic
squamoid morules may be seen
most are low grade with a favourable outcome
high-grade tumours occur
may be mixed, should be classified according to the predominant component
beta-catenin gene mutations (detected by nuclear and cytoplasmic IHC staining

mucinous (“colloid”) adenocarcinoma

extracellular mucin in abundant pools
distend alveolar spaces with destruction of their walls
mucin pools contain clusters of mucin-secreting tumour cells (may be inconspicuous)
more often seen along with other histologic patterns, rather than as a pure pattern

colloid adenocarcinoma classification is used if it is the predominant component

mucinous cystadenocarcinoma

rare
should be classified as colloid adenocarcinoma with cystic changes

signet ring adenocarcinoma
clear cell adenocarcinoma

Immunophenotype:

Marker:	Sensitivity:	Specificity:
PAS pos & PASD neg	80%
AE1/AE3
CAM 5.2
EMA
CEA
CK7	Most	Positive in up to 30% of squamous cell carcinomas
TTF-1*	75% of invasive (Negative in most invasive mucinous adenocarcinomas) (less common in solid pattern)	Also positive in: Small cell CA Large cell NEC Carcinoids Thyroid CA
Napsin A	Comparable to TTF-1	Sometimes expressed in RCC
CK20 (neg)*	Most Positive in ~54% of invasive mucinous adenocarcinoma
Thyroglobulin (neg)
P63 (neg)	May be positive in up to 30% of adenoCA (Focal / weak)
P40 (neg)	More specific than p63, but can be seen in adenoCA (focal / weak)
34Beta12 (CK903) (neg)	Frequently positive in solid pattern adenoCA
EGFR mutation specific antibodies (not recommended)	76% for L858R 60% or less for exon 19 deletions	High
ALK (5A4 or D5F3 clones) FDA approved Ventana ALK (D5F3) CDx Assay as a companion diagnostic for crizotinib May be used to screen for ALK rearrangement with confirmation by FISH before initiating ALK-targeted therapy
ROS1	“Robust screening tool”	Suboptimal Positive results need to be confirmed by another technique
BRAF (V600E)	Variable	Variable
RET Not well evaluated yet
PD-L1 Preliminary studies show an encouraging predictive association with response to PD1 antibody therapy. Further assay validation is needed. Antibody and scoring cutoffs have not been established. Currently no validated, commercially available assay for PD-L1 expression that is predictive of outcome

· *mucinousBAC is typically CK20 pos and TTF-1 neg

· Biopsies:

· Cases positive for an adenocarcinoma marker (i.e. TTF1 or napsin A) and/or mucin, with a negative squamous marker (i.e. p40 or p63) should be classified as NSCC, favour adenocarcinoma

· If an adenocarcinoma marker such as TTF-1 is positive, the tumour should be classified as NSCC, favour adenocarcinoma, regardless of any expression of squamous markers.

· If intracytoplasmic mucin can be demonstrated in a poorly-differentiated NSCC with a mucin stain in at least two tumour cells in the biopsy (and in the absence of IHC markers for adenocarcinoma or squamous cell carcinoma) the diagnosis of adenocarcinoma is appropriate

· If ADC marker (i.e. TTF-1) and/or mucin +ve, and SqCC marker neg, can call NSCC, favour ADC

· Resections:

· Undifferentiated carcinomas that express pneumocyte immunohistochemical markers, and/or mucin expression, are classified as adenocarcinoma (previously large cell carcinomas)

· Although primary lung adenoCA can be TTF-1 negative, additional IHC studies may be helpful to exclude metastasis (i.e. CDX2, CK20, ER, or PR)

Molecular features:

> 50% contain an identifiable genetic alteration, some of which can be targeted by a specific therapeutic inhibitor
KRAS mutations (20-40%) (extremely rare in other histologic types)

Mutations at codon 12
Mutually exclusive with EGFR and ALK alterations typically
Methylation often present in early stages of cancer
RAS is negatively regulated by the catalytic reaction of RAS GTPase-activating proteins (RAS-GAPs), which enhances RAS GTPase activity [193]
Epidemiology:

More frequent in smokers (30%) than non-smokers (5%)
smokers186, 192,194
the incidenceof KRAS mutations increased as smoke exposure increased193
Recently, we evaluated the frequency of KRAS mutations inlung adenocarcinomas from nearly 500 patients, of whom 17%had never smoked cigarettes (30). We noted that KRAS mutations occurred in 22% of the overall population andin15% oflung adenocarcinomas from never-smokers.194
KRAS transition mutations (G/A) were more common in patients who hadnever smoked cigarettes. In contrast, transversion mutations(G/T or G/C) were more common in former/current smokers194
Thus, unlike EGFR mutations, which occur more frequently in tumors from never-smokers (31), KRAS tumorstatus cannot be easily predicted on the basis of smoking historyalone.194
KRAS mutations occur predominantlyin Caucasian patients rather than in East Asians; theincidence of KRAS mutation is ∼30% in Caucasian patients and ∼10% in East Asian patients with adenocarcinoma193
African Americans aresignificantly less likely to harbor EGFR mutations (2%),whereas the frequency of KRAS mutations (23%) does notdiffer from that in Caucasians [55].

Gross:

Hilar location frequently

Histology:

Invasive mucinous adenocarcinoma (formerly mucinous BAC) show a very strong correlation with KRAS mutations
Overall histology cannot be relied upon for predicting KRAS mutation

Prognosis:

Their impact on overall survival remains controversial.
Soh et al. found that gene dosage is associated with prognostic impact [43].193
of 237 lung adenocarcinoma patients,six patients who harbored both a KRAS mutation and CNGhad significantly shortened survival (P=0.04) [43].193
KRAS mutations could be associated with Progressive Disease status.200

Prediction of response to therapy:

No demonstrably efficacious treatments
Lack of response to EGFR inhibitors

the decision to treat with an EGFR TKI can no longer be made without an EGFR result, and the role of KRAS testing in this context has diminished.

because KRAS and EGFR mutations are mutually exclusive, a rapid and inexpensive KRAS assay may be performed initially to exclude KRAS-mutated tumors from EGFR mutation testing as part of an algorithm designed to maximize testing efficiency, provided that the sample is sufficient to perform the KRAS test without sacrificing EGFR and ALK testing, and that the totality of clinically relevant molecular results can be obtained within the target TAT.

absence of a KRAS mutation does not add clinically useful information to the EGFR mutation result and should not be used as a determinant of EGFR TKI therapy.
it is notclear whether the response to EGFR-TKIs differs betweentumors harboring KRAS mutations and those harboringneither KRAS nor EGFR mutations193
Objective response to EGFR TKI can be seen in 0% to 3% of patients with KRAS mutations and 26% of patients with KRAS wild type.
In 2005, we examined the tumor KRASstatus in patients with NSCLC who had been treated with eitherdrug as a single agent (2). Collectively, none of 21 patients whosedisease responded radiographically had KRASmutations, while 9of 38 patients with refractory disease had KRAS mutations (P 50.02). Multiple other groups have reported similar ﬁndings (Table1)194
KRAS mutations are mutually exclusive to EGFR mutations

Of note, the response rate for patients treated with carboplatin and paclitaxel did not differ signiﬁcantly by KRASmutation status (26% versus 23%)194

EGFR mutations (20%)

The EGFR gene is located on the short arm of chromosome 7 (7p11.2) and encodes a 170-kDa type I transmembrane growth factor receptor with TK activity (10)218
Strongly associated with clinical response to EGFR tyrosine kinase inhibitors such as gefitinib, erlotinib, afatinib, small-molecule (tyrosine) kinase inhibitors

Exon 18 Gly719#

Epidemiologic associations:

never smokers or light smokers (never smoked, 47% compared with ever smoked, 7%),
predominantly in female

(male, 10% compared with female,38%)

ethnicity:

10-20% for caucasians, 50% for East Asians
East Asians, 32% compared with Caucasians, 7%, African Americans aresignificantly less likely to harbor EGFR mutations (2%),whereas the frequency of KRAS mutations (23%) does notdiffer from that in Caucasians [55].

Histologic associations:

TTF-1 positive (close relationship)
much more common in adenocarcinoma than other histologic types
lepidic, papillary, or acinar histology (more likely)

less likely in poorly-differentiated, mucinous, or solid histology (but still found with significant frequeny in all grades)
histologic subtype should not be used to determine which samples should be tested for EGFR

very infrequent in squamous cell carcinoma (but some are reported)
very infrequent in invasive mucinous adenocarcinoma (~3%)
not detected in small cell carcinoma nor in carcinoids
rare reports in salivary gland-type tumours, large cell carcinomas, sarcomatoid carcinomas, large cell neuroendocrine carcinoma

EGFR Mutation correlates with copy number changes
Prognosis

Several studies have reported that patients with untreated NSCLC with EGFR mutations have a more favourableprognosis than patients with wild-type EGFR.201
In the BR.21 trial, untreated patients in the placebo group with EGFRmutant tumours had a median survival of 8·3 months(range 3·3–11·1) compared with 3·3 months (2·5–6·8)for patients with wild-type EGFR.1(201)
in the TRIBUTE trial: Patients with mutated tumors (29/228=13%) had signiﬁcantly better clinicaloutcomes in all assessed end points, including survival, compared with patients with wildtype EGFR tumors, regardlessof the therapygiven218
treatment with erlotinib pluschemotherapy versus chemotherapy alone inadvanced NSCLC,218
EGFR mutations were detected in 13% of tumors and were associated with longer survival,irrespective of treatment (P .001).220
We analyzed 397 patients with lung adenocarcinoma who underwent potentially curative pulmonary resection. Univariate analysis showed that patients with EGFR mutations survived for a longer period than those without mutations (p 0.0046).221
Multivariate analysis using the Cox proportional hazards model revealed that smoking status (p 0.0310) and disease stage (p 0.0001) were independent prognostic factors. However,none of the gene mutations was independent prognostic factors(EGFR, p 0.3225; KRAS, p 0.8500; TP53, p 0.3191).221

Prediction of response to therapy:

Benefits for first-line treatment with gefitinib, erlotinib, and afatinib
Gefitinib, afatinib, and erlotinib response:

EGFR molecular testing should be used to select patients for EGFR-targeted TKI therapy

Testing should not be chosen based on clinical characteristics

RR 68% (response rate)

these patients almost invariably experience recurrence or progression while on treatment after a median of 8 to 16 months, a clinical phenomenon termed acquired resistance

PFS 12 months

Significantly longer than those receiving platinum-based chemotherapy

OS not improved compared to chemotherapy so far
EGFR wild-type tumours respond better to conventional chemotherapy than to EGFR TKI

Platinum-based chemotherapy:

EGFR wild-type tumours respond better to conventional chemotherapy than to EGFR TKI

In 2004, the results of the National Cancer Institute of Canada Clinical Trials Group BR.21study led to the global approval of erlotinib asthe ﬁrst targeted therapy for advanced NSCLC(6, 7).218
At approximately the same time, two groups (8, 9) reported the discovery of mutations in the tyrosine kinase (TK) domain ofthe EGFR gene in NSCLC. Furthermore, thepresence of these mutations appeared to correlate with sensitivity to the EGFR inhibitor geﬁ-tinib.218
which mutations are responsive / non-responsive?

L858R and exon 19 deletion mutations show the greatest sensitivity to small-molecule epidermal growth factor receptor (EGFR) inhibitors
Among all EGFR mutations, four types arestrongly correlated with TKI sensitivity in vitroand in vivo. These are point mutations in exons 18 (G719A/C) and 21 (L858R andL861Q) and in-frame deletions in exon 19(218)

Exon 18 Gly719 is sensitive for example

Exon 20 EGFR activating mutations are generally associated with resistance to EGFR tyrosine kinase inhibitors such as erlotinib, afatinib, and gefitinib, although insertions at or before position 768 can be associated with sensitivity

E.g. A763_Y764insFQEA mutant, which is sensitive to erlotinib and gefitinib.

Recently, a rare exon 22 mutation (E884K) that may confer differential sensitivity to different EGFR small-molecule inhibitors was reported (23)218
There is limited data on response to EGFR tyrosine kinase inhibitors for many of the uncommon EGFR activating mutations.
Whereas V689M, N700D,L718P, V765A, V783A, A839T, and K846R areassociated with response to geﬁtinib, E709Q/L,A763V, N826S, and V752I are associated withlack of response (62).218

mutations conferring resistance during treatment

biopsy at time of relapse may be required to determine the best course of therapy
The most common secondary mutation is the T790M substitution of methionine for threonine on codon790 (85–87).218 (60-70% using sensitive methods)

caused by a single base substitution, C to T, at nucleotide 2369
This mutation occurs in cis on the same allele as the original activating mutations; it may be either an exon 19 deletion oran exon 21 L858R mutation218
studies showed that the T790M mutation preferentially increasesthe afﬁnity of the receptor for ATP, therebyreducing the effectiveness of the TKI (25)218
Although T790M mutation induces resistance to geﬁtinib, it can still be inhibited by some irreversible EGFR inhibitors (88)218
accounts for approximately 50% to 60% of acquired resistance to EGFR TKI therapy (51)218
If seen in untreated/pretreated patients, may be present in the germline and indicate a hereditary cancer syndrome, in which case genetic counseling is suggested.

Most studies have only rarely detected T790M in pretreatment samples.
Notably,some studies have detected the T790M mutation at low levels in a few patients prior to initiation of EGFR TKI therapy,suggesting thatsubclones bearing these mutations are preexistent (83), as is true in other cancers, such aschronic myelogenous leukemia, that respond totargeted TKIs.218

Resistance mutations in the drug target markedly diminish the potency of the inhibitor againstthe kinase. Examples include EGFR T790M and BCR-ABL T315I.223
EGFR T790M is essentially the sole resistance mutation observed in the clinic.223
Recent data suggest that AR patients with the T790M mutation can derive continued clinical benefit from the first-line EGFR TKI

L747S, D761Y, and T854A

owing to their relatively low prevalence, there is not much clinical experience with these

MET amplication may account for approximately 10% to 20% of acquired resistance to geﬁtinib and erlotinib (110).218

there is currently a lack of a precise definition of clinically significant MET amplification in this setting and more research is needed before guidelines can be formulated

ERBB2 amplification has been reported in a subset
SCLC histology and associated ‘‘SCLCtype’’ radiosensitivity and chemosensitivity have been observed in some AR cases

EGFR Mutation details

TK domain of EGFR; the mutations are characterized by short deletions in exon 19 and point mutations (G719S, L858R,and L861Q) in exons19 and 21 (Figure 2)218
According to their nucleotide changes,the mutations have been classified into three types.

Class I mutations include short in-frame deletions that result in the loss of four to six amino acids (E746 to S752) encoded by exon 19.

Small deletions in the LREA motif of exon 19

Class II mutations are single-nucleotide substitutions that may occur throughout exons 18 to 21.

By far most common is L858R (leucine to arginine) (exon 21)

Class III mutations are in-frame duplications and/or insertions that occur mostly in exon 20 (20, 21).218

About 3% of EGFR mutations occur at codon 719 and cause substitution of glycine to cysteine, alanine, or serine (G719X) [54]. In addition, about 3% are in-frame insertion mutations in exon 20 [193]

Molecular biology:

EGFR belongs to theHER/erbB family of receptor tyrosine kinases (RTKs), which includes HER1 (EGFR/erbB1),HER2 (neu, erbB2), HER3 (erbB3), and HER4(erbB4).218
Intracellular signaling is mediated mainly through the RAS-RAF-MEK-MAPK pathway, the PI3KPTEN-AKT pathway, and the signal transducer and activator of transcription (STAT)pathway (13). Downstream EGFR signaling ultimately leads to increased proliferation, angiogenesis, metastasis, anddecreased apoptosis(Figure 1) (14).218
EGFR mutation has been shown to activate the antiapoptotic Akt and signal transducers and activators of transcription pathways

EGFR mutation testing guidelines (CAP/IASLC/AMP/ASCO)

ALK gene rearrangement (2-7% in US):

inv(2)(p21p23)

Inversion on chromosome 2p resulting in EML4-ALK fusion gene
fusion gene encoding the amino-terminal portion of EML4 (2p21) and the intracellular region of ALK (2p23), genes that are normally approximately 13 Mb apart.8

other less common variant fusions have been reported

translocations with other chromosomes (KIF5B-ALK, TFG-ALK)

KIF5B-ALK
TFG-ALK
KLC1-ALK

NOT the NPM-ALK transloation characterized in ALCL

Epidemiology:

Never smokers
Younger age
No sharp differences in prevalence according to sex and ethnic origin

Histology:

Adenocarcinoma, adenosquamous

Very infrequent in squamous cell carcinoma

Subtype is not strongly predictive

Solid histology maybe
Signet ring histology maybe

Inevitable acquired resistance to ALK-targeted inhibitors:

ALK kinase domain mutation (often)

ALK mutations that confer acquired resistance to crizotinib

L1152R, C1156Y, F1174L, L1196M, L1198P, D1203N, and G1269A have been reported
Testing for secondary mutations in ALK associated with acquired resistance to ALK inhibitors is not currently required for clinical management (the mechanism of resistance does not clearly predict response to next line therapy)
Ceritinib has shown efficacy in these relapsed patients and has been approved for those who developed resistance to, or could not tolerate, crizotinib

Testing guidelines (CAP/IASLC/AMP/ASCO)
Prediction of response to therapy:

ALK rearrangement predicts response to therapy with a targeted ALK inhibitor, such as crizotinib or ceritinib.
Some evidence suggests the type of FISH pattern (breakapart versus 5' probe deletion) may have implications for treatment response and outcomes.
Polysomy involving the ALK locus confirms that fluorescence in situ hybridization (FISH) scoring was carried out in tumor cells but does not predict response to therapy with targeted ALK inhibitors.
ALK IHC:

Absence of ALK protein expression in cancer cells suggests that this tumor is unlikely to harbor ALK rearrangement and to respond to treatment with a targeted inhibitor, such as crizotinib and ceritinib.
ALK protein expression in cancer cells (based on platform criteria) predicts the presence of ALK rearrangement and response to therapy with a targeted inhibitor, such as crizotinib and ceritinib.
Tumors with faint cytoplasmic labeling should be designated as equivocal. This result can rarely occur both with and without mutation.

mean response rate to EGFR TKI for patients with EGFR mutations is 68%
crizotinib (ALK/MET/ROS1 inhibitor):

RR 57%
PFS 6 months or greater in 72%
FDA approved for advanced-stage, ALK-positive lung cancer
ALK molecular testing should be used to select patients for ALK-targeted TKI therapy

Testing should not be chosen based on clinical characteristics

BRAF mutation (5%)

V600E (half of BRAF mutations in lung cancer)

Proportion of non-V600E mutations is higher in lung cancer than other cancers

dabrafenib (BRAF inhibitor) combined with trametinib (MEK inhibitor) showed efficacy in Phase II clinical trials (63% overall response rate in V600E)
smokers

ROS1 gene rearrangement (1-2% of NSCLC)

ROS1 rearrangement predicts a high response rate to therapy with a targeted inhibitor, such as crizotinib.
Fusion partners: SLC34A2, CD74, TPM3, GOPC (FIG), SDC4, EZR, LRIG3, KDELR2, CCDC6
FISH, IHC, PCR, NGS may be used (no “gold standard” method defined yet)
Polysomy involving the ROS1 locus confirms that fluorescence in situ hybridization (FISH) scoring was carried out in tumor cells but does not predict response to therapy with targeted inhibitors.
ROS1 IHC:

Absence of ROS1 protein expression in cancer cells suggests that this tumor is unlikely to harbor ROS1rearrangement and to respond to treatment with a targeted inhibitor, such as crizotinib.
ROS1 protein expression in cancer cells is highly sensitive for a rearrangement involving ROS1 but is not entirely specific. Therefore, confirmatory molecular methods should be used when ROS1 protein expression is detected.
Tumors with faint cytoplasmic labeling should be designated as equivocal. This result can rarely occur both with and without mutation.

MET mutations (3%)

Splicing variants
Insertion-deletion mutations resulting in exon 14 deletion

May be associated with MET amplification

Associated with response to crizotinib and cabozantinib

RET mutations and gene fusions (1-2%)

RET rearrangement is associated with response to targeted RET inhibitor therapies, such as cabozantinib and vandetinib
Absence of RET rearrangement in cancer cells suggests that this tumor is unlikely to respond to treatment with a targeted RET inhibitor.
Never smokers (almost exclusively)
Cabozantinib (multi-targeted inhibitor) shows promise in phase II trials

Phase 3 trials not conducted yet

FISH

The most common rearrangement, KIF5B-RET, is subtle intrachromosomal inversion and may be difficult to detect practically by FISH

PCR, NGS

ERBB2 mutation

exon 20 insertion mutation
phase I promising for HER-inhibitors

MAP2K1 (MEK1) (rare)

May predict sensitivity to MEK inhibitors

PIK3CA mutation (rare)
AKT1 mutation (rare)
PTEN loss
NRG1/NTRK1 (TRKA) fusion (rare)
beta-catenin gene mutations

fetal adenocarcinoma histology

MET amplification

Rare cases of de novo MET amplification have been associated with profound responses to crizotinib
May be associated with exon 14 deletion (see above)

EGFR copy number gain / amplification

Increased EGFR gene copy number (polysomy or amplification) is observed in about 40% of cases, with a range of 8% to 66%
EGFR TKI response rates for patients with EGFR polysomy/amplification is 30%

Methylation rates of APC, CDH13, and RARb (significantly higher in ADC than in SCC)
Changes common to all lung CA

Other features:

Metastasize widely and earlier than SCC
Atypical adenomatous hyperplasia is precursor lesion for some
Prognostic factors:

AIS and MIA – 100% or near 100% survival with complete resection
Size of central scar – smaller is better prognosis
Vascular invasion is worse prognosis
Papillary pattern >25% is worse prognosis
Fetal adenocarcinoma – most are low grade and favorable outcome

Predictive factors:

EGFR mutation – predictive of responsiveness to EGFR tyrosine kinase inhibitors
ADC histology is a strong predictor for improved outcome with pemetrexed therapy compared with SCC

References:

Robbins 2005
Travis WD. Pathology of Lung Cancer. Clinics in Chest Medicine 2002; 23(1): 65-81.
WHO blue book 2004
Travis et al. International association for the study of lung cancer / American Thoracic Society / European Respitratory Society International Multidisciplinary Classification of Lung Adenocarcinoma. Journal of Thoracic Oncology 2011;6(2):244-285.
Thomas RK, Weir B, Meyerson M. Genomic approaches to lung cancer. Clin Cancer Res (2006); 12(14 Suppl): 4384s-91s.
Travis et al. WHO Classification of Tumours of the Lung, Pleura, Thymus, and Heart, 4^th ed. (2015)
CAP – Template for Reporting Results of Biomarker Testing of specimens from patients with non-small cell carcinoma of the lung (June 2016)

CAP draft lung biomarker template protocol (Sep. 2015)