EGFR-mutation–positive tumors would be expected to experience better outcomes related to both tumor response and tolerability. Soon after erlotinib was approved by the US Food and

Drug Administration (FDA), investigators noticed that patients

who were women, Asian, and nonsmokers tended to respond to

erlotinib therapy more frequently than other populations. However, it was subsequently realized that presence of mutations in

EGFR was a better predictor of response to erlotinib therapy

than demographic factors and smoking status. The prevalence

of EGFR mutations is between 10% and 15% for the overall population, but is 35% in white nonsmokers, and 65% for Asian

nonsmokers.47 Hence, the probability that L.L., a nonsmoking

white woman, has a tumor with this somatic mutation is higher

than for other categories of patients such as smokers or men.

Further, the observed incidence of these mutations is less than

3% for patients with squamous cell histology; therefore, it is not

recommended to test patients with this histology.22 L.L. was diagnosed with adenocarcinoma, which is a nonsquamous subtype,

and mutation testing is thus warranted on that basis alone.

The question arises that patients who fall into the category

with multiple clinical predictors of response could be started

empirically on erlotinib therapy without the need to test for

EGFR mutations. In L.L.’s case, she would be classified as having three of these clinical predictors: female, nonsmoker, and

adenocarcinoma. However, studies with populations of patients

from developed countries with NSCLC showed that those who

had three or more of these characteristics experienced a 49%

response rate, whereas those with sensitizing EGFR mutations

2217Lung Cancer Chapter 94

experienced much higher 67% response rate.35 This indicates

that EGFR mutation is a better predictor of response to erlotinib

therapy than presence of multiple clinical predictors.

Application of this EGFR testing is useful for small molecule

tyrosine kinase inhibitors (e.g., erlotinib), but not for antibody

molecules (e.g., cetuximab) that also block EGFR. The antibody

molecules function to block the receptor externally to the cell surface, and they also stimulate immune responses (e.g., antibodydependent cell-mediated cytotoxicity, complement) against the

tumor. It is likely that even without presence of the mutation(s),

the antibody molecule still binds, and can function through

several mechanisms to promote an anti-tumor effect. Small

molecule inhibitors, however, also bind receptor internally to

the cell surface and can inhibit the active ATP binding region,

which is constitutively activated in tumors with EGFR mutations.

For L.L., it is standard of care to send the tumor specimen for

EGFR mutation analysis.

CASE 94-2, QUESTION 3: Analysis of tumor tissue detected

L858R mutation in exon 21, which is commonly associated

with high sensitivity to the EGFR tyrosine kinase inhibitor

(TKI) erlotinib.35 L.L. was then started on erlotinib therapy

and within 8 weeks of therapy she had obtained a partial response. What benefit from erlotinib therapy can L.L.

expect in terms of prolonged survival?

Although there is short-term benefit for patients with EGFRmutation–positive tumors to receive erlotinib, it is not yet clear

if there is a longer-term benefit. The response rate for erlotinib

treatment is approximately 67%, time to disease progression is

11.8 months, and overall survival is approximately 24 months.

As of yet, there are no data to support that treatment with

small molecule inhibitors prolongs overall survival, relative to

doublet platinum-based therapy. It is much easier to demonstrate an effect on tumor response and time to progression (such

as with erlotinib) than it is for overall survival in most cancer studies. Showing improvement in overall survival is difficult

because study subjects who are randomly assigned to one arm

of a study can exhibit progressive disease, and could hypothetically cross over to receive the competing treatment once they

are classified as off study. The competing treatment could prolong overall survival and confound interpretation of the overall survival benefit for the test treatment (e.g., erlotinib). Further study is required to establish the benefit of erlotinib treatment on overall survival. There is also no evidence to show

that addition of erlotinib to chemotherapy is superior to either

treatment alone. Hence, at this time, treatment for L.L. (singleagent erlotinib) would be expected to prolong her progressionfree survival, but it is not yet known if erlotinib prolongs overall

survival.

CASE 94-2, QUESTION 4: As mentioned earlier, patients

with stage IV NSCLC should receive radiotherapy prior to

chemotherapy, provided they can tolerate aggressive treatment. However, L.L. did not receive radiotherapy. What

characteristic about this patient might preclude her from

receiving radiation prior to erlotinib?

The rationale for omitting radiotherapy has more to do

with the likely toxicity of the combination of radiation and

chemotherapy, rather than any characteristics of L.L. She clearly

has good performance status, and there is no indication of comorbidities other than moderate renal dysfunction therefore she

would likely tolerate aggressive chemotherapy. Two recent clinical trials suggest that radiotherapy and small molecule EGFR

inhibitor therapy may not be safe to give either sequentially or

together. Both studies were conducted with the EGFR small

molecule inhibitor, gefitinib, a quinazoline with similar properties to erlotinib. In the first study, gefitinib versus placebo was

given after a platinum-based concurrent radiotherapy regimen.

Enrollment for this phase III study was stopped early due to the

results of an interim analysis showing median survival time for

those getting placebo (35 months) was greater than for those

getting gefitinib (23 months). The gefitinib arm was associated

with an increased number of pulmonary deaths.48 The second

study was a phase I study evaluating the safety of concurrent gefitinib, docetaxel, and three-dimensional conformal radiotherapy.

Dose-limiting pulmonary toxicity was observed in this second

study.49 For both studies, the EGFR inhibitor was given concurrently and/or after cytotoxic chemotherapy and radiotherapy.

Therefore, the toxicity may be the result of all three (cytotoxic

chemotherapy, radiotherapy, and EGFR inhibitor). Radiotherapy

delivered to the lungs is associated with pulmonary fibrosis and

small molecule EGFR inhibitor therapy is associated with the

development of pneumonitis. Though definitive proof does not

exist for this interaction, it would not be advisable to give L.L.

radiotherapy prior to her erlotinib treatment.

CASE 94-2, QUESTION 5: If L.L.’s tumor was classified as

having wildtype EGFR instead, what treatment should L.L.

receive?

If the biopsy results had shown that the tumor EGFR was

wildtype (i.e., no mutation), then L.L. would be considered for a

platinum-based doublet such as one shown in Table 94-6. Because

her disease is nonsquamous, the adjunct agent selected would

most likely be pemetrexed. Scagliotti et al. randomly assigned

patients with advanced-stage NSCLC to receive either cisplatin

plus gemcitabine versus cisplatin plus pemetrexed. Although

the median survival time was similar between the two arms,

there were differences depending on tumor histology. Patients

who received pemetrexed and had adenocarcinoma experienced

a median overall survival of 12.6 months, whereas those who

received gemcitabine had shorter survival. In contrast, those with

squamous carcinoma who received pemetrexed experienced a

9.4-month median overall survival, whereas those who received

gemcitabine had longer survival.31 Therefore, if L.L.’s tumor had

been wildtype for EGFR, she could receive cisplatin and pemetrexed as first-line treatment because her disease is adenocarcinoma. Consideration could also be given for using cisplatin and

paclitaxel together with bevacizumab. As mentioned previously,

the latter agent is approved for use only in patients with nonsquamous cell NSCLC.34 Those who respond to treatment with

four to six cycles of cytotoxic therapy can receive maintenance

therapy afterward.

CASE 94-2, QUESTION 6: As discussed in Chapter 90,

Adverse Effects of Chemotherapy and Targeted Agents,

the adverse effect profiles for cisplatin and carboplatin differ from each other. If a platinum-based regimen had been

selected instead of erlotinib for L.L., what characteristics

about L.L. could be used to guide the selection of carboplatin versus cisplatin?

Cisplatin is predominantly associated with ototoxicity,

nephrotoxicity, and neurotoxicity, whereas carboplatin is predominantly associated with myelosuppression. This would be

her first cycle of chemotherapy; therefore, she is not expected

to have depleted reserves of myeloid progenitor cells. Her estimated creatinine clearance is 48 mL/minute and based upon this,

2218Section 17 Neoplastic Disorders

would be considered to have moderate renal impairment. In this

case, carboplatin may be preferred over cisplatin because it is

associated with less renal toxicity and it is dosed according to the

Calvert formula, which accounts for renal function:

Total Carboplatin dose (mg) = AUC (GFR + 25) (Eq. 94-1)

AUC is the area under the concentration-time curve and GFR

is glomerular filtration rate.50 A typical dose of carboplatin in

this setting would be determined based on an AUC to account

for renal function instead of the typical weight (or m2)-based

dosing. AUC targets are commonly 4 to 6 mg/mL · minute for

most doublet regimens. If a patient with moderate renal function were to receive a fixed dose that is similar to that given to

a patient with normal renal function, then systemic exposure

would be higher. This higher exposure could result in greater

toxicity, notably greater myelosuppression. By dosing the drug

according to the patient’s renal function, the risk for overdosing

a patient with decreased renal function is minimized. Personnel, usually in the pharmacy, must calculate the actual dose that

should be prepared. For example, if L.L.’s carboplatin dose is

ordered as 5 mg/mL · min, then her dose would be calculated as

dose = 5 mg/mL · minute (48 mL/min + 25), which is 365 mg.

In contrast, if her renal function were within normal range for

estimated creatinine clearance (e.g., 100 mL/minute), her carboplatin dose would be 625 mg (∼70% higher). Several different

methods may used to estimate renal function, such as the Cockroft and Gault equation, and the dose calculations will differ

according to the method chosen. Usually, the choice of method

is institution-specific, because there is no evidence to show that

one estimation method is superior to the other. By consistently

using the same formula, providers reduce variability in systemic

exposure and thus increase the predictability of tolerance to the

drug. If the patient’s GFR is estimated based upon serum creatinine measurements by the isotope dilution mass spectrometry method, then the GFR should be capped to a maximum of

125 mL/minute, because serum creatinine values can be underestimated when they fall below 0.7 mg/mL, as can happen in some

patients. The FDA issued this safety alert to avoid administration

of high doses to patients with normal renal function and thus

likely avoiding drug-related toxicity.51

In conclusion, L.L. should receive erlotinib as first-line therapy. If her tumor had contained wildtype EGFR (instead of

mutant), or if her disease becomes refractory to erlotinib, then

a platinum-based doublet (i.e., cisplatin or carboplatin) could be

considered. She has moderate renal impairment; therefore, an

agent such as cisplatin would not be favored because it is associated with a high incidence of renal toxicity. Carboplatin would

be a safer choice because the dose would be selected based upon

her renal function, and is associated with a lower incidence of

renal toxicity than cisplatin.

CASE 94-2, QUESTION 7: L.L. experienced minor grade

diarrhea and skin rash and otherwise tolerated erlotinib

treatment very well without any impact on her daily activities and quality of life. L.L. continued erlotinib therapy, and

at 9 months there is no evidence of disease progression. The

plan is to continue this therapy until the disease relapses.

In general, therapy with erlotinib is well tolerated, especially

relative to cytotoxic chemotherapy. Diarrhea and rash are the

two most common adverse effects of EGFR TKI therapy.52 The

diarrhea can be treated with loperamide in most cases. Rash

requires intervention in approximately one-third of cases, and it

is desirable to treat it with agents such as 2% topical clindamycin,

minocycline, or doxycycline, and topical 1% hydrocortisone

(discussed in Chapter 90, Adverse Effects of Chemotherapy and

Targeted Agents). Erlotinib dose reduction (in 50-mg decrements) could also be considered, although this would likely

shorten the anticancer benefit experienced with this agent.53

As mentioned earlier, NSCLC is primarily a disease in those

older than 60 years. L.L. is 85 years old, and there is concern

that patients, particularly more than 80 years old, may not tolerate cytotoxic chemotherapy as well as younger patients. Hence,

elderly patients are often undertreated. However, recent studies

suggest that survival benefits are greater in elderly patients who

receive doublet therapy versus single agent.54 More studies of

various regimens are needed that would focus on treatment of

elderly patients, particularly effects on survival, quality of life,

and tolerability. Such studies would investigate the best clinical

parameters that enable prediction of response and tolerability

to therapies, and also the doses associated with the best outcomes. Because the majority of patients have wildtype EGFR,

cytotoxic chemotherapy is first-line. Therefore, considerations

must be given to patient preferences, co-morbidities, and performance status. It is not uncommon for those with advanced

age to live alone. Caretakers who could help the patient monitor for and treat adverse effects of chemotherapy (dehydration

from diarrhea or vomiting, febrile neutropenia, etc.) would be

an essential consideration, particularly in this population. Even

though L.L. is able to continue treatment beyond 9 months, most

patients eventually exhibit progressive disease within 1 to 2 years.

These patients can often be re-biopsied. Approximately 40%

of these cases develop secondary mutations in EGFR (T790M)

that render the tumor resistant to erlotinib treatment.55 At such

time, consideration can be given for second-line treatment, and

if there is concern regarding tolerability of combination cytotoxic chemotherapy, then single agent therapy would be another

second option.

Small Cell Lung Cancer

EPIDEMIOLOGY

Small cell lung cancer (SCLC) accounts for approximately 15%

of all lung cancer histology, and affects both sexes in equal distribution. The disease is much more highly attributable to smoking than NSCLC. As the numbers of people who smoke have

decreased in the United States since its peak in the 1960s, the

incidence of SCLC has also declined. Relative to NSCLC, these

tumors generally have a more rapid doubling time, a higher

growth fraction, and early development of widespread metastases. As a consequence, SCLC is highly sensitive to chemotherapy and radiotherapy; however, most patients eventually die from

recurrent disease.56,57

PATHOPHYSIOLOGY

SCLC is a malignancy thought to be derived from neuroendocrine cells in the bronchus. SCLC is readily diagnosed on small

specimens such as bronchoscopic biopsies, fine needle aspirates,

core biopsies, and cytology. As the name suggests, these tumors

consist of small cells with limited volume of cytoplasm, poorly

defined cell borders, and finely granular chromatin.58 The cells

may be round, oval, or spindle-shaped, and, as previously mentioned, the mitotic count is high. Other characteristics, which

may enable characterization as SCLC include immunoreactivity for cytokeratin (AE1/AE3), epithelial membrane antigen,

thyroid transcription factor 1, and selected markers of neuroendocrine differentiation (CD56, chromogranin, and synaptophysin). The diagnosis is not based solely upon the presence of

the latter, because approximately 10% of NSCLC also stain positively for these markers. Hence, the diagnosis of SCLC requires

2219Lung Cancer Chapter 94

microscopic evaluation of tissue histology, and tests for presence

of molecular markers may also aid in the diagnosis.1

CLINICAL PRESENTATION

SCLCs usually arise centrally (i.e., in the chest region) and present

as a large hilar mass with bulky mediastinal lymphadenopathy

that can cause cough and dyspnea. Very rarely do patients diagnosed with SCLC present with the primary tumor located in the

lung periphery, as can be the case with NSCLC. Due to smoking,

patients with lung cancer may have had previously existing symptoms such as cough and even dyspnea that are related to presence of other smoking-related diseases such as COPD. Therefore,

these previously existing symptoms may not prompt patients to

seek medical attention except in the further management of the

underlying condition.

DIAGNOSIS AND OVERVIEW OF TREATMENT

In general, diagnostic procedures are similar to those used to

diagnose NSCLC. Staging of SCLC is used to determine prognosis and treatment. The median overall survival for patients

with limited stage disease ranges from 17 to 26 months and, for

extensive stage, 3 to 12 months.59 (These data were collected

from an analysis of 14 studies of SCLC, thus wide ranges for

survival are reported.) Surgery (e.g., thoracotomy) is appropriate for less than 5% of patients with early-stage disease, mainly

due to the tendency for these tumors to be bulky and to metastasize quickly. The limited role for surgery, therefore, enables

the use of a simple two-stage system instead of the TNM system

used for other solid tumors. In the Veterans Administration Lung

Study Group staging system, limited stage disease is defined as

disease confined to the ipsilateral hemithorax and encompassed

in a tolerable radiation field, and extensive disease is defined as

disease beyond the ipsilateral hemithorax including malignant

pleural, pericardial effusion or hematogenous metastases.60 As

discussed subsequently, patients with limited stage disease are

treated with a combined modality approach (i.e., chemotherapy

and radiation), and extensive stage, chemotherapy. Only approximately 30% of patients present with limited stage, and the rest

with extensive stage disease.61

CLINICAL PRESENTATION AND PATHOPHYSIOLOGY

OF SMALL CELL LUNG CANCER

CASE 94-3

QUESTION 1: M.W. is a 63-year-old woman who presented

to her primary doctor with complaint of heartburn and pain

in the right side of her upper abdomen; she also had a

feeling of gas in the stomach. In addition, she