Abstract
Introduction: Lung cancer was considered to be rare at the beginning of the 20th century, but it has now reached almost epidemic proportions. It is the leading cause of cancer deaths in developed countries and is also rising at alarming rates in developing countries.
Introduction: Coronavirus disease 2019 (COVID-19) is an emerging viral infection without any approved treatment. Investigational therapies for COVID-19 may cause clinically important drug-drug interactions (DDIs). We aimed to study drug-drug interactions (DDIs) and their risk factors in hospitalised COVID-19 patients.
Aim: The aim of our study is to establish an effect on serum levels of vascular endothelial growth factor (VEGF) after surgery in lung cancer patients.
Methods:This was a prospective study. For the estimation of VEGF, 50 lung cancer patients were studied. Both preoperative and postoperative levels of VEGF were estimated for all subjects. Blood samples were obtained from all cases both preoperatively and postoperatively (four weeks after surgery). Blood samples of 100 age and sex matched healthy controls were collected from the Outpatient Departments of SKIMS to establish normal serum VEGF levels.
Conclusion: Our findings show that serum VEGF levels are higher as the tumor stage progresses and tumor size increases, which explains the lower serum VEGF levels observed by us in the operable patient group.
Keywords:lung, VEGF, cancer, angiogenesis.
INTRODUCTION
Lung cancer was considered to be rare at the beginning of the 20th century, but it has now reached almost epidemic proportions. It is the leading cause of cancer deaths in developed countries and is also rising at alarming rates in developing countries (1). It has a worldwide incidence of 14%, whereas it accounts for 6.8% of all cancers in India (2). It is the leading cancer of both sexes in three of the Urban Cancer Registries (Bhopal, Delhi and Mumbai) in India (3).
In NSCLC, angiogenesis has been evaluated by intratumoral microvessel count or expression of angiogenic markers. Angiogenesis is controlled by stimulatory factors, e.g., vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF), platelet derived endothelial cell growth factor (PDEC-GF)/ thymidine phosphorylase (TP) and inhibitory factors (e.g., angiostatin, endostatin whose balance determines the degree of angiogenesis (4). Tumors can up- or down-regulate these factors to produce an environment in which angiogenesis occurs. The VEGF/VEGFR signalling pathway is frequently up-regulated in lung cancer, and VEGF overexpression is associated with tumor progression (5). High VEGF levels have been identified as an independent prognostic factor correlated with poor prognosis in patients with lung cancer (6). VEGF family consists of seven members, including VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGF-E, VEGF-F, and placental growth factor. Among angiogenic stimulators, VEGF-A plays an essential role in both vasculogenesis and angiogenesis. Placental growth factor (PlGF) shares eight cysteine residues in a VEGF homology domain (7, 8).
Although the importance of VEGF expression in tumor tissues is known, the clinical and prognostic significance of serum VEGF in patients with cancer is not defined. The clinical utility of serum VEGF in predicting clinical course and prognosis in NSCSL is under investigation as well as in other types of cancers. There is a possible implication that serum VEGF level may be an independent prognostic factor among the known clinical factors if they are consistently elevated in tumors as opposed to other conditions, including normal conditions. Therefore, the aim of this study was to investigate the relationship between preoperative and postoperative serum VEGF levels.
MATERIALS AND METHODS
Chemicals and reagents
The various kits and reagents were procured from standard sources.
Patients and controls
Patients attending the Department of Cardiovascular and Thoracic Surgery, Medical Oncology and Internal Medicine of Sher-I-Kashmir Institute of Medical Sciences (SKIMS), India, were screened for lung cancer. A total of 50 histologically confirmed cases were included in the study. Blood samples from all cases were obtained both preoperatively and postoperatively (four weeks after surgery). Blood samples of 100 age and sex matched healthy controls were collected from the Outpatient Departments of SKIMS to establish the normal serum VEGF levels. Patients and controls had a mean age of 62.5 and 59 years, respectively. A written informed consent was obtained from all cases and controls. All patients who have received chemotherapy or had any other type of cancer were excluded from the study.
The demographic and clinic-pathological characteristics of each patient were recorded in a questionnaire to collect information about gender, age, dwelling, smoking status, histopathology, grade of differentiation, tumor stage, family history, etc. The collection and use of blood samples for this study has been approved by the Institute Ethical Committee.
Sample collection/storage
Peripheral blood (¡« 5 mL) was obtained from each subject in EDTA containing vials (200 ¦ÌL of 0.5 M, pH 8.0) and centrifuged at 3,000 rpms for 10 minutes in a refrigerated centrifuge. The resulting plasma was stored in small aliquots at -70¡ãC until use.
Enzyme linked immunosorbent assay (ELISA)
Principle and application
The enzyme-linked immunosorbent assay (ELISA) is a test that uses antibodies and color change to identify a substance.
Procedure for ELISA
All precautions were taken to ensure contamination free quantification of VEGF. All samples were run in duplicate to ensure statistically accurate results.
Equipments and reagents
1. ELISA kit
a. VEGF Microplate (Item A): 96 wells (12 strips x 8 wells) coated with anti-human VEGF
b. Wash buffer concentrate (20x) (Item B): 25 mL of 20x concentrated solution.
c. Standards (Item C): two vials, recombinant human VEGF
d. Assay diluent A (Item D): 30 mL, 0.09% sodium azide as preservative; for standard/sample (serum/plasma) diluents
e. Assay diluent B (Item E): 15 mL of 5x concentrated buffer; for standard/sample (cell culture medium/urine) diluents
f. Detection antibody VEGF (Item F): two vials of biotinylated anti-human VEGF (each vial is enough to assay half microplate)
g. HRP-Streptavidin conentrate (Item G): 200 ìL 300x concentrated HRP conjugated streptavidin
h. TMB one-step substrate reagent (Item H): 12 mL of 3,3’,5,5’- tetramethylbenzidine (TMB) in buffered solution
i. Stop solution (Item I): 8 mL of 0.2 M sulfuric acid
2. Microplate reader capable of measuring absorbance at 450 nm
3. Precision pipettes to deliver 2 ìL to 1 mL volumes
4. Adjustable 1-25 mL pipettes for reagent preparation
5. 100 mL and 1 liter graduated cylinders
6. Absorbent paper
7. Distilled or deionized water
8. Log-log graph paper or computer and software for ELISA data analysis
9. Tubes to prepare standard or sample dilutions
Reagent preparation
i. Bring all reagents and samples to room temperature (18 - 25°C) before use.
ii. Sample dilution: assay diluent A (Item D) should be used for dilution of serum/plasma samples. 1x assay diluent B (Item E) should be used for dilution of culture supernatants and urine. Suggested dilution for normal serum/plasma: 2-5 fold.
iii. Assay diluent B should be diluted 5-fold with deionized or distilled water.
iv. Preparation of standard: Briefly spin the vial of Item C and then add 640 μL of assay diluent A (for serum/plasma samples) or 1x assay diluent B (for cell culture medium and urine) into Item C vial to prepare a 50 ng/mL standard. Dissolve the powder thoroughly by a gentle mix. Add 60 μL of 50 ng/mL VEGF standard from the vial of Item C, into a tube with 440 μl of assay diluent A or 1x assay diluent B to prepare a 6,000 pg/mL standard solution. Pipette 400 μL of assay diluent A or 1x assay diluent B into each tube. Use the stock standard solution to produce a dilution series (shown below). Mix each tube thoroughly before the next transfer. Assay diluent A or 1x assay diluent B serves as the zero standard (0 pg/mL).
v. If the wash concentrate (20x) (Item B) contains visible crystals, warm to room temperature and mix gently until dissolved. Dilute 20 mL of wash buffer concentrate into deionized or distilled water to yield 400 mL of 1x wash buffer.
vi. Briefly spin the detection antibody vial (Item F) before use. Add 100 μL of 1x assay diluent B into the vial to prepare a detection antibody concentrate. Pipette up and down to mix gently (the concentrate can be stored at 4°C for five days). The detection antibody concentrate should be diluted 100-fold with 1x assay diluent B and used in step 4 of Part VI Assay Procedure.
vii. Briefly spin the HRP-Streptavidin concentrate vial (Item G) and pipette up and down to mix gently before use. HRP-Streptavidin concentrate should be diluted 300-fold with 1x assay diluent B.
ELISA protocol
ELISA for VEGF estimation was done as per manufacturer’s protocol. Briefly, the following steps were followed:
1. All reagents and samples were brought to room temperature (18-25°C) before use. All standards and samples were in duplicate.
2. 100 μL of each standard (see Reagent Preparation step 2) and sample was added into appropriate wells. Then wells were covered and incubated for 2.5 hours at room temperature or overnight at 4°C with gentle shaking.
3. Solution was discarded and washed four times with 1x wash solution (300 μL). After the last wash, remaining wash buffer was removed by aspirating or decanting by inverting the plate and blotting it against clean paper towels.
4. 100 μL of 1x prepared biotinylated antibody (Reagent preparation step 6) was added to each well and incubated for 1 hour at room temperature with gentle shaking.
5. Discarded the solution. Repeated the wash as in step 3.
6. 100 μL of prepared Streptavidin solution (see Reagent Preparation step 7) was added to each well and incubated for 45 minutes at room temperature with gentle shaking.
7. Discarded the solution. Repeated the wash as in step 3.
8. 100 μL of TMB one-step substrate reagent (Item H) was added to each well and incubated for 30 minutes at room temperature in the dark with gentle shaking.
9. Added 50 μL of stop solution (Item I) to each well and was read at 450 nm immediately.
Statistical analysis
The continuous variables of the study have been in terms of descriptive statics and categorical variables in terms of frequency & percentage. The standard statical tests like chi-square test, Fisher`s Exact test, McNemar test, Wilcoxon ranksum test, and also repeated measurement analysis have been used to analyse the data. All the results so obtained have been discussed on 5% level of significance that is P valve less than 0.05 considered significant.
RESULTS
A total of 50 newly diagnosed lung cancer patients who underwent surgical resection at the Sher-I-Kashmir Institute of Medical Sciences (SKIMS) were selected for the study. This study was conducted in the department of cardiovascular and thoracic surgery from August 2012 to September 2014.
Characteristics of the study subjects
Fifty lung cancer patients were studied to estimate both preoperative and postoperative VEGF levels.
Patients’ various clinicopathological features are summarized in Table 1.
Preoperative expression of VEGF in cases and controls
One hundred healthy age and sex matched volunteers were selected for the control group. The VEGF expression was determined by ELISA in all the 50 lung cancer cases and 100 controls in triplicates. The normal value was determined by establishing a normal distribution curve. The value of 150 ng/mL was taken as normal and other concentrations above it were considered elevated. The average value of triplicates was considered a true value.
The VEGF expression was elevated in 42% of cases. Among the different clinicopathological parameters, VEGF expression was found to be statistically significant for lymph node positive cases (P-value 0.009), higher pathological stage (P-value 0.009) and higher grade (P-value 0.000). The p value was calculated by using the chi-square test. Statistical information for preoperative VEGF expression and various clinicopathological features is shown in Table 2.
Postoperative VEGF expression
To observe the impact of curative surgical resection of tumours on VEGF levels, we estimated the VEGF expression at four weeks after surgery. Again, the VEGF levels were estimated by ELISA technique using standard kit. All sample were run in triplicates to assure quality results. All results were subjected to a statistical analysis. Chi-square test was performed to get the level of significance. We found that all those factors for which VEGF expression was statistically significant preoperatively (lymph node positive, clinical stage and grade) became non-significant (p value >0.05).
Statistical information for the postoperative VEGF expression and various clinicopathological features is shown in Table 3.
Table 4 is showing that VGEF levels were elevated in 21 (42%) patients before surgery, decreased to statically significant levels in 13 (62%) patients after surgery, and remained elevated in eight (38%) patients (p-value <0.001).
Table 5 shows that the mean VGEF level was 365.08 pg/mL preoperatively and 132 pg/mL postoperatively, with a mean difference of 223.5 pg/mL, and it was statically significant (p-value <0.001).
Table 6 is showing the effect of decrease in postoperative VEGF levels by controlling the tumor stage, nodal status and grade through repeated measurement analysis.
The table shows there is no interaction on decrease in postoperative VEGF levels with involvement of nodes, higher stage and grade of tumor; means patients who were having no node involvement, lower stage and grade of tumor, VEGF levels decreased in them also.
DISCUSSION
As the field of angiogenesis research is undergoing an explosive growth and the development of highly sensitive ELISA tests made it possible to analyse circulating angiogenic factors in clinical blood samples, we set out to investigate the diagnostic and prognostic potential of circulating VEGF in lung cancer patients undergoing surgery. The serum VEGF levels were useful for diagnostic in a study by Tamura M et al (9) or prognostic (Matsuyama W et al, Chol-JH et al, Brattstrom D et al) (10-12). Their predictive value with regard to surgery (13) and treatment for advanced lung cancer (14) has been previously explored and represented the topic of this study. Although our study is the first of its kind in estimating the VEGF levels both preoperatively and postoperatively, comparison will be made with studies in which VEGF values have been estimated prior to surgery.
In the present study, the total number of selected patients was 50. Serum VEGF levels were higher in our patient group than healthy controls and they were shown to be statistically significant.
Patient sex and VEGF
Out of all 50 cases, 84% were males and 16% females, which was not comparable to the male:female ratio observed by Qiang et al (7) but in agreement with that reported by Akin et al (8). The VEGF values were not statistically significant for gender (p-value 0.778) and were consistent with the findings of Qiang et al (15). Thus, age has no effect on VEGF expression in lung cancer.
Smoking and VEGF
In our study, we found no statistical association between VEGF values and smoking (p-value 0.617), although 84% of our patients were smokers and smoking was an independent risk factor for lung cancer, as observed by Hecht et al (17).
Lymph node status and VEGF
For patients with elevated VEGF values and lymph node positive status, a strong statistical significance (p-value 0.009) was noted, which is in agreement with studies carried out by Tamura et al (9), but in contradiction with findings reported by Ahmet et al (16). In our study, the VEGF values were elevated in only 22% of lymph node positive cases, while lymph node positivity was seen in 32% of patients and in around 20% of cases had elevated serum VEGF concentrations but negative lymph node status. However, VEGF levels decreased postoperatively and became statistically non-significant. It appeared that serum VEGF increased with lymph node involvement and also may reflect tumor burden. Our results are promising, as VEGF reduction postoperatively gives a clue regarding the use of anti-VEGF where VEGF remains or increases postoperatively, suggesting either a tumor relapse or residual tumor growth or disease aggressiveness.
Staging and VEGF
Since the sample size was low, we clubbed the staging into two groups, one with stage I and the other one as stage II/III. Again, a strong statistical association was observed for elevated VEGF values with higher stage (p-value 0.009). Among our patients, 68% had a stage I tumor, with elevated VEGF levels in only 20% of them, and 32% of patients had stage II/III, with elevated VEGF levels in 22% of them. However, our findings are in contradiction with those provided by earlier studies (7, 8, 10, 12), but in agreement with those conducted by Ohta et al and Imoto et al. Postoperatively, serum VEGF concentrations decreased to significant levels, showing serum VEGF levels increase with tumor growth, and might also reflect the tumor burden. A serum assay can be easily and frequently performed due to its minimal invasiveness.
Tumor histopathology and VEGF
Out of 50 patients, 31 (62%) had histology squamous cell carcinoma, with only 11 (22%) of them having increased serum VEGF levels, and 19 (38%) patients had histology adenocarcinoma and other types, with 10 (20%) of them having elevated VEGF levels. Although adenocarcinoma is the most frequent histological type, squamous cell carcinoma was the most common histological type in our study; this may be attributed to the fact that 84% of our patients were smokers, which is strongly associated with smoking. No statistical significance was observed for histopathology (p-value 0.617). In this case, our study is in agreement with earlier studies carried out by other authors (16). Again, no statistical significance was observed for decrease in VEGF levels postoperatively with regard to histopathology.
Tumor grade and VEGF
Since tumors with different grading show different biological phenomenon, we compared the VEGF levels with different tumor gradings. We clubbed well to moderately differentiated tumors as low grade and poorly differentiated tumors as high grade; 78% of patients were low grade, with increased VEGF levels in only 20% of them, while 24% of patients were high grade, with elevated VEGF levels in 22% of them. A strong statistical significance was observed between patients with higher grade and elevated VEGF levels (p-value 0.000). Soon after surgery, the VEGF levels became non-significant as was the case with lymph node status and tumor stage. Our results are in total disagreement with earlier studies carried out by Akin et al and Ahmet et al (16), which may be attributed to the low sample size in our study.
Other factors and VEGF
For all other factors such as anorexia, chest-pain, hemoptysis, breathlessness, cough and lung involved, no statistical significance was observed both pre- and postoperatively with serum VEGF levels. Common presenting symptoms, in order of decreasing frequency, were as follows: anorexia in 25 (50%) patients, cough [24 (48%)], hemoptysis [20 (40%)], chest pain [18(36%)] and breathlessness [15 (30%)]. On clinical examination pallor was seen in 24 (48%) patients.
As expected, the curative surgery had a great impact in normalizing the circulating VEGF levels. Postoperatively, we found a drastic change in serum VEGF levels. Preoperatively, the VEGF levels were statistically significant for lymph node metastasis, higher pathological stage and higher tumour grade. The mean preoperative and postoperative VEGF levels were 365.08 pg/mL and 152.5 pg/mL, respectively, with a mean difference of 223.5 pg/mL. This was statically significant change (p-value <0.001). This change is a direct indication of circulating VEGF levels as having a prognostic significance, as suggested by studies carried out by Akin et al. There was an overall decreased trend in postoperative VEGF levels, and no statically significant interaction was seen on decrease in postoperative VEGF levels by controlling the tumor stage, grade and nodal status, meaning that VEGF levels decreased in other patients who also had raised levels, but were node negative and lower tumor stage and grade.
Our data is mostly in agreement with previous studies which showed a role of VEGF as a prognostic factor. Since surgery is the best treatment for early stage lung cancer, the role of VEGF estimation postoperatively will be of tremendous help in managing patients suffering from this disease and will help in regulating the kind of therapy required for individual patients. Any kind of recurrence will not only raise the VEGF but will also give a good information regarding the underlying residual tumour. For those with higher stage lung cancers, for whom surgery is not viable option, the VEGF levels will help in diagnosing the aggressiveness of the disease. Cancers with higher VEGF levels will not only give information regarding its invasive potential but will also help in determining which is the suitable therapy for each patient.
Because angiogenesis is controlled between angiogenic and antiangiogenic factors, it is reasonable to use serum or plasma VEGF levels to correlate with the clinical outcome. Antiangiogenic therapy against VEGF and its receptors has been widely applied in the management of lung cancer patients. Serum VEGF levels should be measured and might be helpful to lead an efficient individualized antiangiogenic therapy for patients with lung cancer in the future.
CONCLUSION
In conclusion, we have demonstrated that serum VEGF levels are higher as the tumor stage progresses and tumor size increases, so naturally we had lower serum VEGF levels in the operable patient group.
Conflict of interests: none declared
Financial support: none declared.
Acknowledgements: Thanks to all coauthors, hospital staff and patients who participated in this study.
TABLE 1.
Patients’ clinicopathological features
TABLE 2.
Preoperative VEGF expression and clinicopathological features
TABLE 3.
Postoperative VEGF expression and clinicopathological features
TABLE 4.
McNemar test
TABLE 5.
Wilcoxon-rank sum test
TABLE 6.
Effect of decrease in postoperative VEGF levels
Contributor Information
Shabir Hussain RATHER, Department of Surgery, SKIMS Medical College, Srinagar, Jandk, India.
Showkat Ali BHAT, Department of Surgery, SKIMS Medical College, Srinagar, Jandk, India.
Arshad BASHIR, Department of Surgery, SKIMS Medical College, Srinagar, Jandk, India.
Abdul Majeed DAR, Department of Cardiovascular Surgery, SKIMS Institute of Medical Scineces Soura, Srinagar, Jandk, India.
Muzzafar ZAMAN, Department of Surgery, SKIMS Medical College, Srinagar, Jandk, India.
Naveed NABI, Department of Surgery, SKIMS Medical College, Srinagar, Jandk, India.
Aliya SHAH, Department of Microbiology, GMC Srinagar, India.
Akif Mutahar SHAH, Department of Anaesthesia, GMC Srinagar, India.
References
- 1.Khuri FR, Herbst RS, Fossella FV. Emerging therapies in non-small cell lung cancer. Ann Oncol. 2001;12:739–744. doi: 10.1023/a:1011197500223. [DOI] [PubMed] [Google Scholar]
- 2.Nanda Kumar A. Consolidated report of the population based cancer registries, incidence and distribution of cancer, 1990-1996. National Cancer Registry Programme. New Delhi: Indian Council of Medical Research. 2001. 2001.
- 3.Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics. CA Cancer J Clin. 2000;1:7–33. doi: 10.3322/canjclin.50.1.7. [DOI] [PubMed] [Google Scholar]
- 4.Gridelli C, Maione P, Rossi A, De Marinis F. Role of bevacizumab in the treatment of nonsmall cell lung cancer: current indications and future developments. The Oncologist. 2007;10:1183–1193. doi: 10.1634/theoncologist.12-10-1183. [DOI] [PubMed] [Google Scholar]
- 5.Sun S, Schiller JH, Spinola M, Minna JD. New molecularly targeted therapies for lung cancer. J Clin Invest. 2007;10:2740–2750. doi: 10.1172/JCI31809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Herbst RS, Onn A, Sandler A. Angiogenesis and lung cancer: prognostic and therapeutic implications. J Clin Oncol. 2005;14:3243–3256. doi: 10.1200/JCO.2005.18.853. [DOI] [PubMed] [Google Scholar]
- 7.Brock CS, Lee SM. Anti-angiogenic strategies and vascular targeting in the treatment of lung cancer. Eur Respir J. 2002;3:557–570. doi: 10.1183/09031936.02.00293002. [DOI] [PubMed] [Google Scholar]
- 8.Pang RWC, Poon RTP. Clinical implications of angiogenesis in cancers. Vasc Health Risk Manag. 2006;2:97–108. doi: 10.2147/vhrm.2006.2.2.97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tamura M, Ohta Y, Kajita T, et al. Plasma VEGF concentration can predict the tumor angiogenic capacity in non-small cell lung cancer. Oncol Rep. 2001;8:1097–102. doi: 10.3892/or.8.5.1097. [DOI] [PubMed] [Google Scholar]
- 10.Matsuyama W, Hashiguchi T, Mizoguchi A, et al. Serum levels of vascular endothelial growth factor dependent on the stage progression of lung cancer. Chest. 2000;118:948–951. doi: 10.1378/chest.118.4.948. [DOI] [PubMed] [Google Scholar]
- 11.Choi JH, Kim HC, Lim HY, et al. Vascular endothelial growth factor in the serum of patients with non-small cell lung cancer correlation with platelet and leukocyte counts. Lung cancer. 2001;33:171–179. doi: 10.1016/s0169-5002(01)00200-8. [DOI] [PubMed] [Google Scholar]
- 12.Battstrom D, Bergqvist M, Larsson A, et al. Basic fibroblast growth factor and vascular endothelial growth in sera from non-small cell lung cancer patients. Anticancer Res. 1998;18:1123–1127. [PubMed] [Google Scholar]
- 13.Battstrom D, Bergqvist M, Hesselius P, et al. Elevated preoperative serum levels of angiogenic cytokines correlate to larger primary tumors and poorer survival in non-small cell lung cancer patients. Lung cancer. 2002;37:57–63. doi: 10.1016/s0169-5002(02)00027-2. [DOI] [PubMed] [Google Scholar]
- 14.Kido Y. Vascular endothelial growth factor (VEGF) serum concentration changes during chemotherapy in patients with lung cancer. Kureme Med J. 2001;48:43–47. doi: 10.2739/kurumemedj.48.43. [DOI] [PubMed] [Google Scholar]
- 15.Qiang Lin, Maoquan Li, Zhen-Ya Shen, et al. Prognostic Impact of Vascular Endothelial Growth Factor-A and E-Cadherin Expression in Completely Resected Pathologic Stage I Non-Small Cell Lung Cancer. Jpn J Clin Oncol. 2010;7:670–676. doi: 10.1093/jjco/hyq041. [DOI] [PubMed] [Google Scholar]
- 16.Ahmet Demirkazik, Hakan Akbulut, A Fusun Ulger, et al. Serum vascular endothelial growth factor levels in patients with nonsmall cell lung cancer. Ann Saudi Med. 2004;5 doi: 10.5144/0256-4947.2004.473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hecht SS. Cigarette smoking and lung cancer chemical mechanism and approach to prevention. Lancet Oncol. 2002;J3:461–469. doi: 10.1016/s1470-2045(02)00815-x. [DOI] [PubMed] [Google Scholar]