SEARCH

SEARCH BY CITATION

Keywords:

  • nonsmall cell lung carcinoma;
  • diagnostic imaging;
  • neoplasm staging;
  • positron emission tomography;
  • pathology

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

BACKGROUND:

The aim of the current study was to determine whether the [18F]-2-fluoro-deoxy-D-glucose (FDG) positron emission tomography (PET) standardized uptake value (SUV) in patients with a new diagnosis of stage I lung cancer correlates with a specific cellular component in the primary tumor.

METHODS:

This study was Health Insurance Portability and Accountability Act compliant and approved by our Institutional Review Board with a waiver of informed consent. Forty patients with stage I nonsmall cell lung cancer (NSCLC) who underwent FDG-PET imaging at the time of diagnosis followed by surgical resection were retrospectively identified. Histologic sections of the primary tumor were reviewed by a pathologist without any clinical data and scored according to the percentage of each cellular component (tumor cells, normal stroma, inflammatory cells, necrosis, fibrosis, and other). Each component was compared with maximal (SUVmax) and mean (SUVmean) SUVs using Pearson correlation coefficient analysis.

RESULTS:

The mean SUVmax and SUVmean values for 40 stage I NSCLC tumors were 8.8 and 5.4, respectively. The mean histologic composition of tumor specimens in order of frequency was as follows: tumor cells (38.9%), fibrosis (30.8%), inflammatory cells (14.8%), normal stroma (5.2%), necrosis (5.8%), and other components (4.5%); however, there was considerable variation noted among individual tumors. There was no statistically significant correlation between SUVmax (r = .19; P = .24) or SUVmean (r = .017; P = .29) and the proportion of tumor cells in the tumor mass or any other cellular components.

CONCLUSIONS:

The cellular composition of stage I NSCLC appears to be highly variable, with no correlation noted between a specific tumor cellular component and FDG activity. Cancer 2010. © 2010 American Cancer Society.

[18F]-2-fluoro-deoxy-D-glucose (FDG)-positron emission tomography (PET) imaging has become an essential tool for evaluating patients with lung cancer. It is useful in suggesting the diagnosis of malignancy, particularly when there is an indeterminate abnormality on conventional imaging studies. PET also assists in staging patients, provides prognostic information, may be useful in monitoring response to therapy, and can be used to follow patients after treatment.1-3

The rationale for using FDG-PET in oncology is its ability to measure increased glucose metabolism, a fundamental property of tumor cells. Elevated FDG uptake suggests which lesions or tissues may harbor tumor cells. Previous studies in patients with early stage nonsmall cell lung cancer (NSCLC) found that FDG activity correlated with tumor doubling time and survival; faster growing tumors were more metabolically active, had a higher maximal standardized uptake value (SUVmax), and a worse clinical prognosis.2-14

However, to the best of our knowledge, the precise mechanism of FDG activity and cellular distribution within malignant tumors is unknown. In fact, the cellular components of lung cancers are markedly heterogeneous, comprised of a spectrum of cells including malignant tumor cells, atypical but not outright malignant cells, normal parenchymal cells, inflammatory cells, necrotic material, fibroblasts, and hematopoietic progenitor cells. Although it is often assumed that FDG uptake in lung cancers is primarily within the tumor cells, the correlation between FDG uptake and cell composition in resected human lung tumor specimens is to the best of our knowledge unknown. This study was performed to determine whether there is a relation between FDG activity and specific tumor cellular components in patients with stage I NSCLC.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

This study was approved by our Institutional Review Board; the requirement for informed consent was waived. A retrospective review of all patients within our center's tumor registry identified patients diagnosed with early NSCLC from January 1998 to 2004 (International Association for the Study of Lung Cancer, 7th edition, [T1a-T2a, N0, M0] stage I disease). Patients with an FDG-PET examination at the time of initial staging and no prior treatment, and who subsequently underwent surgical resection with histopathologic diagnosis were identified; 235 patients met eligibility criteria. Forty patients were selected at random for analysis as determined by a 2-tailed correlation power analysis. Surgical pathologic stage I patients were selected for our study to define a relatively homogeneous clinical population and, because the tumors were completely resected, to ensure a sufficient amount tissue for histologic review as well as a more representative specimen than that provided by fine-needle aspirate or other forms of tissue sampling.

PET Imaging

Preoperative FDG-PET examinations were obtained at the time of initial staging. Patients were instructed to fast for a minimum of 4 hours before the examination. Blood glucose levels were obtained and required to be <200 mg/dL before the intravenous administration of 145 μCi/kg (maximum of 200 μCi) of FDG. Imaging began at 1 hour (± 15 minutes) after radiopharmaceutical injection. PET imaging was performed using a GE Advance unit (General Electric Medical Systems, Milwaukee, WI). Attenuation-corrected images of the chest were obtained using a germanium-68 transmission source and protocols in routine use at the time. Patients imaged after June 2003 also had a low-dose computed tomography (CT) scan without contrast enhancement obtained for PET fusion imaging and lesion localization.

The images were reviewed on a dedicated workstation provided by the equipment manufacturer. The SUV for pulmonary lesions was quantitatively obtained as a marker of FDG-PET activity; the SUVmax and mean SUV (SUVmean) were calculated as the maximum and mean SUV values, respectively, within a region of interest drawn around the primary tumor on transaxial FDG-PET. Partial volume correction is not routinely performed at our institution. SUVmax was defined as the maximum tumor concentration of FDG divided by the injected dose, corrected for patient body weight: (SUVmax = maximum activity concentration/{injected dose/body weight}). SUVmean was defined as the mean tumor concentration of FDG divided by the injected dose, corrected for patient body weight: (SUVmean = mean activity concentration/{injected dose/body weight}). SUV measurements were initially obtained at the time of staging PET in the setting of clinically suspected malignancy, without pathologic diagnosis. These values were subsequently confirmed by a second radiologist (with 15 years of experience) during analysis for the purposes of this study.

Pathologic Review

A single hematoxylin and eosin-stained slide from each resected tumor surgical specimen was selected at random. The slides were digitized for virtual microscopy (Aperio ScanScope CS and ImageScope software, Vista, CA). Five random high-power fields showing the tumor (× 20 magnification) for each slide were reviewed by a pathologist (with 30 years of experience) blinded to the PET results and all other clinical data to determine the overall tumor cellular composition as a percentage for each field representing the following 6 categories: tumor cells, normal stromal tissue, inflammatory cells, necrosis, fibrosis, and “other” cell types. The proportion of each cell type was obtained by a conventional morphometric method based on a point-counting technique. The 5 fields were averaged to obtain a representative component percentage for each specimen. Summation of the average over 5 fields for each cell type represents 100% of the tumor histologic construct. Tumor pathologic characteristics were also obtained including histologic subtype, grading, and tumor dimensions. Tumors were treated as spheres for volume calculations: volume = (4/3r3, with r representing the mean tumor radius.

Data Analysis

Histopathologic specimen composition was compared with radiologic and surgical tumor characteristics by Pearson correlation coefficient analysis to assess the relation between FDG uptake (SUVmax and SUVmean) and surgical tumor specimen histologic cellular composition.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Patient Characteristics

A total of 40 patients meeting inclusion criteria were selected at random: 23 men and 17 women, with a mean age of 67 years (range, 39-85 years). Each patient had a single stage I NSCLC that was surgically resected. Adenocarcinoma was the most prevalent malignancy, accounting for 21 (53%) cases, followed by 16 cases of squamous cell carcinoma (40%), and 3 (7%) nonclassified tumors. The majority of tumors were poorly or moderately differentiated, each accounting for 16 (40%) cases. A summary of patient characteristics is provided in Table 1.

Table 1. Summary of Patient Characteristics
CharacteristicNo. of Patients (%)
Sex
 Male23 (55)
 Female17 (45)
Age, y
 <502 (5)
 50-596 (15)
 60-6911 (28)
 ≥7021 (52)
 Mean67
 Range39-85
Tumor histology
 Adenocarcinoma21 (53)
 Squamous16 (40)
 Not classified3 (7)
Tumor differentiation
 Well7 (18)
 Moderately16 (40)
 Poorly16 (40)
 Not classified1 (2)
Stage
 I40 (100)
 IA (T ≤3 cm)27 (67)
 IB (T >3 but ≤5 cm)13 (33)

FDG-PET Results and Tumor Cellular Composition

FDG-PET was performed a median of 7 weeks (range, 1-9 weeks) before surgical resection with a mean SUVmax of 8.8 (range, 1.6-23.9) and mean SUVmean of 5.4 (range, 1.2-14.9) (Table 2). Seven of 40 patients were imaged with PET-CT, whereas in the remaining 33 patients, a PET without CT-attenuation correction was obtained. Thirty-seven patients (93%) had positive FDG-PET results as defined by a tumor SUVmax ≥2.5. Three patients (7%) had tumors with SUVmax <2.5; incidentally, all of these were classified as adenocarcinomas on histology and ranged in size from 0.8 to 4.0 cm. Surgical pathologic evaluation demonstrated a mean tumor long axis size of 2.6 cm (range, 0.8-5.0 cm) corresponding to a mean volume of 9.9 cm3 (range, 0.1-65.5 cm3).

Table 2. Summary of Tumor Characteristics
CharacteristicMean (Range)
  1. SUVmax indicates maximal standardized uptake value; SUVmean, mean SUV.

SUVmax8.8 (1.6-23.9)
SUVmean5.4 (1.2-14.9)
Tumor size long axis at pathology, cm2.6 (0.8-5.0)
Tumor volume, cm39.9 (0.1-65.5)
Histologic cell type 
 Tumor, %38.9 (10-83)
 Normal stroma, %5.2 (0-38)
 Inflammatory, %14.8 (0-60)
 Necrosis, %5.8 (0-39)
 Fibrosis, %30.8 (0-70)
 Other, %4.5 (0-77)

Histologic review of surgical specimens revealed a highly variable tumor cellular composition. Malignant tumor cells were the most abundant cell type, comprising 38.9% of the tumor cellular mass on average (range, 10-83%). Remaining components in order of frequency were as follows: fibrosis (30.8%; range, 0-70%), inflammatory cells (14.8%; range, 0-60%), normal stroma (5.2%; range, 0-38%), necrotic material (5.8%; range, 0-39%), and other components (4.5%; range, 0-77%). Table 2 summarizes key imaging and pathologic tumor features.

Statistical Analysis

Pearson correlation coefficient analysis was performed comparing both SUVmax and SUVmean with each of the cellular components independently (Tables 3 and 4, respectively). There was no statistically significant correlation noted between FDG activity as measured by the SUVmax or SUVmean and a specific tumor cellular component for stage I NSCLC. Statistical analysis did indicate a significant correlation between SUVmean and mean necrotic material within the tumor specimen (P = .006). Representative correlative analysis between SUVmax and mean tumor cell component is illustrated in Figure 1.

thumbnail image

Figure 1. Scatterplot of maximum standardized uptake value (SUVmax) versus mean tumor cellularity (shown as the percentage) is shown.

Download figure to PowerPoint

Table 3. SUVmax Correlation Summary
ParameterrP
  1. SUVmax indicates maximal standardized uptake value.

Mean tumor cellularity, %0.19.24
Mean inflammatory cellularity, %0.13.42
Mean normal stroma, %−0.09.56
Mean necrosis, %,0.27.09
Mean fibrosis, %−0.15.34
Mean other cell types, %−0.34.03
Mean tumor plus inflammatory cellularity, %0.24.12
Mean fibrosis plus inflammatory cellularity, %−0.07.68
Mean tumor cellularity for adenocarcinoma, %0.29.21
Mean tumor cellularity for squamous cell carcinoma, %0.11.67
Mean tumor cellularity for well-differentiated tumors, %0.10.83
Mean tumor cellularity for moderately differentiated tumors, %0.46.06
Mean tumor cellularity for poorly differentiated tumors, %−0.02.93
Table 4. SUVmean Correlation Summary
ParameterrP
  1. SUVmean indicates mean standardized uptake value.

Mean tumor cellularity, %0.17.29
Mean inflammatory cellularity, %0.10.55
Mean normal stroma, %−0.03.85
Mean necrosis, %0.42.006
Mean fibrosis, %−0.18.26
Mean other cell types, %−0.35.02
Mean tumor plus inflammatory cellularity, %0.21.19
Mean fibrosis plus inflammatory cellularity, %−0.11.47
Mean tumor cellularity for adenocarcinoma, %0.28.21
Mean tumor cellularity for squamous cell carcinoma, %0.18.46
Mean tumor cellularity for well-differentiated tumors, %0.10.83
Mean tumor cellularity for moderately differentiated tumors, %0.46.06
Mean tumor cellularity for poorly differentiated tumors, %−0.04.88

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Lung cancer is the leading cause of cancer-related deaths in the United States. An estimated 215,000 patients will be diagnosed with lung cancer this year in the US, with approximately 3 million new cases worldwide. Despite advances in imaging technology over the past 2 decades, including improved anatomic detail available with multidetector computed tomography and utilization of FDG-PET, the majority of patients still present with advanced stage disease, with only 16% of patients presenting with early stage I clinicopathology at the time of diagnosis. Despite a more favorable prognosis with early detection, the 5-year survival rate for patients with stage I lung cancer is still only 50%.15

Many factors influence outcomes and predict the biological behavior of NSCLC, including performance status, weight loss, histology, and molecular markers. However, stage at presentation is currently the most important prognostic indicator. Noninvasive staging is best assessed by FDG-PET,16-22 because it is more sensitive and specific than other imaging modalities, and often directs patient management.23 Some studies have also demonstrated the prognostic utility of FDG-PET, suggesting that as SUVmax increases survival decreases.3, 4, 24 Furthermore, FDG-PET has been promulgated as a more accurate tool to monitor response to therapy and for surveillance of disease recurrence after definitive treatment. These indications remain to be confirmed in prospective clinical trials because currently available data are controversial.25-28

Although clinical studies continue to define the role of PET in NSCLC, to the best of our knowledge the precise mechanism of FDG uptake and distribution among cells within a tumor is not known. FDG activity is a product of several fundamental features including glucose metabolism and the type/number of cells present in the tumor. Prior reports have indicated no statistically significant correlation between FDG uptake and quantitative expression of glucose transporters in NSCLC,29 and the contribution of the various cell types has not to our knowledge been completely investigated.24 Other factors that may affect FDG activity include tumor size, the presence of necrosis, and patient immune-mediated response.30, 31

Furthermore, the ratio of tumor cells to other cell types within a tumor may affect FDG uptake. It has been postulated that FDG uptake within tumor cells may be altered by competitive uptake within macrophages and lymphocytes.32 Tumors with a large inflammatory component could theoretically have a higher SUVmax than a similar tumor without inflammation. Alternatively, an elevated SUVmax may result from a quantitatively small amount of tumor cells that happen to be highly metabolically active (Fig. 2). Conversely, a low SUVmax could be produced by a large number of tumor cells that have relatively low metabolic activity (Fig. 3).

thumbnail image

Figure 2. A 70-year-old man with a nonsmall cell lung cancer of the upper left lobe measuring 3.5 cm is shown. (Top) [18F]-2-Fluoro-deoxy-D-glucose (FDG) -positron emission tomography (PET) demonstrated a hypermetabolic mass with a maximum standardized uptake value (SUVmax) of 14.44. (Middle) Corresponding computed tomography image without contrast enhancement is shown. (Bottom) A representative hematoxylin and eosin-stained histologic specimen is shown at ×20 magnification. Tumor necrosis was most prevalent (39%), with tumor cells accounting for only 26% of the malignancy. The remaining components included fibrosis (18%), inflammatory cells (12%), and normal stroma (5%).

Download figure to PowerPoint

thumbnail image

Figure 3. A 57-year-old man with a nonsmall cell lung cancer of the upper left lobe measuring 1.7 cm is shown. (Top) [18F]-2-Fluoro-deoxy-D-glucose (FDG) -positron emission tomography (PET) demonstrated a hypermetabolic nodule (arrow) with a maximum standardized uptake value (SUVmax) of 4.65. (Middle) A corresponding computed tomography image without contrast enhancement is shown. (Bottom) A representative hematoxylin and eosin-stained histologic specimen is shown at ×20 magnification. Tumor cells accounted for 81% of the malignancy. The remaining components included necrosis (7%), normal stroma (6%), fibrosis (4%), and inflammatory cells (2%).

Download figure to PowerPoint

The current study found that stage I NSCLCs are indeed comprised of a spectrum of heterogeneous cell types. The complex host response to malignant cell transformation affects tumor phenotype, although the contribution from the various components and the mechanisms are poorly understood.31 Consequently, FDG uptake and distribution within the tumor mass is more complicated than initially envisioned.

Although the prevailing thought is an assumed direct relation between FDG uptake and tumor cells, this study could not find a statistical correlation between FDG activity (SUVmax or SUVmean) and quantitative malignant tumor cells or any other individual tumor cellular component. Although animal studies have suggested a correlation between FDG activity and tumor cells, these findings are not clearly applicable to a human population.33 Animal models use cell lines, and the animals usually lack a functional immune system to produce a host response that mimics human tumors.

Statistical analysis did identify a potentially significant positive correlation between SUVmean and tumor necrosis (P = .006), indicating that the higher the SUVmean the more likely the tumor is to contain necrotic material. This runs contrary to the conventional wisdom that tumor necrosis dilutes SUVmean and further underscores the point that it is still not known which cellular component(s) may be responsible for FDG uptake in NSCLC.

The results of the current study have implications for the practice of using PET to assess treatment response. Currently, response rates for NSCLC are determined by change in tumor size, as outlined by Response Evaluation Criteria In Solid Tumors (RECIST).34 However, a tumor that decreases in size after treatment may merely reflect improvement in the tumor-associated inflammatory component, and not necessarily represent a true tumor cell reduction.35 Because tumor cells alone do not contribute to FDG uptake, changes in FDG activity on serial PET studies as a marker of treatment response may not accurately reflect a change in tumor burden. These findings may explain why a change in FDG activity on PET after neoadjuvant therapy for early stage resectable NSCLC did not predict survival.28 Additional studies are needed to determine the clinical utility of serial PET imaging in this patient population.

This study does have several limitations. Only 5 high-powered fields per sample were evaluated, which may not provide an accurate assessment of tumor cell components. The sample size was small, and retrospective reviews potentially introduce a selection bias. In addition, we recognize that some reports suggest potential differences in absolute SUVs between PET and PET-CT, with studies reporting up to 15% higher values with CT-attenuation corrected imaging.36 However, a more recent study reported no significant difference.37 Although the potential for error is introduced, the effect is likely negligible due to the sample size and small degree of variance within our patient population. Analysis of the 33 patients with PET alone did not significantly alter the conclusions. SUV is also reported to vary in lesions at the lung bases due to respiratory motion.38 Of 40 patients, 9 tumors (23%) were located within the basilar segments of the lower lobes, lingula, or right middle lobe; the majority were located within the upper lobes with the right upper lobe being most common; a subset analysis did not change the results.

In summary, this study found that early stage NSCLCs are heterogeneous tumors with a wide variety of cellular components in varying proportions and no statistically significant correlation between tumor cellular components and FDG activity. Although PET has been a useful tool for evaluating lung cancer patients, the exact mechanism of FDG uptake remains unknown. A better understanding of tumor biology and the complex host response could potentially help suggest ways to improve imaging strategies. This is most evident in the PET assessment of treatment response, because a reduction in tumor size or FDG uptake may reflect either a decrease in viable tumor cells or, alternatively, the associated inflammatory component. New radiotracers, such as fluoro-3′-deoxy-L-thymidine (FLT) and fluoromisonidazole (FMISO), among others, may be of more value than FDG in this regard. Additional studies evaluating these agents and to further define factors influencing FDG activity may be helpful to refine the role of PET imaging in the management of patients with NSCLC.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES
  • 1
    Ahuja V, Coleman RE, Herndon J, Patz EF. The prognostic significance of fluorodeoxyglucose positron emission tomography imaging for patients with nonsmall cell lung carcinoma. Cancer. 1998; 83: 918-924.
  • 2
    Dhital K, Saunders CA, Seed PT, O'Doherty MJ, Dussek J. [(18)F]Fluorodeoxyglucose positron emission tomography and its prognostic value in lung cancer. Eur J Cardiothorac Surg. 2000; 18: 425-428.
  • 3
    Downey RJ, Akhurst T, Gonen M, et al. Preoperative F-18 fluorodeoxyglucose-positron emission tomography maximal standardized uptake value predicts survival after lung cancer resection. J Clin Oncol. 2004; 22: 3255-3260.
  • 4
    Cerfolio RJ, Bryant AS, Ohja B, Bartolucci AA. The maximum standardized uptake values on positron emission tomography of a nonsmall cell lung cancer predict stage, recurrence, and survival. J Thorac Cardiovasc Surg. 2005; 130: 151-159.
  • 5
    Duhaylongsod FG, Lowe VJ, Patz EF, Vaughn AL, Coleman RE, Wolfe WG. Lung tumor growth correlates with glucose metabolism measured by fluoride-18 fluorodeoxyglucose positron emission tomography. Ann Thorac Surg. 1995; 60: 1348-1352.
  • 6
    Duhaylongsod FG, Lowe VJ, Patz EF, Vaughn AL, Coleman RE, Wolfe WG. Detection of primary and recurrent lung cancer by means of F-18 fluorodeoxyglucose positron emission tomography (FDG PET). J Thorac Cardiovasc Surg. 1995; 110: 130-139; discussion 139-140.
  • 7
    Hellwig D, Gröschel A, Graeter TP, et al. Diagnostic performance and prognostic impact of FDG-PET in suspected recurrence of surgically treated nonsmall cell lung cancer. Eur J Nucl Med Mol Imaging. 2006; 33: 13-21.
  • 8
    Higashi K, Ueda Y, Arisaka Y, et al. 18F-FDG uptake as a biologic prognostic factor for recurrence in patients with surgically resected nonsmall cell lung cancer. J Nucl Med. 2002; 43: 39-45.
  • 9
    Jeong HJ, Min JJ, Park JM, et al. Determination of the prognostic value of [(18)F]fluorodeoxyglucose uptake by using positron emission tomography in patients with nonsmall cell lung cancer. Nucl Med Commun. 2002; 23: 865-870.
  • 10
    Marom EM, McAdams HP, Erasmus JJ, et al. Staging nonsmall cell lung cancer with whole-body PET. Radiology. 1999; 212: 803-809.
  • 11
    Patz EF, Connolly J, Herndon J. Prognostic value of thoracic FDG PET imaging after treatment for nonsmall cell lung cancer. AJR Am J Roentgenol. 2000; 174: 769-774.
  • 12
    Sasaki R, Komaki R, Macapinlac H, et al. [18F]fluorodeoxyglucose uptake by positron emission tomography predicts outcome of nonsmall-cell lung cancer. J Clin Oncol. 2005; 23: 1136-1143.
  • 13
    Vansteenkiste JF, Stroobants SG, Dupont PJ, et al. Prognostic importance of the standardized uptake value on (18)F-fluoro-2-deoxy-glucose-positron emission tomography scan in nonsmall-cell lung cancer: an analysis of 125 cases. Leuven Lung Cancer Group. J Clin Oncol. 1999; 17: 3201-3206.
  • 14
    Vesselle H, Freeman JD, Wiens L, et al. Fluorodeoxyglucose uptake of primary nonsmall cell lung cancer at positron emission tomography: new contrary data on prognostic role. Clin Cancer Res. 2007; 13: 3255-3263.
  • 15
    Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin. 2008; 58: 71-96.
  • 16
    Gail MH, Eagan RT, Feld R, et al. Prognostic factors in patients with resected stage I nonsmall cell lung cancer. A report from the Lung Cancer Study Group. Cancer. 1984; 54: 1802-1813.
  • 17
    Graziano SL. nonsmall cell lung cancer: clinical value of new biological predictors. Lung Cancer. 1997; 17( suppl 1): S37-S58.
  • 18
    Harpole DH, Herndon JE, Young WG, Wolfe WG, Sabiston DC. Stage I nonsmall cell lung cancer. A multivariate analysis of treatment methods and patterns of recurrence. Cancer. 1995; 76: 787-796.
  • 19
    Ichinose Y, Hara N, Ohta M, et al. Is T factor of the TNM staging system a predominant prognostic factor in pathologic stage I nonsmall-cell lung cancer? A multivariate prognostic factor analysis of 151 patients. J Thorac Cardiovasc Surg. 1993; 106: 90-94.
  • 20
    Johnson BE. Biologic and molecular prognostic factors--impact on treatment of patients with nonsmall cell lung cancer. Chest. 1995; 107: 287S-290S.
  • 21
    Lipford EH, Eggleston JC, Lillemoe KD, Sears DL, Moore GW, Baker RR. Prognostic factors in surgically resected limited-stage, nonsmall cell carcinoma of the lung. Am J Surg Pathol. 1984; 8: 357-365.
  • 22
    Naruke T, Goya T, Tsuchiya R, Suemasu K. Prognosis and survival in resected lung carcinoma based on the new international staging system. J Thorac Cardiovasc Surg. 1988; 96: 440-447.
  • 23
    Fischer B, Lassen U, Mortensen J, et al. Preoperative staging of lung cancer with combined PET-CT. N Engl J Med. 2009; 361: 32-39.
  • 24
    Dooms C, van Baardwijk A, Verbeken E, et al. Association between 18F-fluoro-2-deoxy-D-glucose uptake values and tumor vitality: prognostic value of positron emission tomography in early-stage nonsmall cell lung cancer. J Thorac Oncol. 2009; 4: 822-828.
  • 25
    de Geus-Oei LF, van der Heijden HF, Corstens FH, Oyen WJ. Predictive and prognostic value of FDG-PET in nonsmall-cell lung cancer: a systematic review. Cancer. 2007; 110: 1654-1664.
  • 26
    Hoang JK, Hoagland LF, Coleman RE, Coan AD, Herndon JE, Patz EF. Prognostic value of fluorine-18 fluorodeoxyglucose positron emission tomography imaging in patients with advanced-stage nonsmall-cell lung carcinoma. J Clin Oncol. 2008; 26: 1459-1464.
  • 27
    Kased N, Erasmus JJ, Komaki R, Cox JD. Prognostic value of posttreatment [18F] fluorodeoxyglucose uptake of primary nonsmall cell lung carcinoma treated with radiation therapy with or without chemotherapy: a brief review. J Thorac Oncol. 2008; 3: 534-538.
  • 28
    Tanvetyanon T, Eikman EA, Sommers E, Robinson L, Boulware D, Bepler G. Computed tomography response, but not positron emission tomography scan response, predicts survival after neoadjuvant chemotherapy for resectable nonsmall-cell lung cancer. J Clin Oncol. 2008; 26: 4610-4616.
  • 29
    Marom EM, Aloia TA, Moore MB, et al. Correlation of FDG-PET imaging with Glut-1 and Glut-3 expression in early-stage nonsmall cell lung cancer. Lung Cancer. 2001; 33: 99-107.
  • 30
    Oremek GM, Sauer-Eppel H, Bruzdziak TH. Value of tumour and inflammatory markers in lung cancer. Anticancer Res. 2007; 27: 1911-1915.
  • 31
    Petersen RP, Campa MJ, Sperlazza J, et al. Tumor infiltrating Foxp3+ regulatory T-cells are associated with recurrence in pathologic stage I NSCLC patients. Cancer. 2006; 107: 2866-2872.
  • 32
    Maschauer S, Prante O, Hoffmann M, Deichen JT, Kuwert T. Characterization of 18F-FDG uptake in human endothelial cells in vitro. J Nucl Med. 2004; 45: 455-460.
  • 33
    Brown RS, Leung JY, Fisher SJ, Frey KA, Ethier SP, Wahl RL. Intratumoral distribution of tritiated fluorodeoxyglucose in breast carcinoma: I. Are inflammatory cells important? J Nucl Med. 1995; 36: 1854-1861.
  • 34
    Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst. 2000; 92: 205-216.
  • 35
    Birchard KR, Hoang JK, Herndon JE, Patz EF. Early changes in tumor size in patients treated for advanced stage nonsmall cell lung cancer do not correlate with survival. Cancer. 2009; 115: 581-586.
  • 36
    Nakamoto Y, Osman M, Cohade C, et al. PET/CT: comparison of quantitative tracer uptake between germanium and CT transmission attenuation-corrected images. J Nucl Med. 2002; 43: 1137-1143.
  • 37
    Souvatzoglou M, Ziegler SI, Martinez MJ, et al. Standardised uptake values from PET/CT images: comparison with conventional attenuation-corrected PET. Eur J Nucl Med Mol Imaging. 2007; 34: 405-412.
  • 38
    Erdi YE, Nehmeh SA, Pan T, et al. The CT motion quantitation of lung lesions and its impact on PET-measured SUVs. J Nucl Med. 2004; 45: 1287-1292.