Hubert Vesselle

PET-CT image registration in the chest using free-form deformations

We have implemented and validated an algorithm for three-dimensional positron emission tomography transmission-to-computed tomography registration in the chest, using mutual information as a similarity criterion. Inherent differences in the two imaging protocols produce significant nonrigid motion between the two acquisitions. A rigid body deformation combined with localized cubic B-splines is used to capture this motion. The deformation is defined on a regular grid and is parameterized by potentially several thousand coefficients. Together with a spline-based continuous representation of images and Parzen histogram estimates, our deformation model allows closed-form expressions for the criterion and its gradient. A limited-memory quasi-Newton optimization algorithm is used in a hierarchical multiresolution framework to automatically align the images. To characterize the performance of the method, 27 scans from patients involved in routine lung cancer staging were used in a validation study. The registrations were assessed visually by two expert observers in specific anatomic locations using a split window validation technique. The visually reported errors are in the 0- to 6-mm range and the average computation time is 100 min on a moderate-performance workstation.

Phenotypic heterogeneity of end-stage prostate carcinoma metastatic to bone ☆

To better understand the clinical and pathologic features of end-stage, androgen-independent carcinoma of the prostate (CaP), we performed rapid autopsies on 14 men who died of progressive CaP and recorded relevant clinical data. The timing of tumor progression varied widely. The median time to androgen independence was 2 years (range, 4 months to 13.6 years). The median survival after androgen independence was 1 year (range, 1 month to 3.6 years). Because osseous metastases are prevalent in progressive CaP, up to 20 bone sites were systematically sampled in each patient. Bone metastases were widespread; tumor filled the marrow in an average of 14 bone sites. Tumor histology and expression of prostate-specific antigen (PSA) and chromogranin A (CGA) were examined in all metastases and were compared with the primary tumor. Five histological patterns of metastatic tumor were observed: solid (10 patients), macroacinar (1 patient), microacinar (1 patient), clear cell (1 patient), and comedocarcinoma (1 patient). Gleason grade of the primary tumor did not predict the histological pattern of the metastases. Although >70% of tumor cells expressed PSA, the fraction of PSA-positive cells varied widely in separate metastases in some patients (standard deviation >25). Likewise, the fraction of neuroendocrine (NE) (CGA-positive) tumor cells in different metastases varied widely. For example, between 0 and 95% of tumor cells in different metastases in 1 patient had a NE phenotype. The present study highlights the heterogeneity--histologically and immunophenotypically--of metastatic CaP. Consequently, therapy directed to the phenotype of 1 metastasis may have no effect on other metastases in the same patient because of phenotypic heterogeneity.

Nonrigid multimodality image registration

We have designed, implemented, and validated an algorithm capable of 3D PET-CT registration in the chest, using mutual information as a similarity criterion. Inherent differences in the imaging protocols produce significant non-linear motion between the two acquisitions. To recover this motion, local deformations modeled with cubic B-splines are incorporated into the transformation. The deformation is defined on a regular grid and is parameterized by potentially several thousand coefficients. Together with a spline-based continuous representation of images and Parzen histogram estimates, the deformation model allows for closed-form expressions of the criterion and its gradient. A limited-memory quasi-Newton optimization package is used in a hierarchical multiresolution framework to automatically align the images. To characterize the performance of the algorithm, 27 scans from patients involved in routine lung cancer screening were used in a validation study. The registrations were assessed visually by two observers in specific anatomic locations using a split window validation technique. The visually reported errors are in the 0-6mm range and the average computation time is 100 minutes.

DNA Methylation in Tumor and Matched Normal Tissues from Non-Small Cell Lung Cancer Patients

We used MethyLight assays to analyze DNA methylation status of 27 genes on 49 paired cancerous and noncancerous tissue samples from non-small cell lung cancer (NSCLC) patients who underwent surgical resection. Seven genes (RARB, BVES, CDKN2A, KCNH5, RASSF1, CDH13, and RUNX) were found to be methylated significantly more frequently in tumor tissues than in noncancerous tissues. Only methylation of CCND2 and APC was frequently detected in both cancerous and noncancerous tissues, supporting the hypothesis that the methylation of these two genes is a preneoplastic change and may be associated with tobacco smoking exposure. Methylation of any one of eight genes (RASSF1, DAPK1, BVES, CDH13, MGMT, KCNH5, RARB, or CDH1) was present in 80% of NSCLC tissues but only in 14% of noncancerous tissues. Detection of methylation of these genes in blood might have utility in monitoring and detecting tumor recurrence in early-stage NSCLC after curative surgical resection. (Cancer Epidemiol Biomarkers Prev 2008;17(3):645–54)

Fluorodeoxyglucose Uptake of Primary Non-Small Cell Lung Cancer at Positron Emission Tomography: New Contrary Data on Prognostic Role

Purpose: This prospective study evaluated the prognostic significance of 18F-fluorodeoxyglucose (18F-FDG) uptake in primary non-small cell lung cancer (NSCLC) at positron emission tomography, in a carefully staged population, while correcting for partial volume effects. Experimental Design: Two hundred eight potentially resectable NSCLC patients were referred for FDG positron emission tomography staging after thoracic computed tomography. Each tumor stage was confirmed surgically, or for some stage IV tumors by additional imaging. The tumor maximum pixel-standardized uptake value (maxSUV) and the maxSUV partial volume corrected for lesion size (PVCmaxSUV) were compared with overall survival and disease-free survival using Cox proportional hazards regression. Results: Stage distribution: stage I, 36%; stage II, 15%; stage III, 30%; stage IV, 19%. Patients were followed for a median of 33.6 months, with 90 deaths from NSCLC (median survival for all stages, 43.3 months). With respect to overall survival, the most significant cutoff value for both maxSUV and PVCmaxSUV was 7. MaxSUV ≥7 was significantly associated with an increased risk of death from NSCLC in univariable analysis, whereas PVCmaxSUV ≥7 was only marginally associated. However, in multivariable analyses, neither maxSUV ≥7 nor PVCmaxSUV ≥7 provided significant additional prognostic information over stage, tumor size, and age. In the 103 patients who underwent surgical resection only, surgical stage, but not maxSUV or PVCmaxSUV, was univariably associated with survival or recurrence. SUV definitions based on lean body mass, body surface area, and plasma glucose correction yielded identical results. Conclusions: As expected, tumor stage is prognostic in NSCLC. However, tumor FDG uptake does not provide additional prognostic information. This prospective study contradicts prior reports.

Relationship Between Non-small Cell Lung Cancer FDG Uptake at PET, Tumor Histology, and Ki-67 Proliferation Index

Introduction: We compared primary non-small cell lung cancer (NSCLC) 18F-fluorodeoxyglucose (FDG) uptake at positron emission tomography (PET) to tumor histologic features and Ki-67 proliferation index. This large, prospectively-recruited patient cohort has previously been analyzed based on differences in FDG uptake across stage groups; the current analysis adds further dimensions to this characterization. Materials and Methods: One hundred seventy-eight patients with potentially-resectable NSCLC were scanned with FDG PET before therapy. A partial volume correction algorithm was used to correct FDG uptake values for their dependence on tumor size. Primary tumor resection specimens, core biopsies, and biopsies of metastatic lymph nodes were used to assess each tumor’s NSCLC histologic subtype, degree of differentiation, and Ki-67 proliferation index. Results: Bronchioalveolar carcinomas were found to have lower FDG uptake at PET and lower Ki-67 scores than any other histologic subtype. Non-bronchioalveolar adenocarcinomas had lower FDG uptake and Ki-67 scores than squamous cell carcinomas or large cell undifferentiated carcinomas. Better differentiated NSCLCs had lower FDG uptake and Ki-67 scores than more poorly differentiated NSCLCs. There was a significant positive correlation between FDG uptake and Ki-67 scores. Partial volume correction increased the strength of this correlation, while also diminishing the strong positive correlation between FDG uptake and tumor size. Conclusions: There are significant differences in NSCLC FDG uptake across histologic subtypes and differentiation groups. These differences parallel nearly identical differences in Ki-67 scores, implying that differences in NSCLC tumor cell proliferation may give rise to commensurate differences in tumor glucose metabolism.

DNA hypermethylation of tumors from non-small cell lung cancer (NSCLC) patients is associated with gender and histologic type.

BACKGROUND We previously identified a number of genes which were methylated significantly more frequently in the tumor compared to the non-cancerous lung tissues from non-small cell lung cancer (NSCLC) patients. Detection of methylation profiles of genes in NSCLC could provide insight into differential pathways to malignancy and lead to strategies for better treatment of individuals with NSCLC. METHODS We determined the DNA methylation status of 27 genes using quantitative MethyLight assays in lung tumor samples from 117 clinically well-characterized NSCLC patients. RESULTS Hypermethylation was detected in one of more of the genes in 106 (91%) of 117 cases and was detected at high levels (percentage methylation reference (PMR)> or =4%) in 79% of NSCLC cases. Methylation of APC, CCND2, KCNH5 and, RUNX was significantly more frequent in adenocarcinomas compared to squamous cell carcinomas (SCC), while methylation of CDKN2A was more common in SCC. Hypermethylation of KCNH5, KCNH8, and RARB was more frequent in females compared to males. Hypermethylation of APC and CCND2 was inversely associated with proliferation score assessed by Ki-67 level. CONCLUSIONS Our findings of differential gene hypermethylation frequencies in tumor tissues from patients with adenocarcinoma or squamous cell cancers and in females compared to males suggests that further investigation is warranted in order to more fully understand the potential disparate pathways and/or risk factors for NSCLC associated with histologic type and gender.

Tumor 3′-Deoxy-3′-18F-Fluorothymidine (18F-FLT) Uptake by PET Correlates with Thymidine Kinase 1 Expression: Static and Kinetic Analysis of 18F-FLT PET Studies in Lung Tumors

We report the first, to our knowledge, findings describing the relationships between both static and dynamic analysis parameters of 3′-deoxy-3′-18F-fluorothymidine (18F-FLT) PET and the expression of the proliferation marker Ki-67, and the protein expression and enzymatic activity of thymidine kinase-1 (TK1) in surgically resected lung lesions. Methods: Static and dynamic analyses (4 rate constants and 2 compartments) of 18F-FLT PET images were performed in a cohort of 25 prospectively accrued, clinically suspected lung cancer patients before surgical resection (1 lesion was found to be benign after surgery). The maximal and overall averaged expression of Ki-67 and TK1 were determined by semiquantitative analysis of immunohistochemical staining. TK1 enzymatic activity was determined by in vitro assay of extracts prepared from flash-frozen samples of the same tumors. Results: Static 18F-FLT uptake (partial-volume–corrected maximum-pixel standardized uptake value from 60- to 90-min summed dynamic data) was significantly correlated with the overall (ρ = 0.57, P = 0.006) and maximal (ρ = 0.69, P < 0.001) immunohistochemical expressions of Ki-67 and TK1 (overall expression: ρ = 0.65, P = 0.001; maximal expression: ρ = 0.68, P < 0.001) but not with TK1 enzymatic activity (ρ = 0.34, P = 0.146). TK1 activity was significantly correlated with TK1 protein expression only when immunohistochemistry was scored for maximal expression (ρ = 0.52, P = 0.029). Dynamic analysis of 18F-FLT PET revealed correlations between the flux constant (KFLT) and both overall (ρ = 0.53, P = 0.014) and maximal (ρ = 0.50, P = 0.020) TK1 protein expression. KFLT was also associated with both overall (ρ = 0.59, P = 0.005) and maximal (ρ = 0.63, P = 0.002) Ki-67 expression. We observed no significant correlations between TK1 enzyme activity and KFLT. In addition, no significant relationships were found between TK1 expression, TK1 activity, or Ki-67 expression and any of the compartmental rate constants. Conclusion: The absence of observable correlations of the imaging parameters with TK1 activity suggests that 18F-FLT uptake and retention within cells may be complicated by a variety of still undetermined factors in addition to TK1 enzymatic activity.

Relationship between Non-Small Cell Lung Cancer Fluorodeoxyglucose Uptake at Positron Emission Tomography and Surgical Stage with Relevance to Patient Prognosis

Purpose: Because the tumor stage is the most significant prognostic factor for non-small cell lung cancer (NSCLC) and given that NSCLC [18F]fluorodeoxyglucose (18F-FDG) uptake appears to have prognostic significance, we examined the relationship between NSCLC 18F-FDG uptake and surgical stage. Experimental Design: One hundred seventy-eight patients with a proven diagnosis of NSCLC were enrolled, then imaged with 18F-FDG positron emission tomography and their disease thoroughly staged. Primary tumor size at computed tomography and 18F-FDG uptake were compared to overall tumor stage and to T, N, and M stage descriptors. Tumor uptake was quantitated by maximum pixel-standardized uptake value (maxSUV) and then partial volume corrected for lesion size using recovery coefficients. Results: A significant difference in tumor size was associated with tumors of different TNM stage, T status, N status, or M status. Similarly, the primary tumor maxSUV was significantly associated with TNM stage, T status, and M status. However, we observed no significant difference in the partial-volume-corrected tumor maxSUV for different stages; different T, N, or M descriptors; tumors without evidence of spread (N0M0) versus tumors with nodal spread (N1,2,3M0); or tumors without spread (N0M0) versus all others. Conclusions: We found an association between tumor stage and 18F-FDG maxSUV, but this relationship disappeared after correction of tumor uptake for lesion size. Therefore, if partial-volume-corrected 18F-FDG uptake is prognostic of NSCLC outcome, it is not on the basis of a relationship with tumor stage but through a different mechanism.

The Ki-67 index and survival in non-small cell lung cancer: a review and relevance to positron emission tomography.

PURPOSEWe reviewed the current literature to discover the range of studies covering tissue-based and noninvasive methods for determining tumor stage and the prognostic value of staging in non-small cell lung cancer. DESIGNDespite refinements in staging of non-small cell lung cancer, each stage remains heterogeneous because each stage contains patients who are at higher risk for recurrence than other patients within the same stage. Tissue-based and noninvasive methods have been investigated to complement tumor stage in assessing non-small cell lung cancer prognosis. The prognostic significance of tumor proliferation assessed by Ki-67 protein expression has been demonstrated in non-small cell lung cancer. RESULTRecent positron emission tomography studies have also shown both prognostic value in non-small cell lung cancer uptake of [F-18] fluorodeoxyglucose (FDG) and correlation between non-small cell lung cancer FDG uptake and tumor proliferation. DISCUSSIONWe reviewed the prognostic significance of Ki-67 expression in non-small cell lung cancer and related it to positron emission tomography.