Radiology: Perfusion CT provides lung cancer angiogenesis evaluation
Perfusion CT may offer an effective technique for the evaluation of lung cancer angiogenesis, depicting significant differences in anti-angiogenetic treatment response on the whole tumor mass, according to a study published in the May issue of Radiology.
Lung tumors remain the most deadly neoplasms, resulting in five-year survival of just 10 percent for advanced stage diagnoses despite improvements in treatment. Perfusion CT has played an expanding role in the diagnosis and follow-up of lung cancer, though most such research has focused on the evaluation of restricted areas within large lesions, rather than on whole-tumor perfusion.
“The recent introduction of faster scanners allows wider z-axis coverage and may guarantee a reliable volumetric evaluation of perfusion parameters within large areas of the thorax and abdomen, enabling whole tumor evaluation,” explained Francesco Fraioli, MD, of the department of radiological sciences at the University of Rome “La Sapienza” in Italy, and colleagues.
Fraioli and co-authors investigated the capability of wide-volume perfusion CT to evaluate the effects of chemotherapy and antiangiogenetic treatment on the whole tumor mass in patients with locally advanced adenocarcinoma. Moreover, the authors assessed the correlation between conventional response evaluation criteria in solid tumors (RECIST) and changes in CT numbers in response to the therapy.
Forty-five patients with unresectable tumors that were 20 mm or larger underwent baseline perfusion CT just before treatment and follow-up after 40 days, while 14 patients were available for 90-day follow-up.
Baseline perfusion CT exams were rated of optimal quality in 29 of 45 cases, adequate in 15 patients and poor in one case. For 40-day follow-up, 27 studies were seen as optimal, 17 as adequate and one as poor. Ninety-day follow-up showed no images of poor quality, three rated as adequate and 11 as optimal.
Overall, kappa agreement between the baseline and 40-day studies was strong, at 0.909. Agreement by the two study readers also was strong across perfusion parameters. Blood volume, time to peak and permeability were measured at 0.874, 0.85 and 0.911 for baseline, respectively. At 40-day follow up, blood volume agreement reached 0.893, time to peak at 0.887 and permeability at 0.939.
Strong intraobserver agreement was also examined across blood flow, blood volume, time to peak and permeability, with baseline and follow-up kappa scores ranging from 0.884 to 0.979.
Patient evaluation using RECIST revealed that no individuals exhibited complete responses to therapy, while 20 percent were classified as partially responsive, 9 percent as having progressive disease and 71 percent as being stable. Patients who showed a partial response to treatment had a mean tumor size of 3.8 cm at baseline, which dropped to 2.5 cm following therapy.
Inter- and intraobserver agreement for the evaluation of RECIST parameters was very high from baseline (intraclass correlation coefficient [ICC], 0.96 to 0.98, respectively) through second follow-up (ICC 0.94 and 0.97, respectively).
“In our study, we tried to demonstrate if perfusion CT allows evaluation and in some way prediction of the effects of chemotherapy combined with antiangiogenetic drugs and if changes in CT numbers correlate with the response to therapy as assessed with RECIST criteria,” Fraioli and colleagues wrote.
“As for the first point, we demonstrated that treatment-induced changes in perfusion parameters may be identified at perfusion CT, even if the lack of a reproducibility analysis partially depletes this finding. Moreover, an exhaustive prediction of the treatment outcome on the basis of perfusion CT parameters of the lesions at baseline examinations is presently out of range.”
The authors acknowledged their small sample size and lack of histological validation limited their findings.
“Probably the most interesting aspect of our results,” Fraioli and co-authors argued, “is represented by the discrepancies between RECIST and perfusion CT measurements. In 36 percent of the subjects, the size of the lesion at the 42-day follow-up was considered stable with RECIST evaluation but vascularization was increased, suggesting a poor response to treatment (in all 16 subjects blood volume was increased and in three subjects blood flow was also increased).”
On the other hand, the authors noted that decreased blood flow, blood volume, time to peak and permeability were found in 11 percent of patients with RECIST-classified stable disease, while 24 percent of these patients exhibited a decrease of blood flow and blood volume, “suggesting some sort of response to therapy.”
“In conclusion, our results suggest that perfusion CT may allow evaluation of lung cancer angiogenesis demonstrating alterations in vascularity following treatment; these data may provide additional information about treatment outcome,” Fraioli and colleagues stated.
Lung tumors remain the most deadly neoplasms, resulting in five-year survival of just 10 percent for advanced stage diagnoses despite improvements in treatment. Perfusion CT has played an expanding role in the diagnosis and follow-up of lung cancer, though most such research has focused on the evaluation of restricted areas within large lesions, rather than on whole-tumor perfusion.
“The recent introduction of faster scanners allows wider z-axis coverage and may guarantee a reliable volumetric evaluation of perfusion parameters within large areas of the thorax and abdomen, enabling whole tumor evaluation,” explained Francesco Fraioli, MD, of the department of radiological sciences at the University of Rome “La Sapienza” in Italy, and colleagues.
Fraioli and co-authors investigated the capability of wide-volume perfusion CT to evaluate the effects of chemotherapy and antiangiogenetic treatment on the whole tumor mass in patients with locally advanced adenocarcinoma. Moreover, the authors assessed the correlation between conventional response evaluation criteria in solid tumors (RECIST) and changes in CT numbers in response to the therapy.
Forty-five patients with unresectable tumors that were 20 mm or larger underwent baseline perfusion CT just before treatment and follow-up after 40 days, while 14 patients were available for 90-day follow-up.
Baseline perfusion CT exams were rated of optimal quality in 29 of 45 cases, adequate in 15 patients and poor in one case. For 40-day follow-up, 27 studies were seen as optimal, 17 as adequate and one as poor. Ninety-day follow-up showed no images of poor quality, three rated as adequate and 11 as optimal.
Overall, kappa agreement between the baseline and 40-day studies was strong, at 0.909. Agreement by the two study readers also was strong across perfusion parameters. Blood volume, time to peak and permeability were measured at 0.874, 0.85 and 0.911 for baseline, respectively. At 40-day follow up, blood volume agreement reached 0.893, time to peak at 0.887 and permeability at 0.939.
Strong intraobserver agreement was also examined across blood flow, blood volume, time to peak and permeability, with baseline and follow-up kappa scores ranging from 0.884 to 0.979.
Patient evaluation using RECIST revealed that no individuals exhibited complete responses to therapy, while 20 percent were classified as partially responsive, 9 percent as having progressive disease and 71 percent as being stable. Patients who showed a partial response to treatment had a mean tumor size of 3.8 cm at baseline, which dropped to 2.5 cm following therapy.
Inter- and intraobserver agreement for the evaluation of RECIST parameters was very high from baseline (intraclass correlation coefficient [ICC], 0.96 to 0.98, respectively) through second follow-up (ICC 0.94 and 0.97, respectively).
“In our study, we tried to demonstrate if perfusion CT allows evaluation and in some way prediction of the effects of chemotherapy combined with antiangiogenetic drugs and if changes in CT numbers correlate with the response to therapy as assessed with RECIST criteria,” Fraioli and colleagues wrote.
“As for the first point, we demonstrated that treatment-induced changes in perfusion parameters may be identified at perfusion CT, even if the lack of a reproducibility analysis partially depletes this finding. Moreover, an exhaustive prediction of the treatment outcome on the basis of perfusion CT parameters of the lesions at baseline examinations is presently out of range.”
The authors acknowledged their small sample size and lack of histological validation limited their findings.
“Probably the most interesting aspect of our results,” Fraioli and co-authors argued, “is represented by the discrepancies between RECIST and perfusion CT measurements. In 36 percent of the subjects, the size of the lesion at the 42-day follow-up was considered stable with RECIST evaluation but vascularization was increased, suggesting a poor response to treatment (in all 16 subjects blood volume was increased and in three subjects blood flow was also increased).”
On the other hand, the authors noted that decreased blood flow, blood volume, time to peak and permeability were found in 11 percent of patients with RECIST-classified stable disease, while 24 percent of these patients exhibited a decrease of blood flow and blood volume, “suggesting some sort of response to therapy.”
“In conclusion, our results suggest that perfusion CT may allow evaluation of lung cancer angiogenesis demonstrating alterations in vascularity following treatment; these data may provide additional information about treatment outcome,” Fraioli and colleagues stated.