- Email: [email protected]
Computer tomography in pulmonary invasive aspergillosis in hematological patients with neutropenia: An useful tool for diagnosis and assessment of outcome in clinical trials
European Journal of Radiology 74 (2010) e172–e175
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Computer tomography in pulmonary invasive aspergillosis in hematological patients with neutropenia: An useful tool for diagnosis and assessment of outcome in clinical trials Denis Caillot a,∗ , Valérie Latrabe b , Anne Thiébaut c , Raoul Herbrecht d , Stéphane De Botton e , Arnaud Pigneux b , Franc¸oise Monchecourt f , Lamine Mahi f , Serge Alfandari e , Jean-Franc¸ois Couaillier a a
Dijon University Hospital Center, Dijon, France Bordeaux University Hospital Center, Bordeaux, France c Lyon University Hospital Center, Lyon, France d Strasbourg Regional University Hospital Center, Strasbourg, France e Lille Regional University Hospital Center, Lille, France f Gilead Sciences, Paris, France b
a r t i c l e
i n f o
Article history: Received 22 August 2008 Received in revised form 9 April 2009 Accepted 25 May 2009 Keywords: Thoracic CT scan Pulmonary invasive aspergillosis Neutropenia Acute myeloid leukaemia
a b s t r a c t Background and objective: The exact timing of the evolution of lesion volumes of invasive pulmonary aspergillosis (IPA) on CT scan images could be helpful in the management of hematological patients but has never been evaluated in a prospective study. We analyzed the CT scan data from the prospective Combistrat trial. Design and methods: Volumes of aspergillosis lesions from 30 patients (including 24 acute myeloid leukaemia) with probable (n = 26) or proven (n = 4) IPA according to the EORTC-MSG modified criteria, were measured prospectively on the thoracic CT scans at the enrolment in the study on day 0 (D0), D7, D14 and end of treatment (EOT). Results: For the overall population, the volume of pulmonary aspergillosis lesions increased significantly from D0 to D7 (1.6 fold; p = 0.003). Then this volume decreased significantly from D7 to D14 (1.36 fold at D14 with p = 0.003 for D14 vs. D7, but with p = 0.56 for D14 vs. D0). At EOT (= D17, median value), the volume of lesions was significantly lower than D14 (0.76 fold the initial volume; p < 0.001) but it was not significantly different when compared to D0 (p = 0.11). Conclusions: The results of this prospective study suggest that the sequential analysis of CT scan in neutropenic patients with IPA depicts more precisely the evolution of lesion volumes than comparison to baseline images. Moreover, the systematic use of chest CT appears to be a useful tool for diagnosis and outcome evaluation of IPA in clinical trials. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Invasive aspergillosis (IA) in neutropenic patients remains lifethreatening. Therefore reliable and timely diagnosis is an important objective for rapid treatment initiation because IA jeopardizes the planning of chemotherapy or hematopoietic stem cell transplantation. Thoracic CT scan may detect pulmonary aspergillosis at an early stage of infection [1,2]. Thus, the halo sign (a mass-like infil-
∗ Corresponding author at: Service d’Hématologie, C.H.U. de Dijon, Hôpital du Bocage, Hôpital d’Enfants, 10, boulevard du Maréchal de Lattre de Tassigny, B.P. 1542, 21034 Dijon, France. Tel.: +33 3 80 29 50 41. E-mail address: [email protected] (D. Caillot). 0720-048X/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2009.05.058
trate surrounded by ground-glass opacity) is considered as an early sign of invasive pulmonary aspergillosis (IPA) [3]. Until now the evolution of CT scan images has been poorly studied although a better knowledge of anatomical evolution of lesions could help decisions for treatment continuation or switching. In a previous paper we studied the timing of CT images in hematological patients with neutropenia [4]. We observed that, despite early initiation of antifungal treatment, the size of the lesions of IPA increased during the first days of the disease (between 0 and day 7). We concluded that an increase of the size of aspergillosis lesion during the first days after diagnosis was not associated with a pejorative immediate outcome. This apparent failure of the antifungal treatment could lead clinicians to initiate other antifungal treatments supposed to be more efficient. However, this study selected
D. Caillot et al. / European Journal of Radiology 74 (2010) e172–e175
retrospectively patients with surgically proven invasive aspergillosis (according to the EORTC/MSG criteria) [5]. So, surgery could have subsequently modified the outcome of these patients. Therefore, it was of interest to study in a prospective study whether these observations could be extended to patients with probable IPA most often treated without surgery. We had the opportunity to answer this issue with the Combistrat study that included a majority of patients with probable IPA [6]. The aim of the Combistrat study was to evaluate prospectively a combination of liposomal amphotericin B and caspofungin in comparison with liposomal amphotericin B alone. In this article, we present the analysis of the sequential images of pulmonary lesions performed prospectively during the Combistrat trial. 2. Patients and methods 2.1. Study population The present analysis was based upon thoracic CT scan data gathered from the Combistrat study that has been published elsewhere [6]. Briefly, the Combistrat study was a national, multicenter, pilot, prospective, randomized open trial in patients with proven or probable IA according to criteria of the European Organization for Research and Treatment of Cancer/Mycosis Study Group (EORTC/MSG)[5]. 2.2. CT scan image database The database of digitized images used for this analysis comprised the thoracic CT scans from patients enrolled in the Combistrat study. Thoracic CT scans analyzed in the present study were performed for diagnosis, outcome assessment and follow-up of patients. The Combistrat study included hematological patients with neutropenia with probable or proven IA according to the criteria of the EORTC/MSG. Proven IA diagnosis required identification of typical hyphal elements after histopathological or cytopathological examination with evidence of associated tissue damage (either microscopically or unequivocally by imaging) or growth of Aspergillus organisms from a sample obtained by sterile procedure from a normally sterile site and clinically or radiologically abnormal site consistent with infection (excluding urine and mucous membrane). Probable IA diagnosis required at least one host factor criterion and one microbiological criterion and one major (or two minor) clinical criteria from abnormal site consistent with infection and no other pathogen detected to account for the clinical or radiographic signs of infection. As a modification of EORTC/MSG criteria, diagnosis of probable IA included also patients with recent neutropenia (<500 neutrophils/mm3 ) within 14 days of study inclusion and “halo” or “air crescent” signs on thoracic CT scan [5]. Note that this modification of criteria for probable IA has been recently included in the Revised Definitions of Invasive Fungal Disease of EORTC/MSG [7]. In addition, serum galactomannan assay (PlateliaTM , Bio-Rad Laboratories, Hercules, CA and Bio-Rad, Marnes-la-Coquette, France) with an optical density index of ≥1.0 was considered positive. To ensure uniformity of assessments, the CT scan diagnosis of IA was verified by a centrally data review board including one hematologist and one radiologist who confirmed IA. CT scans, when initially positive, were repeated at days 7, 14, 28 and then once a month, if applicable, for radiographic response assessments. The measurement of lesion volumes was also centralized. As it was hypothesized that aspergillary lesion had an ovoid shape, the calculated volume (in cm3 ) of each aspergillosis
e173
pulmonary lesion was obtained using the following formula: height × length × wide × /6). For several lesions, the total volume was obtained by addition of all elementary volumes. 2.3. Assessment of efficacy Complete (cure) and partial (improvement) overall responses were considered as favorable responses. Stable overall response and failure were considered as unfavorable responses. Complete response was defined as complete resolution of all signs and symptoms attributable to the invasive fungal infection as compared to baseline and complete or near complete clearing of CT scan abnormalities associated with active fungal infection (or persistent residual scarring only) and eradication of the pathogen. Partial response was defined as meaningful improvement of all clinical signs and symptoms attributable to IA and a decrease of at least 50% of radiographic signs. Stable response was defined as minor or no improvement but no worsening of attributable signs, symptoms, radiographic and/or bronchoscopic findings. Failure was defined as progression of infection, based on an increase in the number and/or severity of clinical signs and symptoms attributable to the IA, worsening of CT scan abnormalities consistent with progressive infection and persistently positive cultures, histopathological findings or positive galactomannan assays. 2.4. Statistical analysis The changes of lesion volumes were compared using Wilcoxon’s test. A p-value below 0.05 was considered as statistically significant. 3. Results 3.1. Patients of the Combistrat trial Baseline characteristics and clinical results of the Combistrat study have been described elsewhere [6]. Briefly, among the 32 patients enrolled in the Combistrat trial, 30 met the diagnosis of proven (n = 4) or probable (n = 26) IA according to a centrally data review board. At baseline, all the patients were neutropenic and had clinical signs. The initial thoracic CT scans were abnormal for all patients. In addition, presence of CT halo sign, positive galactomannan assay (two successive positive samples with an optical density index ≥1), culture or direct examination of bronchoalveolar lavage contributed to diagnosis for 19, 12 and 5 patients, respectively. The majority of patients were male (21/30; 70%) and the age ranged from 16 to 75 y (median, 58.5 y). Twenty-four patients had acute myeloblastic leukemia; other underlying conditions were chronic lymphocytic leukemia (n = 3), myeloproliferative disorders (n = 2) and acute lymphoblastic leukemia (n = 1). Median duration of neutropenia at inclusion was 23 days. Overall, at EOT (primary efficacy endpoint), 14 of 30 patients had a favorable partial or complete response. At EOT, 29/30 patients were alive and at week 12, all the patients of combination group were alive while 3 of 15 died in the high-dose group. 3.2. Evolution of the volumes of lesions in the overall population The volumes of the thoracic aspergillosis lesions were calculated for each patient. As evidenced in Fig. 1, for overall population, the volume increased significantly from D0 to D7 (1.6 fold; p = 0.003). After D7, the volume decreased and its variation between D7 and D14 was also statistically significant (1.6 vs. 1.36 fold the initial volume, p = 0.003). Moreover, at EOT (= D17, median value), the volume of lesions was significantly lower than at D14 (0.76 fold the
e174
D. Caillot et al. / European Journal of Radiology 74 (2010) e172–e175
4. Discussion
Fig. 1. Evolution of volume of lesions of pulmonary aspergillosis. The volumes of the thoracic aspergillosis lesions increased significantly from D0 to D7 (1.6 fold). After D7 the volume decreased and the volume variation between D7 and D14 was also statistically significant (1.36 fold the initial volume). Median of end of treatment (EOT) was 17 days after D0 and at EOT the volume was lower than the D14 volume (0.76 the initial volume).
initial volume; p < 0.001). At EOT and D14, the volumes were not significantly different when compared to D0 (p = 0.11 for EOT and p = 0.56 for D14). In patients with favorable outcome at EOT, the total volume of lesions between D0 and D14 was significantly higher (90, 60 and 44 cm3 at D0, D7, D14; median values) as compared to patients with unfavorable outcome (14, 38 and 33 cm3 , respectively; Anova test, p = 0.015). In contrast, as expected, the lesion volumes appeared to be smaller at EOT in patients with favorable outcome in comparison in patients with unfavorable outcome (29 vs. 58 cm3 , respectively). Fig. 2 presents successive thoracic CT scans of a patient with probable invasive pulmonary aspergillosis (IPA).
3.3. Prognostic factors for favorable overall response An increase of the total volume of lesions between D0 and D7 was observed for 19/26 patients (73%). Favorable overall response at EOT was observed for 7/19 (42%) patients with increase of volume and for 5/7 (71%) patients without increase of volume. A positive galactomannan assay (two successive positive samples with an optical density index ≥1) was present at baseline for 12/30 (40%) patients. Favorable overall response at EOT was observed for 5/12 (50%) patients with positive galactomannan assay and 7/18 (39%) patients with negative galactomannan assay at baseline. The halo sign was present at D0 for 19/30 patients (63%). Favorable overall response at EOT was observed for 8/19 (42%) patients with halo at baseline and for 6/11 (54%) patients without halo at baseline.
In a previous retrospective study in neutropenic patients with proven IPA, we showed that the volume of aspergillosis lesions evaluated by sequential thoracic CT scans increased during the first days after antifungal treatment and was not correlated with pejorative outcome [4]. An increase of the size of pulmonary aspergillosis lesions during the first days of treatment was also reported in the recent study of Brodoefel et al. in 40 non-HIV immunocompromised patients with confirmed pulmonary IA [8]. There were 26/30 patients in the Combistrat study with probable IPA. Therefore the present data allowed us to study whether our previous observation of an initial increase of lesion size could be extended to probable IPA in a prospective trial. We observed indeed that 19/26 patients had an increase of lesion volume between D0 and D7. Overall, the increase of median lesion size from D0 to D7 was statistically significant and was followed by a significant decrease from D7 to D14 and then from D14 to EOT. This indicates that the counterintuitive result observed in a majority of patients (increase of lesion volumes) is not associated to a poor prognosis. Moreover this suggests that the evolution of the aspergillosis lesions should be followed by sequential analysis of CT scans and not only by comparison of each CT scan with baseline images. Among the 19/26 patients with a lesion size increase between D0 and D7, a favorable response was observed for 42% of them. This confirms our previous study that a size increase of pulmonary aspergillosis lesions is not systematically pejorative and extents these observations to probable IPA. Therefore, in patients recovering from neutropenia with a concomitant “deterioration” of aspergillosis lesions, this notion should be taken into account before any decision for antifungal treatment change. The reasons for the initial volume increase of aspergillosis lesions remain unclear. It could be hypothesized that the lesion volume increase is related to defense mechanisms with infiltration of immune cells. Therefore these signs of apparent pulmonary deterioration should be considered as markers of bone marrow recovery. Indeed, both the crescent sign and the cavitations are the result of leukocyte recovery with release of proteases at the site of lesions [9]. Moreover, we have observed as Brodoefel et al. that cavitations were strong precursors of beneficial outcome [8,10]. In our previous study we also observed that recovery of peripheral neutrophils was accompanied with appearance of air-crescent signs and lesion volume increase [4]. Miceli et al. proposed also recently that the association of a pulmonary worsening with neutrophil recovery was caused by the restored ability to mount an inflammatory response [11]. This mechanism could be similar to the immune reconstitution inflammation syndrome (IRIS) described in HIV patients who respond to antiretroviral treatment. Miceli et al. proposed recently a galactomannan-based strategy to distinguish between progressive aspergillosis and responding IA with IRIS [11]. Other factors possibly predictive of favorable overall response were analyzed. Thus, in the study of Greene et al., a significant higher proportion of patients with a halo sign evidenced by the baseline CT scan responded to subsequent treatment as compared with patients with other CT imaging findings (52 vs. 29%) [12]. In the present study, the presence of the halo sign was not predictive of favorable overall response at EOT (42% of favorable outcome with halo vs. 54% in the absence of halo). A positive galactomannan assay at baseline did not appear to be predictive of favorable overall response. The results of this prospective study suggest that the sequential analysis of CT scan in neutropenic patients with IPA depicts more precisely the evolution of lesion volumes than comparison to baseline images. Moreover, the systematic use of chest CT appears to be a useful tool for diagnosis and outcome evaluation of IPA in clinical trials.
D. Caillot et al. / European Journal of Radiology 74 (2010) e172–e175
e175
Fig. 2. Evolution of thoracic CT scan images of a patient with invasive pulmonary aspergillosis. The images A through D are successive thoracic CT scan images of a patient with diagnosis of probable IPA. Diagnosis of IPA was based on positive galactomannan and CT images (bronchoalveolar lavage was negative for direct examination and culture while positive for galactomannan detection). There was an increase of the total lesion volume between D0 (Image A) and D7 (Image B) and then a decrease at D14 (Image C). Typical halo sign and crescent air sign were visible. Sequelae lesions were observed at D42 (Image D).
Conflict of interest D. Caillot has been a consultant for Schering; R. Herbrecht has been a consultant to Pfizer, Schering-Plough, Gilead Sciences, MSD, Astellas and has received research support from Pfizer; S. de Botton has been a consultant to Schering-Plough; S. Alfandari received honoraria from Amgen and Gilead Sciences; L. Mahi and F. Monchecourt are employees of Gilead Sciences. All other authors: no conflicts. Acknowledgements We thank Franck Sévenier, Isabelle Pascal and Aurélie Moyne (Fovéa, Paris) for data analysis, Francis Beauvais, Aymeric Duvivier and Bénédicte Bonnet (Gilead Sciences, France) for their support and assistance. References [1] Hauggaard A, Ellis M, Ekelund L. Early chest radiography and CT in the diagnosis, management and outcome of invasive pulmonary aspergillosis. Acta Radiol 2002;43:292–8. [2] Greene R. The radiological spectrum of pulmonary aspergillosis. Med Mycol 2005;43(Suppl. 1):S147–54. [3] Kuhlman JE, Fishman EK, Siegelman SS. Invasive pulmonary aspergillosis in acute leukemia: characteristic findings on CT, the CT halo sign, and the role of CT in early diagnosis. Radiology 1985;157:611–4.
[4] Caillot D, Couaillier JF, Bernard A, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic computed tomography scans in patients with neutropenia. J Clin Oncol 2001;19:253– 9. [5] Ascioglu S, Rex JH, de Pauw B, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis 2002;34:7– 14. [6] Caillot D, Thiébaut A, Herbrecht R, et al. Liposomal amphotericin B in combination with caspofungin for invasive aspergillosis in patients with hematological malignancies: a randomized pilot study (Combistrat Trial). Cancer 2007;110:2740–6. [7] De Pauw B, Walsh TJ, Donnelly JP, et al. Revised Definitions of Invasive Fungal Disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. Clin Infect Dis 2008;46:1813–21. [8] Brodoefel H, Vogel M, Hebart H, et al. Long-term CT follow-up in 40 non-HIV immunocompromised patients with invasive pulmonary aspergillosis: kinetics of CT morphology and correlation with clinical findings and outcome. AJR Am J Roentgenol 2006;187:404–13. [9] Abramson S. The air crescent sign. Radiology 2001;218:230–2. [10] Caillot D, Casasnovas O, Bernard A, et al. Improved management of invasive pulmonary aspergillosis in neutropenic patients using early thoracic computed tomographic scan and surgery. J Clin Oncol 1997;15:139– 47. [11] Miceli MH, Maertens J, Buve K, et al. Immune reconstitution inflammatory syndrome in cancer patients with pulmonary aspergillosis recovering from neutropenia: proof of principle, description, and clinical and research implications. Cancer 2007;110:112–20. [12] Greene RE, Schlamm HT, Oestmann JW, et al. Imaging findings in acute invasive pulmonary aspergillosis: clinical significance of the halo sign. Clin Infect Dis 2007;44:373–9.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Computer tomography in pulmonary invasive aspergillosis in hematological patients with neutropenia: An useful tool for diagnosis and assessment of outcome in clinical trials Denis Caillot a,∗ , Valérie Latrabe b , Anne Thiébaut c , Raoul Herbrecht d , Stéphane De Botton e , Arnaud Pigneux b , Franc¸oise Monchecourt f , Lamine Mahi f , Serge Alfandari e , Jean-Franc¸ois Couaillier a a
Dijon University Hospital Center, Dijon, France Bordeaux University Hospital Center, Bordeaux, France c Lyon University Hospital Center, Lyon, France d Strasbourg Regional University Hospital Center, Strasbourg, France e Lille Regional University Hospital Center, Lille, France f Gilead Sciences, Paris, France b
a r t i c l e
i n f o
Article history: Received 22 August 2008 Received in revised form 9 April 2009 Accepted 25 May 2009 Keywords: Thoracic CT scan Pulmonary invasive aspergillosis Neutropenia Acute myeloid leukaemia
a b s t r a c t Background and objective: The exact timing of the evolution of lesion volumes of invasive pulmonary aspergillosis (IPA) on CT scan images could be helpful in the management of hematological patients but has never been evaluated in a prospective study. We analyzed the CT scan data from the prospective Combistrat trial. Design and methods: Volumes of aspergillosis lesions from 30 patients (including 24 acute myeloid leukaemia) with probable (n = 26) or proven (n = 4) IPA according to the EORTC-MSG modified criteria, were measured prospectively on the thoracic CT scans at the enrolment in the study on day 0 (D0), D7, D14 and end of treatment (EOT). Results: For the overall population, the volume of pulmonary aspergillosis lesions increased significantly from D0 to D7 (1.6 fold; p = 0.003). Then this volume decreased significantly from D7 to D14 (1.36 fold at D14 with p = 0.003 for D14 vs. D7, but with p = 0.56 for D14 vs. D0). At EOT (= D17, median value), the volume of lesions was significantly lower than D14 (0.76 fold the initial volume; p < 0.001) but it was not significantly different when compared to D0 (p = 0.11). Conclusions: The results of this prospective study suggest that the sequential analysis of CT scan in neutropenic patients with IPA depicts more precisely the evolution of lesion volumes than comparison to baseline images. Moreover, the systematic use of chest CT appears to be a useful tool for diagnosis and outcome evaluation of IPA in clinical trials. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Invasive aspergillosis (IA) in neutropenic patients remains lifethreatening. Therefore reliable and timely diagnosis is an important objective for rapid treatment initiation because IA jeopardizes the planning of chemotherapy or hematopoietic stem cell transplantation. Thoracic CT scan may detect pulmonary aspergillosis at an early stage of infection [1,2]. Thus, the halo sign (a mass-like infil-
∗ Corresponding author at: Service d’Hématologie, C.H.U. de Dijon, Hôpital du Bocage, Hôpital d’Enfants, 10, boulevard du Maréchal de Lattre de Tassigny, B.P. 1542, 21034 Dijon, France. Tel.: +33 3 80 29 50 41. E-mail address: [email protected] (D. Caillot). 0720-048X/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2009.05.058
trate surrounded by ground-glass opacity) is considered as an early sign of invasive pulmonary aspergillosis (IPA) [3]. Until now the evolution of CT scan images has been poorly studied although a better knowledge of anatomical evolution of lesions could help decisions for treatment continuation or switching. In a previous paper we studied the timing of CT images in hematological patients with neutropenia [4]. We observed that, despite early initiation of antifungal treatment, the size of the lesions of IPA increased during the first days of the disease (between 0 and day 7). We concluded that an increase of the size of aspergillosis lesion during the first days after diagnosis was not associated with a pejorative immediate outcome. This apparent failure of the antifungal treatment could lead clinicians to initiate other antifungal treatments supposed to be more efficient. However, this study selected
D. Caillot et al. / European Journal of Radiology 74 (2010) e172–e175
retrospectively patients with surgically proven invasive aspergillosis (according to the EORTC/MSG criteria) [5]. So, surgery could have subsequently modified the outcome of these patients. Therefore, it was of interest to study in a prospective study whether these observations could be extended to patients with probable IPA most often treated without surgery. We had the opportunity to answer this issue with the Combistrat study that included a majority of patients with probable IPA [6]. The aim of the Combistrat study was to evaluate prospectively a combination of liposomal amphotericin B and caspofungin in comparison with liposomal amphotericin B alone. In this article, we present the analysis of the sequential images of pulmonary lesions performed prospectively during the Combistrat trial. 2. Patients and methods 2.1. Study population The present analysis was based upon thoracic CT scan data gathered from the Combistrat study that has been published elsewhere [6]. Briefly, the Combistrat study was a national, multicenter, pilot, prospective, randomized open trial in patients with proven or probable IA according to criteria of the European Organization for Research and Treatment of Cancer/Mycosis Study Group (EORTC/MSG)[5]. 2.2. CT scan image database The database of digitized images used for this analysis comprised the thoracic CT scans from patients enrolled in the Combistrat study. Thoracic CT scans analyzed in the present study were performed for diagnosis, outcome assessment and follow-up of patients. The Combistrat study included hematological patients with neutropenia with probable or proven IA according to the criteria of the EORTC/MSG. Proven IA diagnosis required identification of typical hyphal elements after histopathological or cytopathological examination with evidence of associated tissue damage (either microscopically or unequivocally by imaging) or growth of Aspergillus organisms from a sample obtained by sterile procedure from a normally sterile site and clinically or radiologically abnormal site consistent with infection (excluding urine and mucous membrane). Probable IA diagnosis required at least one host factor criterion and one microbiological criterion and one major (or two minor) clinical criteria from abnormal site consistent with infection and no other pathogen detected to account for the clinical or radiographic signs of infection. As a modification of EORTC/MSG criteria, diagnosis of probable IA included also patients with recent neutropenia (<500 neutrophils/mm3 ) within 14 days of study inclusion and “halo” or “air crescent” signs on thoracic CT scan [5]. Note that this modification of criteria for probable IA has been recently included in the Revised Definitions of Invasive Fungal Disease of EORTC/MSG [7]. In addition, serum galactomannan assay (PlateliaTM , Bio-Rad Laboratories, Hercules, CA and Bio-Rad, Marnes-la-Coquette, France) with an optical density index of ≥1.0 was considered positive. To ensure uniformity of assessments, the CT scan diagnosis of IA was verified by a centrally data review board including one hematologist and one radiologist who confirmed IA. CT scans, when initially positive, were repeated at days 7, 14, 28 and then once a month, if applicable, for radiographic response assessments. The measurement of lesion volumes was also centralized. As it was hypothesized that aspergillary lesion had an ovoid shape, the calculated volume (in cm3 ) of each aspergillosis
e173
pulmonary lesion was obtained using the following formula: height × length × wide × /6). For several lesions, the total volume was obtained by addition of all elementary volumes. 2.3. Assessment of efficacy Complete (cure) and partial (improvement) overall responses were considered as favorable responses. Stable overall response and failure were considered as unfavorable responses. Complete response was defined as complete resolution of all signs and symptoms attributable to the invasive fungal infection as compared to baseline and complete or near complete clearing of CT scan abnormalities associated with active fungal infection (or persistent residual scarring only) and eradication of the pathogen. Partial response was defined as meaningful improvement of all clinical signs and symptoms attributable to IA and a decrease of at least 50% of radiographic signs. Stable response was defined as minor or no improvement but no worsening of attributable signs, symptoms, radiographic and/or bronchoscopic findings. Failure was defined as progression of infection, based on an increase in the number and/or severity of clinical signs and symptoms attributable to the IA, worsening of CT scan abnormalities consistent with progressive infection and persistently positive cultures, histopathological findings or positive galactomannan assays. 2.4. Statistical analysis The changes of lesion volumes were compared using Wilcoxon’s test. A p-value below 0.05 was considered as statistically significant. 3. Results 3.1. Patients of the Combistrat trial Baseline characteristics and clinical results of the Combistrat study have been described elsewhere [6]. Briefly, among the 32 patients enrolled in the Combistrat trial, 30 met the diagnosis of proven (n = 4) or probable (n = 26) IA according to a centrally data review board. At baseline, all the patients were neutropenic and had clinical signs. The initial thoracic CT scans were abnormal for all patients. In addition, presence of CT halo sign, positive galactomannan assay (two successive positive samples with an optical density index ≥1), culture or direct examination of bronchoalveolar lavage contributed to diagnosis for 19, 12 and 5 patients, respectively. The majority of patients were male (21/30; 70%) and the age ranged from 16 to 75 y (median, 58.5 y). Twenty-four patients had acute myeloblastic leukemia; other underlying conditions were chronic lymphocytic leukemia (n = 3), myeloproliferative disorders (n = 2) and acute lymphoblastic leukemia (n = 1). Median duration of neutropenia at inclusion was 23 days. Overall, at EOT (primary efficacy endpoint), 14 of 30 patients had a favorable partial or complete response. At EOT, 29/30 patients were alive and at week 12, all the patients of combination group were alive while 3 of 15 died in the high-dose group. 3.2. Evolution of the volumes of lesions in the overall population The volumes of the thoracic aspergillosis lesions were calculated for each patient. As evidenced in Fig. 1, for overall population, the volume increased significantly from D0 to D7 (1.6 fold; p = 0.003). After D7, the volume decreased and its variation between D7 and D14 was also statistically significant (1.6 vs. 1.36 fold the initial volume, p = 0.003). Moreover, at EOT (= D17, median value), the volume of lesions was significantly lower than at D14 (0.76 fold the
e174
D. Caillot et al. / European Journal of Radiology 74 (2010) e172–e175
4. Discussion
Fig. 1. Evolution of volume of lesions of pulmonary aspergillosis. The volumes of the thoracic aspergillosis lesions increased significantly from D0 to D7 (1.6 fold). After D7 the volume decreased and the volume variation between D7 and D14 was also statistically significant (1.36 fold the initial volume). Median of end of treatment (EOT) was 17 days after D0 and at EOT the volume was lower than the D14 volume (0.76 the initial volume).
initial volume; p < 0.001). At EOT and D14, the volumes were not significantly different when compared to D0 (p = 0.11 for EOT and p = 0.56 for D14). In patients with favorable outcome at EOT, the total volume of lesions between D0 and D14 was significantly higher (90, 60 and 44 cm3 at D0, D7, D14; median values) as compared to patients with unfavorable outcome (14, 38 and 33 cm3 , respectively; Anova test, p = 0.015). In contrast, as expected, the lesion volumes appeared to be smaller at EOT in patients with favorable outcome in comparison in patients with unfavorable outcome (29 vs. 58 cm3 , respectively). Fig. 2 presents successive thoracic CT scans of a patient with probable invasive pulmonary aspergillosis (IPA).
3.3. Prognostic factors for favorable overall response An increase of the total volume of lesions between D0 and D7 was observed for 19/26 patients (73%). Favorable overall response at EOT was observed for 7/19 (42%) patients with increase of volume and for 5/7 (71%) patients without increase of volume. A positive galactomannan assay (two successive positive samples with an optical density index ≥1) was present at baseline for 12/30 (40%) patients. Favorable overall response at EOT was observed for 5/12 (50%) patients with positive galactomannan assay and 7/18 (39%) patients with negative galactomannan assay at baseline. The halo sign was present at D0 for 19/30 patients (63%). Favorable overall response at EOT was observed for 8/19 (42%) patients with halo at baseline and for 6/11 (54%) patients without halo at baseline.
In a previous retrospective study in neutropenic patients with proven IPA, we showed that the volume of aspergillosis lesions evaluated by sequential thoracic CT scans increased during the first days after antifungal treatment and was not correlated with pejorative outcome [4]. An increase of the size of pulmonary aspergillosis lesions during the first days of treatment was also reported in the recent study of Brodoefel et al. in 40 non-HIV immunocompromised patients with confirmed pulmonary IA [8]. There were 26/30 patients in the Combistrat study with probable IPA. Therefore the present data allowed us to study whether our previous observation of an initial increase of lesion size could be extended to probable IPA in a prospective trial. We observed indeed that 19/26 patients had an increase of lesion volume between D0 and D7. Overall, the increase of median lesion size from D0 to D7 was statistically significant and was followed by a significant decrease from D7 to D14 and then from D14 to EOT. This indicates that the counterintuitive result observed in a majority of patients (increase of lesion volumes) is not associated to a poor prognosis. Moreover this suggests that the evolution of the aspergillosis lesions should be followed by sequential analysis of CT scans and not only by comparison of each CT scan with baseline images. Among the 19/26 patients with a lesion size increase between D0 and D7, a favorable response was observed for 42% of them. This confirms our previous study that a size increase of pulmonary aspergillosis lesions is not systematically pejorative and extents these observations to probable IPA. Therefore, in patients recovering from neutropenia with a concomitant “deterioration” of aspergillosis lesions, this notion should be taken into account before any decision for antifungal treatment change. The reasons for the initial volume increase of aspergillosis lesions remain unclear. It could be hypothesized that the lesion volume increase is related to defense mechanisms with infiltration of immune cells. Therefore these signs of apparent pulmonary deterioration should be considered as markers of bone marrow recovery. Indeed, both the crescent sign and the cavitations are the result of leukocyte recovery with release of proteases at the site of lesions [9]. Moreover, we have observed as Brodoefel et al. that cavitations were strong precursors of beneficial outcome [8,10]. In our previous study we also observed that recovery of peripheral neutrophils was accompanied with appearance of air-crescent signs and lesion volume increase [4]. Miceli et al. proposed also recently that the association of a pulmonary worsening with neutrophil recovery was caused by the restored ability to mount an inflammatory response [11]. This mechanism could be similar to the immune reconstitution inflammation syndrome (IRIS) described in HIV patients who respond to antiretroviral treatment. Miceli et al. proposed recently a galactomannan-based strategy to distinguish between progressive aspergillosis and responding IA with IRIS [11]. Other factors possibly predictive of favorable overall response were analyzed. Thus, in the study of Greene et al., a significant higher proportion of patients with a halo sign evidenced by the baseline CT scan responded to subsequent treatment as compared with patients with other CT imaging findings (52 vs. 29%) [12]. In the present study, the presence of the halo sign was not predictive of favorable overall response at EOT (42% of favorable outcome with halo vs. 54% in the absence of halo). A positive galactomannan assay at baseline did not appear to be predictive of favorable overall response. The results of this prospective study suggest that the sequential analysis of CT scan in neutropenic patients with IPA depicts more precisely the evolution of lesion volumes than comparison to baseline images. Moreover, the systematic use of chest CT appears to be a useful tool for diagnosis and outcome evaluation of IPA in clinical trials.
D. Caillot et al. / European Journal of Radiology 74 (2010) e172–e175
e175
Fig. 2. Evolution of thoracic CT scan images of a patient with invasive pulmonary aspergillosis. The images A through D are successive thoracic CT scan images of a patient with diagnosis of probable IPA. Diagnosis of IPA was based on positive galactomannan and CT images (bronchoalveolar lavage was negative for direct examination and culture while positive for galactomannan detection). There was an increase of the total lesion volume between D0 (Image A) and D7 (Image B) and then a decrease at D14 (Image C). Typical halo sign and crescent air sign were visible. Sequelae lesions were observed at D42 (Image D).
Conflict of interest D. Caillot has been a consultant for Schering; R. Herbrecht has been a consultant to Pfizer, Schering-Plough, Gilead Sciences, MSD, Astellas and has received research support from Pfizer; S. de Botton has been a consultant to Schering-Plough; S. Alfandari received honoraria from Amgen and Gilead Sciences; L. Mahi and F. Monchecourt are employees of Gilead Sciences. All other authors: no conflicts. Acknowledgements We thank Franck Sévenier, Isabelle Pascal and Aurélie Moyne (Fovéa, Paris) for data analysis, Francis Beauvais, Aymeric Duvivier and Bénédicte Bonnet (Gilead Sciences, France) for their support and assistance. References [1] Hauggaard A, Ellis M, Ekelund L. Early chest radiography and CT in the diagnosis, management and outcome of invasive pulmonary aspergillosis. Acta Radiol 2002;43:292–8. [2] Greene R. The radiological spectrum of pulmonary aspergillosis. Med Mycol 2005;43(Suppl. 1):S147–54. [3] Kuhlman JE, Fishman EK, Siegelman SS. Invasive pulmonary aspergillosis in acute leukemia: characteristic findings on CT, the CT halo sign, and the role of CT in early diagnosis. Radiology 1985;157:611–4.
[4] Caillot D, Couaillier JF, Bernard A, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic computed tomography scans in patients with neutropenia. J Clin Oncol 2001;19:253– 9. [5] Ascioglu S, Rex JH, de Pauw B, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis 2002;34:7– 14. [6] Caillot D, Thiébaut A, Herbrecht R, et al. Liposomal amphotericin B in combination with caspofungin for invasive aspergillosis in patients with hematological malignancies: a randomized pilot study (Combistrat Trial). Cancer 2007;110:2740–6. [7] De Pauw B, Walsh TJ, Donnelly JP, et al. Revised Definitions of Invasive Fungal Disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. Clin Infect Dis 2008;46:1813–21. [8] Brodoefel H, Vogel M, Hebart H, et al. Long-term CT follow-up in 40 non-HIV immunocompromised patients with invasive pulmonary aspergillosis: kinetics of CT morphology and correlation with clinical findings and outcome. AJR Am J Roentgenol 2006;187:404–13. [9] Abramson S. The air crescent sign. Radiology 2001;218:230–2. [10] Caillot D, Casasnovas O, Bernard A, et al. Improved management of invasive pulmonary aspergillosis in neutropenic patients using early thoracic computed tomographic scan and surgery. J Clin Oncol 1997;15:139– 47. [11] Miceli MH, Maertens J, Buve K, et al. Immune reconstitution inflammatory syndrome in cancer patients with pulmonary aspergillosis recovering from neutropenia: proof of principle, description, and clinical and research implications. Cancer 2007;110:112–20. [12] Greene RE, Schlamm HT, Oestmann JW, et al. Imaging findings in acute invasive pulmonary aspergillosis: clinical significance of the halo sign. Clin Infect Dis 2007;44:373–9.