PROTEOMIC PROFILE CORRELATES TO MOLECULAR RESPONSE IN PATIENTS WITH CHRONIC MYELOID LEUKEMIA
(Abstract release date: 05/19/16)
EHA Library. Sgherza N. 06/09/16; 134712; PB1812
Dr. Nicola Sgherza
Contributions
Contributions
Abstract
Abstract: PB1812
Type: Publication Only
Background
Recent advances in the proteomic field have allowed us to better understand the biology of several cancer types and/or discover new candidate biomarkers. Very few data are available about the serum protein expression of patients with Chronic Myeloid Leukemia (CML).
Aims
The aim of our pilot study was to evaluate a correlation between depth of molecular response (MR) and proteomic profile of CML patients, and if possible to find novel potential biomarkers complementary to currently existing proven tools to monitor therapy response.
Methods
Thirty sera from peripheral blood (PB) were sampled from 8 patients in MR1 response, 6 in MR2, 11 in MR3, and 5 in MR4. For 11 patients serum from bone marrow (BM) was also available. In particular 3 were sampled from patients in MR1, 3 in MR2, 4 in MR3, 1 in MR4. The one patient at diagnosis was included in MR1 group. Samples were divided into 2 groups according to the achievement of MR3: Group A, which comprised 14 samples from patients with a MR lower than MR3, and Group B, including 16 samples from patients who experienced a MR greater than or equal to MR3.The association of proteomic profile with MR was investigated using the SELDI ToF Mass Spectrometry platform. Each specimen was analyzed in duplicate. Expression Differences Mapping analysis was applied in order to generate a cluster peaks list, which describes how a singular peak is expressed in the specimen spectrum. Finally, with the aim to give a preliminary identity to the most differentially expressed peaks, an in silico identification was attempted using the Mascot database search available at www.matrixscience.com.
Results
Comparing PB sera from group A and group B, only a peak at 5075 Da had a p-value=0.05. Comparing sera from patients with different MR (MR1 vs MR2 vs MR3 vs MR4), statistical analysis found two features at 11092 Da and once again that at 5075 Da as differentially expressed and statistically significant (P-value=0.0034 and 0.0084 respectively) between MR1 and MR4. In particular the peptide at 11092 Da was overexpressed in MR4 patients. Also the peak at 5075 Da was highlighted as statistically significant, but its significance was less important than that of 11092 Da due to a slightly worse sensitivity (80%). Sera from PB were compared with those extracted from BM. Interestingly, no difference was highlighted. Regarding the two peaks (5075 Da and 11072 Da) we hypothesized that they were closely related, probably as a fragmentation product. With the aim to quickly give a preliminary identity to the peptide, we attempted an “in silico” mass peptide finger print using the open-source tool available at www.matrixscience.com. The bioinformatic tool identified the peptide with a high probability score, as a truncated part of a larger nuclear protein ZMIZ2. It would interact with the Wnt/β-catenin pathway participating in its activation. As suggested by data from literature, we can try to hypothesize that, unlike non responders, in the serum of responsive patients a part of ZMIZ2 is present, as a consequence of the degradation of the entire protein; this finding results in an under-regulation of the Wnt/β-catenin pathway with consequently a better response.
Conclusion
These preliminary data suggest that the peptide at 11092 Da is very closely related to a good response. Our efforts are now oriented to definitively confirming the identity of the peptide and to clarify its role.
Session topic: E-poster
Type: Publication Only
Background
Recent advances in the proteomic field have allowed us to better understand the biology of several cancer types and/or discover new candidate biomarkers. Very few data are available about the serum protein expression of patients with Chronic Myeloid Leukemia (CML).
Aims
The aim of our pilot study was to evaluate a correlation between depth of molecular response (MR) and proteomic profile of CML patients, and if possible to find novel potential biomarkers complementary to currently existing proven tools to monitor therapy response.
Methods
Thirty sera from peripheral blood (PB) were sampled from 8 patients in MR1 response, 6 in MR2, 11 in MR3, and 5 in MR4. For 11 patients serum from bone marrow (BM) was also available. In particular 3 were sampled from patients in MR1, 3 in MR2, 4 in MR3, 1 in MR4. The one patient at diagnosis was included in MR1 group. Samples were divided into 2 groups according to the achievement of MR3: Group A, which comprised 14 samples from patients with a MR lower than MR3, and Group B, including 16 samples from patients who experienced a MR greater than or equal to MR3.The association of proteomic profile with MR was investigated using the SELDI ToF Mass Spectrometry platform. Each specimen was analyzed in duplicate. Expression Differences Mapping analysis was applied in order to generate a cluster peaks list, which describes how a singular peak is expressed in the specimen spectrum. Finally, with the aim to give a preliminary identity to the most differentially expressed peaks, an in silico identification was attempted using the Mascot database search available at www.matrixscience.com.
Results
Comparing PB sera from group A and group B, only a peak at 5075 Da had a p-value=0.05. Comparing sera from patients with different MR (MR1 vs MR2 vs MR3 vs MR4), statistical analysis found two features at 11092 Da and once again that at 5075 Da as differentially expressed and statistically significant (P-value=0.0034 and 0.0084 respectively) between MR1 and MR4. In particular the peptide at 11092 Da was overexpressed in MR4 patients. Also the peak at 5075 Da was highlighted as statistically significant, but its significance was less important than that of 11092 Da due to a slightly worse sensitivity (80%). Sera from PB were compared with those extracted from BM. Interestingly, no difference was highlighted. Regarding the two peaks (5075 Da and 11072 Da) we hypothesized that they were closely related, probably as a fragmentation product. With the aim to quickly give a preliminary identity to the peptide, we attempted an “in silico” mass peptide finger print using the open-source tool available at www.matrixscience.com. The bioinformatic tool identified the peptide with a high probability score, as a truncated part of a larger nuclear protein ZMIZ2. It would interact with the Wnt/β-catenin pathway participating in its activation. As suggested by data from literature, we can try to hypothesize that, unlike non responders, in the serum of responsive patients a part of ZMIZ2 is present, as a consequence of the degradation of the entire protein; this finding results in an under-regulation of the Wnt/β-catenin pathway with consequently a better response.
Conclusion
These preliminary data suggest that the peptide at 11092 Da is very closely related to a good response. Our efforts are now oriented to definitively confirming the identity of the peptide and to clarify its role.
Session topic: E-poster
Abstract: PB1812
Type: Publication Only
Background
Recent advances in the proteomic field have allowed us to better understand the biology of several cancer types and/or discover new candidate biomarkers. Very few data are available about the serum protein expression of patients with Chronic Myeloid Leukemia (CML).
Aims
The aim of our pilot study was to evaluate a correlation between depth of molecular response (MR) and proteomic profile of CML patients, and if possible to find novel potential biomarkers complementary to currently existing proven tools to monitor therapy response.
Methods
Thirty sera from peripheral blood (PB) were sampled from 8 patients in MR1 response, 6 in MR2, 11 in MR3, and 5 in MR4. For 11 patients serum from bone marrow (BM) was also available. In particular 3 were sampled from patients in MR1, 3 in MR2, 4 in MR3, 1 in MR4. The one patient at diagnosis was included in MR1 group. Samples were divided into 2 groups according to the achievement of MR3: Group A, which comprised 14 samples from patients with a MR lower than MR3, and Group B, including 16 samples from patients who experienced a MR greater than or equal to MR3.The association of proteomic profile with MR was investigated using the SELDI ToF Mass Spectrometry platform. Each specimen was analyzed in duplicate. Expression Differences Mapping analysis was applied in order to generate a cluster peaks list, which describes how a singular peak is expressed in the specimen spectrum. Finally, with the aim to give a preliminary identity to the most differentially expressed peaks, an in silico identification was attempted using the Mascot database search available at www.matrixscience.com.
Results
Comparing PB sera from group A and group B, only a peak at 5075 Da had a p-value=0.05. Comparing sera from patients with different MR (MR1 vs MR2 vs MR3 vs MR4), statistical analysis found two features at 11092 Da and once again that at 5075 Da as differentially expressed and statistically significant (P-value=0.0034 and 0.0084 respectively) between MR1 and MR4. In particular the peptide at 11092 Da was overexpressed in MR4 patients. Also the peak at 5075 Da was highlighted as statistically significant, but its significance was less important than that of 11092 Da due to a slightly worse sensitivity (80%). Sera from PB were compared with those extracted from BM. Interestingly, no difference was highlighted. Regarding the two peaks (5075 Da and 11072 Da) we hypothesized that they were closely related, probably as a fragmentation product. With the aim to quickly give a preliminary identity to the peptide, we attempted an “in silico” mass peptide finger print using the open-source tool available at www.matrixscience.com. The bioinformatic tool identified the peptide with a high probability score, as a truncated part of a larger nuclear protein ZMIZ2. It would interact with the Wnt/β-catenin pathway participating in its activation. As suggested by data from literature, we can try to hypothesize that, unlike non responders, in the serum of responsive patients a part of ZMIZ2 is present, as a consequence of the degradation of the entire protein; this finding results in an under-regulation of the Wnt/β-catenin pathway with consequently a better response.
Conclusion
These preliminary data suggest that the peptide at 11092 Da is very closely related to a good response. Our efforts are now oriented to definitively confirming the identity of the peptide and to clarify its role.
Session topic: E-poster
Type: Publication Only
Background
Recent advances in the proteomic field have allowed us to better understand the biology of several cancer types and/or discover new candidate biomarkers. Very few data are available about the serum protein expression of patients with Chronic Myeloid Leukemia (CML).
Aims
The aim of our pilot study was to evaluate a correlation between depth of molecular response (MR) and proteomic profile of CML patients, and if possible to find novel potential biomarkers complementary to currently existing proven tools to monitor therapy response.
Methods
Thirty sera from peripheral blood (PB) were sampled from 8 patients in MR1 response, 6 in MR2, 11 in MR3, and 5 in MR4. For 11 patients serum from bone marrow (BM) was also available. In particular 3 were sampled from patients in MR1, 3 in MR2, 4 in MR3, 1 in MR4. The one patient at diagnosis was included in MR1 group. Samples were divided into 2 groups according to the achievement of MR3: Group A, which comprised 14 samples from patients with a MR lower than MR3, and Group B, including 16 samples from patients who experienced a MR greater than or equal to MR3.The association of proteomic profile with MR was investigated using the SELDI ToF Mass Spectrometry platform. Each specimen was analyzed in duplicate. Expression Differences Mapping analysis was applied in order to generate a cluster peaks list, which describes how a singular peak is expressed in the specimen spectrum. Finally, with the aim to give a preliminary identity to the most differentially expressed peaks, an in silico identification was attempted using the Mascot database search available at www.matrixscience.com.
Results
Comparing PB sera from group A and group B, only a peak at 5075 Da had a p-value=0.05. Comparing sera from patients with different MR (MR1 vs MR2 vs MR3 vs MR4), statistical analysis found two features at 11092 Da and once again that at 5075 Da as differentially expressed and statistically significant (P-value=0.0034 and 0.0084 respectively) between MR1 and MR4. In particular the peptide at 11092 Da was overexpressed in MR4 patients. Also the peak at 5075 Da was highlighted as statistically significant, but its significance was less important than that of 11092 Da due to a slightly worse sensitivity (80%). Sera from PB were compared with those extracted from BM. Interestingly, no difference was highlighted. Regarding the two peaks (5075 Da and 11072 Da) we hypothesized that they were closely related, probably as a fragmentation product. With the aim to quickly give a preliminary identity to the peptide, we attempted an “in silico” mass peptide finger print using the open-source tool available at www.matrixscience.com. The bioinformatic tool identified the peptide with a high probability score, as a truncated part of a larger nuclear protein ZMIZ2. It would interact with the Wnt/β-catenin pathway participating in its activation. As suggested by data from literature, we can try to hypothesize that, unlike non responders, in the serum of responsive patients a part of ZMIZ2 is present, as a consequence of the degradation of the entire protein; this finding results in an under-regulation of the Wnt/β-catenin pathway with consequently a better response.
Conclusion
These preliminary data suggest that the peptide at 11092 Da is very closely related to a good response. Our efforts are now oriented to definitively confirming the identity of the peptide and to clarify its role.
Session topic: E-poster
{{ help_message }}
{{filter}}