Savaraj N

Background: Melanomas harboring BRAF mutation (V600E) is known to recur following treatment with BRAF inhibitors (BRAFi) despite a high initial response rate. While the addition of MEK inhibitor can increase both response rate and duration of response. Relapsed is still inevitable. We have found that both BRAFi-resistant melanoma (BR) and BRAFi/ MEKi resistant (BMR) cells are vulnerable to arginine depletion. This is due to the fact that they cannot readily undergo autophagy upon arginine deprivation and unable to turn on argininosuccinate synthetase 1 (ASS1) to synthesize arginine. In this report, we have studied the underlying mechanisms.

Methods: 4 pairs of melanoma cell lines (parental and its BRAFi/MEKi resistant variants): A375/A375BMR, Skmel28/Skmel28BMR, A2058/A2058BMR, and MEL1220/Mel120BMR) were used for the study. MEL1220 was established in our laboratory. BMR resistant cell line was generated by continuous exposure of parental to BRAF inhibitor (Vemurafenib) and MEK inhibitor (Trametinib).

Results/Discussion: We demonstrated that downregulation of AMPK-1 is a major factor contributing to impairment of autophagy. In addition, decreased AMPK-1 expression also results in metabolic reprogramming in BMR cells toward more dependency on arginine and less on glycolysis evidenced by upregulated arginine transporter CAT-2 and downregulated glucose transporter GLUT1. Notably, when naïve melanoma cells become BMR cells by long-term exposure to BRAFi/MEKi, a stepwise degradation of AMPK-1 is initiated via ubiquitin-proteasome system (UPS). Using proteomic analysis, we discovered that a novel E3 ligase, RING finger 44 (RNF44), can interact with AMPK-1 and promote AMPK-1 degradation in both BR (BRAF resistance) and BMR cells. Elevated RNF44 and downregulated AMPK-1 expressions are also found in both BR BMR cells. Elevated RNF44 and downregulated AMPK-1 expressions are also found in both BR and BMR and tumor samples from BR and BMR patients.

Conclusions: Our results suggest that BRAFi resistance-driven proteasomal degradation of AMPK-1 is crucial to govern vulnerability to arginine deprivation in melanoma cells; therefore, depleting exogenous arginine can be considered as a novel salvage treatment for melanoma patients who fail BRAF/MEK inhibitors. Furthermore arginine deprivation also results in an increase PD-L1 expression which make them more sensitive to check point inhibitors (supported by VA Merit Review Award (1BX003328) to Savaraj and 1R01CA152197 to Kuo, Feun and Savaraj.

Background: Melanomas harboring BRAF mutation (V600E) is known to recur following treatment with BRAF inhibitors (BRAFi) despite a high initial response rate. While the addition of MEK inhibitor can increase both response rate and duration of response. Relapsed is still inevitable. We have found that both BRAFi-resistant melanoma (BR) and BRAFi/ MEKi resistant (BMR) cells are vulnerable to arginine depletion. This is due to the fact that they cannot readily undergo autophagy upon arginine deprivation and unable to turn on argininosuccinate synthetase 1 (ASS1) to synthesize arginine. In this report, we have studied the underlying mechanisms.

Methods: 4 pairs of melanoma cell lines (parental and its BRAFi/MEKi resistant variants): A375/A375BMR, Skmel28/Skmel28BMR, A2058/A2058BMR, and MEL1220/Mel120BMR) were used for the study. MEL1220 was established in our laboratory. BMR resistant cell line was generated by continuous exposure of parental to BRAF inhibitor (Vemurafenib) and MEK inhibitor (Trametinib).

Results/Discussion: We demonstrated that downregulation of AMPK-1 is a major factor contributing to impairment of autophagy. In addition, decreased AMPK-1 expression also results in metabolic reprogramming in BMR cells toward more dependency on arginine and less on glycolysis evidenced by upregulated arginine transporter CAT-2 and downregulated glucose transporter GLUT1. Notably, when naïve melanoma cells become BMR cells by long-term exposure to BRAFi/MEKi, a stepwise degradation of AMPK-1 is initiated via ubiquitin-proteasome system (UPS). Using proteomic analysis, we discovered that a novel E3 ligase, RING finger 44 (RNF44), can interact with AMPK-1 and promote AMPK-1 degradation in both BR (BRAF resistance) and BMR cells. Elevated RNF44 and downregulated AMPK-1 expressions are also found in both BR BMR cells. Elevated RNF44 and downregulated AMPK-1 expressions are also found in both BR and BMR and tumor samples from BR and BMR patients.

Conclusions: Our results suggest that BRAFi resistance-driven proteasomal degradation of AMPK-1 is crucial to govern vulnerability to arginine deprivation in melanoma cells; therefore, depleting exogenous arginine can be considered as a novel salvage treatment for melanoma patients who fail BRAF/MEK inhibitors. Furthermore arginine deprivation also results in an increase PD-L1 expression which make them more sensitive to check point inhibitors (supported by VA Merit Review Award (1BX003328) to Savaraj and 1R01CA152197 to Kuo, Feun and Savaraj.

Background: Melanomas harboring BRAF mutation (V600E) is known to recur following treatment with BRAF inhibitors (BRAFi) despite a high initial response rate. While the addition of MEK inhibitor can increase both response rate and duration of response. Relapsed is still inevitable. We have found that both BRAFi-resistant melanoma (BR) and BRAFi/ MEKi resistant (BMR) cells are vulnerable to arginine depletion. This is due to the fact that they cannot readily undergo autophagy upon arginine deprivation and unable to turn on argininosuccinate synthetase 1 (ASS1) to synthesize arginine. In this report, we have studied the underlying mechanisms.

Methods: 4 pairs of melanoma cell lines (parental and its BRAFi/MEKi resistant variants): A375/A375BMR, Skmel28/Skmel28BMR, A2058/A2058BMR, and MEL1220/Mel120BMR) were used for the study. MEL1220 was established in our laboratory. BMR resistant cell line was generated by continuous exposure of parental to BRAF inhibitor (Vemurafenib) and MEK inhibitor (Trametinib).

Results/Discussion: We demonstrated that downregulation of AMPK-1 is a major factor contributing to impairment of autophagy. In addition, decreased AMPK-1 expression also results in metabolic reprogramming in BMR cells toward more dependency on arginine and less on glycolysis evidenced by upregulated arginine transporter CAT-2 and downregulated glucose transporter GLUT1. Notably, when naïve melanoma cells become BMR cells by long-term exposure to BRAFi/MEKi, a stepwise degradation of AMPK-1 is initiated via ubiquitin-proteasome system (UPS). Using proteomic analysis, we discovered that a novel E3 ligase, RING finger 44 (RNF44), can interact with AMPK-1 and promote AMPK-1 degradation in both BR (BRAF resistance) and BMR cells. Elevated RNF44 and downregulated AMPK-1 expressions are also found in both BR BMR cells. Elevated RNF44 and downregulated AMPK-1 expressions are also found in both BR and BMR and tumor samples from BR and BMR patients.

Conclusions: Our results suggest that BRAFi resistance-driven proteasomal degradation of AMPK-1 is crucial to govern vulnerability to arginine deprivation in melanoma cells; therefore, depleting exogenous arginine can be considered as a novel salvage treatment for melanoma patients who fail BRAF/MEK inhibitors. Furthermore arginine deprivation also results in an increase PD-L1 expression which make them more sensitive to check point inhibitors (supported by VA Merit Review Award (1BX003328) to Savaraj and 1R01CA152197 to Kuo, Feun and Savaraj.

Purpose: The time to response to immunotherapy is long and not suitable for rapidly growing BRAF resistant (BR) or BMR ( BRAF and MEK) resistant melanoma. We exploit a new approach to treat these tumors.

Background: We have previously shown that BR/BMR cells do not express argininosuccinate synthetase (ASS) and arginine deprivation induced apoptosis instead of autophagy (Oncotarget 7:14). We plan to exploited this alteration to treat BR/BMR.

Methods: Five BR cells A375, MEL-1220, A2058, UACC-62, and SK-MEL28 were established (Oncotarget). Arginine deprivation was achieved using arginine degrading enzyme (ADI-PEG20, Polaris) which degrade arginine to citrulline.

Results: BR cells are hypersensitive to ADI-PEG20. Treatment resulted in 10-30% increase in apoptosis compared to their parental cells. The mechanisms involved are as follows: All BR cells do not express ASS, and have attenuated glucose uptake. They acquire exogenous arginine by expressing high levels of arginine transporter CAT2. The mechanism for low levels of ASS is due to diminished c-Myc, a positive regulator of ASS. Additionally, AMPK-α1 (govern autophagy and glucose uptake) was attenuated in BR cells. This is proved by knockdown and overexpress AMPK-α1. Immunohistochemical staining further confirmed lower levels of AMPK-α1 in tumor tissues (average H-scores of ASS and AMPK in parental tissues vs BR (BMR) tissues are 58.2 vs 7.8, and 146 vs 78.3, respectively, P < .03). Importantly, these findings also apply to tumor from BMR patients. Interestingly, treatment with ADI-PEG20 leads to robust expression of immune checkpoint PD-L1 in both parental and BR cells and PBMCs from BR patients. Importantly, macrophage polarization may involve in metabolic reprogramming.

Conclusions: Attenuation of AMPK-α1-in BR results in diminished autophagy and metabolic alteration. These BR cells depend less on glucose but more on arginine, and hence vulnerable to arginine deprivation. Additionally, arginine deprivation upregulates PD-L1 expression and leads to sensitivity to anti PDL-1 antibody. Combination of ADI-PEG20 with checkpoint inhibitors can lead to robust antitumor effect in BR and BMR patients. Supported by VA Merit Review (BLR&D BX00333280-01) and R01CA152197.

Purpose: The time to response to immunotherapy is long and not suitable for rapidly growing BRAF resistant (BR) or BMR ( BRAF and MEK) resistant melanoma. We exploit a new approach to treat these tumors.

Background: We have previously shown that BR/BMR cells do not express argininosuccinate synthetase (ASS) and arginine deprivation induced apoptosis instead of autophagy (Oncotarget 7:14). We plan to exploited this alteration to treat BR/BMR.

Methods: Five BR cells A375, MEL-1220, A2058, UACC-62, and SK-MEL28 were established (Oncotarget). Arginine deprivation was achieved using arginine degrading enzyme (ADI-PEG20, Polaris) which degrade arginine to citrulline.

Results: BR cells are hypersensitive to ADI-PEG20. Treatment resulted in 10-30% increase in apoptosis compared to their parental cells. The mechanisms involved are as follows: All BR cells do not express ASS, and have attenuated glucose uptake. They acquire exogenous arginine by expressing high levels of arginine transporter CAT2. The mechanism for low levels of ASS is due to diminished c-Myc, a positive regulator of ASS. Additionally, AMPK-α1 (govern autophagy and glucose uptake) was attenuated in BR cells. This is proved by knockdown and overexpress AMPK-α1. Immunohistochemical staining further confirmed lower levels of AMPK-α1 in tumor tissues (average H-scores of ASS and AMPK in parental tissues vs BR (BMR) tissues are 58.2 vs 7.8, and 146 vs 78.3, respectively, P < .03). Importantly, these findings also apply to tumor from BMR patients. Interestingly, treatment with ADI-PEG20 leads to robust expression of immune checkpoint PD-L1 in both parental and BR cells and PBMCs from BR patients. Importantly, macrophage polarization may involve in metabolic reprogramming.

Conclusions: Attenuation of AMPK-α1-in BR results in diminished autophagy and metabolic alteration. These BR cells depend less on glucose but more on arginine, and hence vulnerable to arginine deprivation. Additionally, arginine deprivation upregulates PD-L1 expression and leads to sensitivity to anti PDL-1 antibody. Combination of ADI-PEG20 with checkpoint inhibitors can lead to robust antitumor effect in BR and BMR patients. Supported by VA Merit Review (BLR&D BX00333280-01) and R01CA152197.

Purpose: The time to response to immunotherapy is long and not suitable for rapidly growing BRAF resistant (BR) or BMR ( BRAF and MEK) resistant melanoma. We exploit a new approach to treat these tumors.

Background: We have previously shown that BR/BMR cells do not express argininosuccinate synthetase (ASS) and arginine deprivation induced apoptosis instead of autophagy (Oncotarget 7:14). We plan to exploited this alteration to treat BR/BMR.

Methods: Five BR cells A375, MEL-1220, A2058, UACC-62, and SK-MEL28 were established (Oncotarget). Arginine deprivation was achieved using arginine degrading enzyme (ADI-PEG20, Polaris) which degrade arginine to citrulline.

Results: BR cells are hypersensitive to ADI-PEG20. Treatment resulted in 10-30% increase in apoptosis compared to their parental cells. The mechanisms involved are as follows: All BR cells do not express ASS, and have attenuated glucose uptake. They acquire exogenous arginine by expressing high levels of arginine transporter CAT2. The mechanism for low levels of ASS is due to diminished c-Myc, a positive regulator of ASS. Additionally, AMPK-α1 (govern autophagy and glucose uptake) was attenuated in BR cells. This is proved by knockdown and overexpress AMPK-α1. Immunohistochemical staining further confirmed lower levels of AMPK-α1 in tumor tissues (average H-scores of ASS and AMPK in parental tissues vs BR (BMR) tissues are 58.2 vs 7.8, and 146 vs 78.3, respectively, P < .03). Importantly, these findings also apply to tumor from BMR patients. Interestingly, treatment with ADI-PEG20 leads to robust expression of immune checkpoint PD-L1 in both parental and BR cells and PBMCs from BR patients. Importantly, macrophage polarization may involve in metabolic reprogramming.

Conclusions: Attenuation of AMPK-α1-in BR results in diminished autophagy and metabolic alteration. These BR cells depend less on glucose but more on arginine, and hence vulnerable to arginine deprivation. Additionally, arginine deprivation upregulates PD-L1 expression and leads to sensitivity to anti PDL-1 antibody. Combination of ADI-PEG20 with checkpoint inhibitors can lead to robust antitumor effect in BR and BMR patients. Supported by VA Merit Review (BLR&D BX00333280-01) and R01CA152197.

Background: Since approval of kinase inhibitors and programmed death receptor-1 (PD-1) and CTLA-4 blocking antibody, therapeutic approach to treat metastatic melanoma have demonstrated better efficacy then traditional chemotherapy. While kinase inhibitors target specific mutations and have demonstrated response rate around 50%, resistance to treatment ultimately develops. Treatment with check point inhibitor can produce durable response which translated into long-term survival, but the key question is that the marker(s) to predict response is not
clear yet. Other questions are: whether both drugs have equal efficacy, target similar domain on PD-1, and possible cross resistance. Here, we report a melanoma patient who clearly progressed after 10 doses of nivolumab. The treatment was changed to pembrolizumab with good response.

Case: We report a case of a 76-year-old man with aggressive metastatic melanoma and a 3-year history of treatment. After lymph node dissection in Dec 2013 for a T4bN3 disease of neck (BRAF V600E positive), he was planned for postoperative radiation. Unfortunately, his malignancy recurred very quickly in Jan 2014 with dermal infiltration and subcutaneous nodule. Initial treatment with vemurafenib was not tolerated. Subsequently, he was given trametinib plus dabrafenib with some response, then melanoma recurred after 10 months. Ipilimumab was started, patient tolerated it well, but disease progressed. Nivolumab was started, but disease progressed with lymph nodes, liver, lung and spleen metastasis. While there are no data, we decided to try pembrolizumab and the patient showed a significant response, with disappearing of the metastasis. The only metastatic site that remained was in spleen. Interestingly, biopsy from metastatic lesion in the spleen showed 40% of tumor cells were positive for PD-L1 in the membrane but immune cells were negative. Patient has received a total of 29 cycles of pembrolizumab and is doing well.

Conclusions: This case provides an interesting observation of long duration of response of pembrolizumab after failing nivolumab. This could be due to some pharmacological differences between these inhibitors as well as variations of immune cells in the tumor microenvironment. Further study is needed to clarify this issue.

Background: Since approval of kinase inhibitors and programmed death receptor-1 (PD-1) and CTLA-4 blocking antibody, therapeutic approach to treat metastatic melanoma have demonstrated better efficacy then traditional chemotherapy. While kinase inhibitors target specific mutations and have demonstrated response rate around 50%, resistance to treatment ultimately develops. Treatment with check point inhibitor can produce durable response which translated into long-term survival, but the key question is that the marker(s) to predict response is not
clear yet. Other questions are: whether both drugs have equal efficacy, target similar domain on PD-1, and possible cross resistance. Here, we report a melanoma patient who clearly progressed after 10 doses of nivolumab. The treatment was changed to pembrolizumab with good response.

Case: We report a case of a 76-year-old man with aggressive metastatic melanoma and a 3-year history of treatment. After lymph node dissection in Dec 2013 for a T4bN3 disease of neck (BRAF V600E positive), he was planned for postoperative radiation. Unfortunately, his malignancy recurred very quickly in Jan 2014 with dermal infiltration and subcutaneous nodule. Initial treatment with vemurafenib was not tolerated. Subsequently, he was given trametinib plus dabrafenib with some response, then melanoma recurred after 10 months. Ipilimumab was started, patient tolerated it well, but disease progressed. Nivolumab was started, but disease progressed with lymph nodes, liver, lung and spleen metastasis. While there are no data, we decided to try pembrolizumab and the patient showed a significant response, with disappearing of the metastasis. The only metastatic site that remained was in spleen. Interestingly, biopsy from metastatic lesion in the spleen showed 40% of tumor cells were positive for PD-L1 in the membrane but immune cells were negative. Patient has received a total of 29 cycles of pembrolizumab and is doing well.

Conclusions: This case provides an interesting observation of long duration of response of pembrolizumab after failing nivolumab. This could be due to some pharmacological differences between these inhibitors as well as variations of immune cells in the tumor microenvironment. Further study is needed to clarify this issue.

Background: Since approval of kinase inhibitors and programmed death receptor-1 (PD-1) and CTLA-4 blocking antibody, therapeutic approach to treat metastatic melanoma have demonstrated better efficacy then traditional chemotherapy. While kinase inhibitors target specific mutations and have demonstrated response rate around 50%, resistance to treatment ultimately develops. Treatment with check point inhibitor can produce durable response which translated into long-term survival, but the key question is that the marker(s) to predict response is not
clear yet. Other questions are: whether both drugs have equal efficacy, target similar domain on PD-1, and possible cross resistance. Here, we report a melanoma patient who clearly progressed after 10 doses of nivolumab. The treatment was changed to pembrolizumab with good response.

Case: We report a case of a 76-year-old man with aggressive metastatic melanoma and a 3-year history of treatment. After lymph node dissection in Dec 2013 for a T4bN3 disease of neck (BRAF V600E positive), he was planned for postoperative radiation. Unfortunately, his malignancy recurred very quickly in Jan 2014 with dermal infiltration and subcutaneous nodule. Initial treatment with vemurafenib was not tolerated. Subsequently, he was given trametinib plus dabrafenib with some response, then melanoma recurred after 10 months. Ipilimumab was started, patient tolerated it well, but disease progressed. Nivolumab was started, but disease progressed with lymph nodes, liver, lung and spleen metastasis. While there are no data, we decided to try pembrolizumab and the patient showed a significant response, with disappearing of the metastasis. The only metastatic site that remained was in spleen. Interestingly, biopsy from metastatic lesion in the spleen showed 40% of tumor cells were positive for PD-L1 in the membrane but immune cells were negative. Patient has received a total of 29 cycles of pembrolizumab and is doing well.

Conclusions: This case provides an interesting observation of long duration of response of pembrolizumab after failing nivolumab. This could be due to some pharmacological differences between these inhibitors as well as variations of immune cells in the tumor microenvironment. Further study is needed to clarify this issue.