Search

e20016 Background: An elevated serum LDH level is an adverse prognostic factor in NDMM. However, this category includes quantitative serum LDH levels that range from just over the upper limit of normal (ULN) to levels that may be 2 or more-fold higher than the ULN. This binary classification of serum LDH level of “normal versus elevated” fails to discriminate between the different disease biology that exists among NDMM patients with elevated serum LDH levels. Thus, we attempted to further stratify NDMM patients by the level of their serum LDH and determine its impact on OS. Methods: The cohort included patients diagnosed with NDMM from the Mayo Clinic, Rochester from 2003 - 2017 who were treated with novel agent induction therapy and had serum LDH levels measured at the time of diagnosis. The serum LDH levels were stratified into three levels: Normal (LDH < 222 U/L), Elevated (LDH 223-444 U/L), and Very Elevated (LDH >444 U/L or >2x upper limit of normal). Survival analysis was performed using the Kaplan-Meier survival analysis and compared via the log-rank method. Results: The cohort consists of 1,196 NDMM patients with a median age of 65 (22 – 95). R-ISS classification and cytogenetic risk were available for 968 and 970 patients respectively. The median serum LDH level was (162 U/L (3- 1260)) and an elevated LDH was present in 199 patients (17%). The median OS for patients with normal (N = 997; 83%), elevated (N = 170; 13%) and very elevated (N = 29; 3%) LDH levels were 76 months, 57 months and 23 months respectively (P < 0.001). The impact of these different levels of LDH on OS by R-ISS stage and cytogenetic risk is shown in the Table. Conclusions: A very small subset of NDMM patients has very elevated LDH levels that confer an exceptionally poor OS irrespective of R-ISS stage and cytogenetic risk. Future studies elucidating their disease biology responsible for such poor OS outcomes are warranted.[Table: see text]

e19524 Background: Multiple myeloma (MM) is a plasma cell neoplasia characterized by a diffuse clonal plasma cell infiltration of the bone marrow. Serum lactate dehydrogenase (LDH) is a relevant biomarker in MM and in the era of novel agents, LDH confirmed its negative impact on survival in newly diagnosed setting (Bal et al., presented at ASH2018). Methods: This single center retrospective study included 208 patients with a diagnosis of MM who underwent ASCT at our center between January 2008 - Mart 2018 were prospectively analyzed. We identified patients with baseline serum LDH values (ULN = 247 U/L). We compare baseline characteristics and outcomes of ASCT according to LDH values. Results: All clinical data were available in 208 cases (High LDH: 92 (44.2%), Normal LDH: 116 (55.8%)) There were 87 (41.8 %) female and 87 (41.8 %) male patients. The median age at diagnosis of MM was 63 years (range, 37-77 years). The median time of follow-up was 48 months (range, 2.8-197.5 months). Of the 92 patients with high LDH who underwent ASCT, median age was 65 years and ECOG performance status was 1 and 58.6 % were male. There was no statistically differences in performance status, induction treatment regimens and the number of treatment line prior to ASCT. Induction therapy consisted of 3 drugs in 91.3% and 2 drugs in 8.7 % in high LDH group (p = NS). Forty-eight (53.3%) patients with high LDH received a bortezomib-based induction and 35 patients (40.2 %) achieved ³ very good partial remission (VGPR) after the induction therapy. During follow up, 45 relapses were occurred and all relapsed patients died. According to the long-rank test, estimated 5-year OS was not significantly different between normal and high LDH categories (63.3 %±0.6 % vs. 64.6 %±0.5 %; p = 0.99). Median PFS was 101.4 months (95% CI: 73.7-129) in high LDH group and 92.1 months (95% CI: 71.1-113.1) in normal LDH group (p = 0.99). Conclusions: Elevated LDH was confirmed as a poor prognostic factor in previous reports. In our transplanted patients with high LDH cohort, clinical outcomes are favorable and similar those have normal LDH. Therefore, further studies of larger randomized cohorts are required to clarify the impact of pre-transplant LDH levels on ASCT outcomes.

e21054 Background: Up to 50% of patients undergoing resection for advanced melanoma experience recurrence. Identification of preoperative prognostic biomarkers is needed to ascertain risk of relapse and guide postoperative management. Lactate dehydrogenase (LDH) represents a strong prognostic factor in unresectable metastatic (stage IV) melanoma, but its relevance in patients with resected stage III or IV disease remains unknown. Methods: We retrospectively analyzed data from patients with stage III and IV melanoma who had undergone complete resection of disease and received follow-up treatment at Mayo Clinic, Rochester between January 1, 2000 and January 31, 2012. Clinical data were collected from electronic records. Survival data were estimated using the Kaplan-Meier method. Associations of preoperative LDH with time to relapse and death were evaluated using Cox proportional hazards regression models and summarized with hazard ratios and 95% confidence intervals. Results: A total of 154 subjects with resectable stage III or IV melanoma were included in the study. Median age at the time of resection was 58; 54 (35.1%) were female. One-hundred sixteen (75.3%) patients were classified as stage III and 38 (24.7%) stage IV. Adjuvant systemic treatment was administered in 75 (48.7%) patients and adjuvant radiation in 32 (20.7%). Median duration of follow-up was 4.0 years. Sixteen (10.3%) patients had preoperative LDH above the upper limit of normal. Each 50-unit increase in LDH was associated with a 15% increased risk of relapse (HR 1.15; p = 0.040) and 23% increased risk of death (HR 1.23; p = 0.001). After adjusting for age, gender, stage, number of sites, adjuvant systemic treatment, and adjuvant radiation, preoperative LDH remained associated with time to death (HR 1.25; p = 0.002). Preoperative LDH greater than the upper limit of normal was associated with increased hazard of death, both with univariate (HR 2.44; p = 0.005) and multivariate (HR 2.17; p = 0.017) analyses. Conclusions: This study supports the role of elevated preoperative LDH as a predictor of inferior outcomes in patients with advanced melanoma. Further study to correlate LDH to outcomes in the era of adjuvant immunotherapy is required.

3036 Background: Ipilimumab, a CTLA4 blocking antibody, has been shown to improve survival in metastatic melanoma patients in phase III trials. However, only a subset of patients experiences long-term survival benefit. Therefore, we analysed two independent retrospective sets of patients treated in ‘real world’ settings to identify prognostic factors that correlate with better outcome after ipilimumab therapy. Methods: Advanced melanoma patients, eligible for ipilimumab therapy, were treated in the Dutch and UK Expanded Access Programs (EAP) and after European licensing on the 3mg/kg schedule. Baseline characteristics and peripheral blood parameters were assessed and patients were monitored for the occurrence of adverse events and responses. Results: A total of 166 patients were enrolled in the Dutch EAP. Best overall response rate and disease control rate were (DCR) 17% and 35%. Median follow-up was 17.9 months, with a median progression free survival of 2.9 months. Median overall survival (OS) was 7.5 months, and OS at 1 year 37.8% and at 2 years 22.9%. Univariate analysis revealed that gender, age, Breslow thickness, baseline and week 6 lymphocyte count (ALC), and prior treatment had no influence on OS, while mWHO, M stage, baseline LDH, S100, erythrocyte sedimentation rate (ESR), and the slope of ALC did. In a multivariate model only baseline LDH and ESR remained as significant independent prognostic factors. The relevance of LDH was validated in an independent cohort of 68 patients from the UK. Stratifying the patients in the NL cohort according to LDH levels (≤upper limit normal (ULN), 54.4% of patients versus >2xULN, 16.9% of patients) separated into groups of patients observing DCR of 48% versus 14%, and median OS of 13.7 versus 3.1 months, while toxicity was similar in both groups (48% and 41%). None of the patients was alive at 15 months after treatment if baseline LDH was >2xULN in both the NL and UK cohorts. Conclusions: Overall survival upon ipilimumab treatment is lower in an expanded access population as compared to phase III trials. LDH >2xULN at baseline is a potential prognosticator in patients receiving ipilimumab and should be considered in guiding ipilimumab treatment initiation.

8047 Background: The prognostic significance of high serum LDH is well established for both indolent and aggressive lymphomas at diagnosis. The performance characteristics of LDH in predicting relapse after treatment in patients (pts) with DLBCL has not been well studied. The determination of utility of LDH monitoring in pts with DLBCL would have widespread practice implications. Methods: We searched the Nebraska Lymphoma Study Group database to identify pts with DLBCL who were treated with a rituximab-based regimen, from 2000 to 2010, and sustained a complete response (CR). For the pts who relapsed (RP), after sustaining a CR, we collected the LDH level at relapse and the LDH level 3 months prior (considered baseline). For pts who never relapsed (NR), we collected the last 2 LDH levels at follow-up; levels had to be at least 3 months apart. The relative increase of LDH was compared, to baseline, among RP vs. NR. Results: We identified 129 pts, 15 pts were excluded from the analysis as their LDH results were not available, 27 relapsed and 87 didn’t. Median age of pts was 57 years. Only 9/27 RP (33%) had increase in LDH above upper limit of normal (ULN) at relapse. The mean increase in LDH at relapse was 1.2 fold above the ULN for RP vs. 0.83 for NR (p=0.59). The mean increase in LDH, from baseline, was 1.1 fold in NR vs. 1.3 in RP (p=0.3). The likelihood ratio (LR) of relapse was 4.65 for pts who had 1.5 fold increase in LDH above baseline (1.5xLDH) vs. those who didn’t (p=0.03). The sensitivity, specificity, positive and negative predictive values of 1.5xLDH for detecting relapse, compared to clinical and imaging findings were 0.18, 0.95, 0.55, and 0.79 respectively. Also, 1.5xLDH at relapse was significantly associated with the presence of fever (LR=5.74; p=0.03) but not with other symptoms including drenching night sweats, anemia, or new/progressive lymphadenopathy. Conclusions: A 1.5 fold increase in LDH, over a period of 3 months, is associated with increased likelihood of relapse from DLBCL. Modest elevations in LDH (

8093 Background: In patients with symptomatic multiple myeloma, the clinical features, responses to treatment, and survival times vary. Well established predictors of survival include the International Staging System (ISS), cytogenetic abnormalities, and response to therapy. Long recognized has been the association of high serum lactate dehydrogenase (LDH) with advanced disease and shorter survival. We focused here on the impact of high LDH on staging and prognosis in order to guide the role of recent advances in therapy. Methods: We evaluated 1,247 patients with newly diagnosed, symptomatic myeloma from 10/74 to 7/11. Our goal was to determine the prognostic value of high LDH (>300 IU/L) in relation to ISS stage. We also compared the frequencies of anemia, hypercalcemia, and response to therapy in patients with high LDH with those of patients with Stage III disease and normal LDH values. Results: All 1,139 patients with normal LDH lived significantly longer than the 108 patients with elevated values (47 vs. 16 months, p 11.5 mg/dl (41 vs. 27%, p

514 Background: No predictive marker for efficacy of a Bevacizumab-based therapy has been identified in metastatic colorectal cancer. Soluble LDH levels seem to be related to increased angiogenesis. The focus of this analysis is whether high levels of LDH can be related to higher benefit of bevacizumab-based therapy in this setting. Methods: Eligible patients were metastatic colorectal cancer patients treated in 1st line setting with a chemotherapy doublet and bevacizumab in the last year. A historical control group was also formed by collecting data of patients treated with only a chemotherapy doublet. Pre-treatment LDH serum levels were collected for all patients. Cut-off level for soluble LDH was determined via ROC curve. Primary end-point was difference in progression free survival between the two groups. Secondary end-points were overall survival and response rates. Results: 82 patients treated with first line containing bevacizumab were eligible for our analysis. 138 patients were identified as control group. The two groups were comparable for major clinical characteristics. LDH cut-off level was stated at 588 mg/dl. In the control group PFS of 4.2 and 8 months were observed for LDH high and low levels (p=0.0003,HR:0.2973,95%CI:0.0318 to 0.3543). OS were 19.6 and 34.9 months (p=0.0014,HR:0.2484,95%CI: 0.0188 to 0.3884). No significant differences were seen for response rates (with 20% vs 39% partial responses, p=0.1671). In the group of patients treated with Bevacizumab, PFS of 8.5 and 7.3 months were observed respectively for LDH high and low levels (p=0.2,HR:0.6360,95%CI:0.2528 to 1.4185). OS were 26.6 and 22 months (p=0.7,HR:0.8480,95%CI:0.2307 to 2.9563). Improved response rate for LDH high patients could be seen (58% vs 14% partial responses, p=0.0243). Conclusions: These data confirm the role of soluble LDH as marker of worse outcome for unselected patients and also suggest a role as potential marker of increased effectiveness of Bevacizumab based therapy.

e15621 Background: Though serum LDH level is frequently elevated in the patients with advanced gastric cancer, its clinical significance is still elusive. Moreover, the relationship between the change of serum LDH level after chemotherapy and the response to the treatment has not been studied, yet. We analyzed serum LDH level as a prognostic factor for the patients with advanced stomach cancer. Methods: We assessed serum LDH level before chemotherapy for the patients who were planned to receive palliative chemotherapy. We re-assessed their serum LDH level at the time when the response to chemotherapy was evaluated after 2–4 cycles of treatment. The survival duration and the response to chemotherapy for the patients with low serum LDH level were compared to the survival duration and the response to chemotherapy for the patients with high serum LDH level. The relationship between the change of serum LDH level and the response to the treatment was evaluated, too. Results: Total 118 patients were entered into this study and 114 patients were evaluable for their response to chemotherapy. Pre-treatment serum LDH level was normal in 88 patients and elevated in 30 patients. The response rate in the patients with high serum LDH level was significantly higher than the response rate in the patients with normal serum LDH level (34.5% versus 15.3%, p < 0.05). However, the patients with normal serum LDH level lived longer than the patients with high serum LDH level (median: 378 days versus 206 days, p < 0.001). The normalizing of the elevated serum LDH level after chemotherapy was related to the good response to treatment (response rate 50.0% versus 18.8%, p < 0.05). Conclusions: For the patients with advanced gastric cancer, high serum LDH level was related to better response to chemotherapy but shorter survival duration. The normalization of elevated serum LDH level after chemotherapy was related to good response to treatment. No significant financial relationships to disclose.

16086 Background: Increase in serum LDH levels predicts a poor outcome in a spectrum of neoplastic diseases. Based on a clinical observation that several RCC patients receiving sunitinib developed ...

3530 Background: It is well known that angiogenesis and glycolysis are regulated by hypoxic conditions. Recent clinical trials (CONFIRM1 and CONFIRM2) have shown that patients with mCRC with high serum LDH benefited from PTK787/ZK 222584, a VEGF receptor tyrosine kinase inhibitor. We tested the hypothesis that patients with high serum LDH have increased intratumoral expression of genes involved with hypoxia (hypoxia inducible factor (HIF1a and 2a) and lactate dehydrogenase A (LDHA) and glycolysis (glucose transporter 1 (Glut-1) and genes involved in angiogenesis such as vascular endothelial growth factor A (VEGFA) and neuropilin 1 (NRP1) in patients with mCRC. Methods: 78 formalin fixed paraffin embedded (FFPE) tumor samples from 36 patients (20 males, 16 females: Median age 59 years (range 29–84) with mCRC who underwent first line therapy (not from CONFIRM trials) were analyzed. In addition, tumor gene expression was correlated with serum LDH levels from the same group of patients. FFPE tissues were dissected using laser-captured microdissection and analyzed LDHA, VEGFA, HIF1a, HIF2a, Glut-1 and NRP1 mRNA expression using a quantitative real-time RT-PCR method. Gene expression values (relative mRNA levels) are expressed as ratios between the target gene and internal reference gene (beta-actin). Results: Spearman Rank Correlation Analysis of Associations Between serum LDH levels and Gene Expression values. Conclusions: Our results demonstrate that intratumoral gene expression of LDHA, HIF1a and HIF2a, Glut-1 and VEGFA are significantly correlated. Patients with high serum LDH have increased intratumoral gene expression of VEGFA. These results support the hypothesis that serum LDH levels may serve as a surrogate marker for activation of the HIF related genes in the tumor. These observations may explain the efficacy of PTK787 in metastatic colorectal cancer patients with high serum LDH levels. [Table: see text] [Table: see text]

18001 Background: Bcl-2 inhibits apoptosis, confers resistance to chemotherapy in many cancers, is often over-expressed in advanced melanoma, and is a prime candidate for pro-apoptotic targeted therapy. Recent studies by the EORTC and others show that lactate dehydrogenase (LDH) is a powerful prognostic determinant in advanced melanoma patients (pts). The chief treatment effect of oblimersen (Obl) + dacarbazine (DTIC) was obtained from the largest cohort of normal LDH pts studied to date; detailed efficacy and safety data are reported. Methods: Previously untreated pts (N = 771) were randomized to receive DTIC 1000 mg/m2 IV Q 3W alone or preceded by Obl (7 mg/kg/d × 5D). Pts were stratified pre-randomization by performance status, liver metastases, and visceral metastases and/or elevated LDH. Serum LDH was ≤ the cutoff of 1.1 × upper limit of normal in 508 pts. Treatment group characteristics were well matched. Results: Combining Obl and DTIC significantly increased all efficacy outcomes in pts with baseline (BL) nonelevated (NE) LDH, including overall survival (median 11.4 vs 9.7 mo; P = .018), progression-free survival (3.1 vs. 1.6 mo; P = .0007), objective response (17.2% vs 9.3%; P = .009), complete response (3.4% vs 0.8%; P = .041), and durable response ≥6 mo (9.6% vs 4.0%; P = .014). In contrast, there were no significant differences in outcomes for pts with elevated BL LDH (n = 252). BL LDH was the only stratification variable significantly associated with a treatment interaction effect. Compared with the entire study population, pts with NE LDH in both groups had a lower incidence of treatment-emergent adverse events (TEAEs) with an outcome of death, serious TEAEs, and discontinuations due to TEAEs. Conclusions: With extended follow-up, this large trial confirms that serum LDH (prospectively stratified pre-randomization) is probably the most important prognostic factor in advanced melanoma. Given the demonstrable lack of benefit and high potential to yield misleading evidence regarding therapeutic activity, pts with elevated LDH should probably be excluded from trials of new therapies until establishing efficacy in pts with nonelevated LDH. No significant financial relationships to disclose.