The Therapeutic Landscape for KRAS-Mutated Colorectal Cancers

Colorectal cancer is one of the world's most prevalent and lethal cancers. Mutations of the KRAS gene occur in ~40% of metastatic colorectal cancers. While this cohort has historically been difficult to manage, the last few years have shown exponential growth in the development of selective inhibitors targeting KRAS mutations. Their foremost mechanism of action utilizes the Switch II binding pocket and Cys12 residue of GDP-bound KRAS proteins in G12C mutants, confining them to their inactive state. Sotorasib and Adagrasib, both FDA-approved for the treatment of non-small cell lung cancer (NSCLC), have been pivotal in paving the way for KRAS G12C inhibitors in the clinical setting. Other KRAS inhibitors in development include a multi-targeting KRAS-mutant drug and a G12D mutant drug. Treatment resistance remains an issue with combination treatment regimens including indirect pathway inhibition and immunotherapy providing possible ways to combat this. While KRAS-mutant selective therapy has come a long way, more work is required to make this an effective and viable option for patients with colorectal cancer.

Adjuvant chemotherapy guided by circulating tumor DNA analysis in stage II colon cancer: The randomized DYNAMIC trial

LBA100

Background: The role of adjuvant chemotherapy (CT) in stage II colon cancer continues to be debated. The presence of circulating tumor DNA (ctDNA) after surgery predicts very poor recurrence-free survival (RFS), while its absence predicts a low recurrence risk. For ctDNA-positive cases the benefit of adjuvant CT is unknown. DYNAMIC was designed to assess if a ctDNA-guided approach could reduce the use of adjuvant CT without compromising recurrence risk. Methods: DYNAMIC is a multi-center randomized controlled phase II trial. Eligible patients had resected stage II colon cancer and were suitable for adjuvant CT. Patients were randomly assigned 2:1 to ctDNA-guided management or standard management (clinician-guided based on conventional criteria), after stratification for T stage and participating center location. Criteria for clinical low versus high risk were predefined. The Safe-SeqS tumor-informed personalized ctDNA assay was used. For ctDNA-guided management, a ctDNA-positive result at 4 or 7 weeks after surgery prompted oxaliplatin-based or fluoropyrimidine CT; ctDNA-negative patients were not treated. The primary efficacy endpoint was non-inferiority in RFS rate at 2 years. A key secondary endpoint was adjuvant CT use. The target sample size of 450 provided 80% power with 95% confidence to confirm non-inferiority between the two arms with a margin of 8.5%. Results: Of 455 patients randomized between Aug 2015 and Aug 2019, 302 were assigned to ctDNA-guided and 153 to standard management. Median follow-up was 37 months. In the ctDNA-guided arm, ctDNA analysis was successful in all but three patients; only two patients did not receive ctDNA-guided management. In the intention to treat population, fewer patients overall in the ctDNA-guided arm received adjuvant CT compared to standard management (15.3% vs 27.9%, odds ratio 2.14; P = 0.002), with the largest difference seen in patients with T4 or poorly differentiated tumors (odds ratios 6.22 and 6.31, respectively). Of those who received treatment, oxaliplatin-based doublet was more frequently administered than fluoropyrimidine alone for ctDNA-guided compared to standard management patients (62.2% vs. 9.8%; P < 0.001). ctDNA-guided management was non-inferior to standard management for 2-year RFS (93.5% vs 92.4%, difference 1.1%, 95% confidence interval, -4.1% to 6.2%). Following adjuvant CT, 3-year RFS for ctDNA-positive patients was 86.4%. Without adjuvant CT, 3-year RFS for ctDNA-negative patients was 92.5%, with a 3-year RFS of 96.7% in the clinical low risk subgroup. Conclusions: A ctDNA-guided approach to stage II colon cancer reduced adjuvant chemotherapy use without compromising recurrence-free survival. The low recurrence rate in ctDNA-positive patients who received chemotherapy suggests a survival benefit from adjuvant therapy. ctDNA-negative patients are unlikely to benefit from chemotherapy. Clinical trial information: ACTRN12615000381583.

SPAR: A randomized placebo-controlled phase 2 trial of simvastatin in addition to standard chemotherapy and radiation in preoperative treatment for rectal cancer: An AGITG clinical trial

TPS3646

Background: In retrospective studies statin use during preoperative chemo-radiation (pCRT) for rectal cancer is associated with improved overall survival, pathological tumor response and treatment toxicity. In vivo preclinical studies show that statins radiosensitize cancer cells, with improved tumor control and reduced radiation-induced gastrointestinal (GI) and skin toxicities. A prospective randomized trial is justified to confirm these clinically important benefits. Tumor regression following pCRT has strong prognostic significance, as assessed radiologically (MRI-based tumor regression grading [mrTRG]) prior to, or pathologically (pathTRG) following, surgery. Using mrTRG after each treatment phase in total neoadjuvant therapy (TNT) programs could assess incremental tumor regression and optimize patient management including surgery. Methods: Design: This double-blind phase 2 trial is recruiting 222 patients planned to receive long-course fluoropyrimidine-based pCRT for rectal adenocarcinoma at 17 sites in New Zealand and Australia. Patients are randomized to simvastatin 40mg or placebo daily for 90 days, starting 1 week prior to pCRT, with minimization for major prognostic variables. Pelvic MRI at baseline and 6-8 weeks after pCRT will assess mrTRG. A protocol amendment allows TNT using consolidation chemotherapy after pCRT; MRI timing is unchanged. Primary objective: comparison of rates of grades 1-2 mrTRG following pCRT with simvastatin or placebo, considering mrTRG in 4 ordered categories (1, 2, 3, 4-5). Secondary objectives: comparison of rates of grades 1-2 pathTRG in resected tumors; incidence of > grade 2 acute GI and non-GI adverse events (AE); incidence of late GI AE; compliance with intended pCRT and trial medication; proportion of patients undergoing surgical resection post-pCRT; 3-year local recurrence rate, disease-free and cancer-specific survival; and pathological scores for radiation colitis. Tertiary and correlative objectives: association between mrTRG and pathTRG grouping; inter-observer scoring agreement on mrTRG and pathTRG; comparison of the association between tumor CD3+ and/or CD8+ T-cell infiltrates in diagnostic biopsies and pathTRG; intensity and distribution of subsets of infiltrating T-cells in irradiated resected normal and malignant tissue; and the effect of simvastatin on markers of systemic inflammation (modified Glasgow prognostic score and the neutrophil-lymphocyte ratio). Eligibility criteria exclude statin use within 6 weeks prior to trial entry, patients intolerant of statins, and planned use of oxaliplatin or biological agents during pCRT. Trial recruitment commenced April 2018 and 95 of 222 patients have been recruited as at 21 January 2022. Clinical trial information: ACTRN12617001087347.

Australasian Gastrointestinal Trials Group (AGITG) CONTROL NET Study: 177Lu-DOTATATE peptide receptor radionuclide therapy (PRRT) and capecitabine plus temozolomide (CAPTEM) for pancreas and midgut neuroendocrine tumours (pNETS, mNETS)—Final results

4122

Background: CAPTEM is an accepted regimen for patients (pts) with advanced pNETs. Single agent PRRT is now a standard of care for progressive WHO Grade 1/2 mNETs. High activity was seen with PRRT/CAPTEM in a single arm Phase I/II trial. This study aims to determine the activity of combining CAPTEM with PRRT in mNETs and pNETs pts. Methods: Non-comparative randomised open label parallel group phase II trial with 2:1 randomisation to PRRT/CAPTEM (experimental arm) vs. PRRT (mNETs control) and CAPTEM (pNETS control). PRRT/CAPTEM: 7.8GBq 177Lu Octreotate (Lutate) given intravenously (IV) on day 10 every 8 weeks for 4 cycles, with concurrent oral capecitabine 750mg/m2 b.i.d. days 1-14 and temozolomide 75mg/m2 b.i.d. days 10-14 every 56 day cycle, up to 4 cycles. PRRT alone: 7.8GBq 177Lu Octreotate (Lutate) given intravenously (IV) on day 1 every 8 weeks for 4 cycles. CAPTEM alone: Oral capecitabine 750mg/m2 b.i.d. days 1-14 and days 29-42; Oral temozolomide 75mg/m2 b.i.d. days 10-14 and 38-42 every 56 day (8w) cycle. Primary endpoint: Progression free survival (PFS). mNETS: At 15 months, assuming PFS 66.4% in control arm; target PFS ³ 80%; pNETS: At 12 months, assuming PFS 60% in control arm; target PFS ³ 75%. Secondary endpoints: Objective tumor response rate (complete or partial) (OTRR), overall survival (OS), adverse events (AEs). Results: 75 pts enrolled (Dec 2015 – Nov 2018): mNETs 33 PRRT/CAPTEM, 14 PRRT, median follow up (mFU) 60.3 months; pNETS 19 PRRT/CAPTEM, 9 CAPTEM, mFU 57.5 months (mo). Late Grade 3/4 haematologic AEs: mNETS: 2/32 (6%) PRRT/CAPTEM pts and 4/13 (31%) PRRT pts. Events included myelodysplastic syndrome (40 mo), leukaemia (60 mo), pancytopenia (50 mo), anaemia (32 mo), thrombocytopenia (7 mo). No late haematologic G3/4 AEs were reported in the pNETS cohort. No late renal toxicity was identified in all study arms. Conclusions: CONTROL NETs is the first randomized trial to demonstrate efficacy for PRRT in pNETs, in addition to a standard of care. Extended follow up confirms durable CAPTEM/PRRT activity, with superior PFS in pNETs. Late haematologic toxicity was seen in both mNET PRRT arms but was not higher with additional CAPTEM. The activity of CAPTEM/PRRT in pNETs should be tested in the phase III setting. Clinical trial information: ACTRN12615000909527. [Table: see text]

Options beyond BRAF targeted therapy in second-line treatment of patients with BRAFV600E mutant (BRAFmt) metastatic colorectal cancer (mCRC)

47

Background: BRAFmt is a negative prognostic factor in mCRC but also identifies a patient population that may benefit from BRAF targeted therapy. Results from recent trials (BEACON and SWOG1406) demonstrate improved survival outcomes in second- and third-line settings when combining a BRAF inhibitor, an EGFR inhibitor (EGFRi) +/- a MEK inhibitor. In both trials, irinotecan and cetuximab was the control arm with a dismal response rate of 2-4% and progression free survival (PFS) of only 2 months. This suggests chemotherapy plus an EGFRi may not be the optimal approach where BRAF-targeted therapies are not available or have failed. Methods: Data from July 2009 to September 2021 was analysed from TRACC, a multi-site Australian mCRC comprehensive prospective registry enrolling consecutive patients. Patient characteristics, treatment and survival outcomes were examined for patients treated with chemotherapy (CT) alone, with bevacizumab (BEV) or with an EGFRi. Results: Of 2046 registry patients, 256 (13%) harboured a BRAFmt. 72 BRAFmt patients had received second-line (28%) treatment, including CT alone (n = 28), CT plus BEV (n = 26), and CT plus EGFRi (n = 18). Baseline characteristics are shown in the table. Median second-line PFS was 3.3, 4.7 and 1.8 months, for CT alone, CT plus BEV and CT plus EGFRi respectively. Median overall survival (OS) was 8.7, 7.9 and 2.5 months respectively. In multivariate analysis, PFS when treated with CT plus EGFRi trended inferior to CT alone (p = 0.054) and CT plus BEV (p = 0.061), whereas for OS, treatment with CT plus EGFRi was inferior to CT alone (p = 0.038) and CT plus BEV (p = 0.015). Poor PFS was associated with age ≥ 65 years (HR 3.03, p < 0.001) and ECOG ≥ 2 (HR 2.62, p = 0.004), but not associated with a right side primary (p = 0.17), mismatch repair (MMR) status (p = 0.86), or ≥ 3 organs with metastases (p = 0.32). Poor OS was associated with age ≥ 65 years (HR 3.11, p < 0.001), ECOG ≥ 2 (HR 7.32, p < 0.001), right side primary (HR 3.03, p = 0.002) and proficient MMR status (HR 3.57, p = 0.018), but not associated with ≥ 3 organs with metastases (p = 0.17). Conclusions: Less than one-third of BRAFmt mCRC patients received second-line therapy in a real-world setting, indicating an urgency to explore activity of BRAF targeted therapy in the first line setting. Treated patients received limited benefit, with CT plus EGFRi PFS outcomes comparable to BEACON control arm (1.8 vs 2.0 months) and trending inferior to other options. The best OS outcomes were achieved with CT alone or CT plus BEV.[Table: see text]

Variable implementation of optimal therapeutic strategies in metastatic colorectal cancer: Reviewing rates of liver resection and triplet chemotherapy across Australian hospitals as potential quality indicators

248

Background: Quality indicators (QI) are essential to monitor the efficacy of cancer care and to guide quality improvement, however many are derived from ‘expert consensus’ and are not validated against outcomes. Moreover, the majority of oncological QI are defined in the surgical setting, with only a paucity of QI for the treatment of metastatic disease. We aimed to define and validate novel QI for metastatic colorectal cancer (mCRC) based on therapeutic approaches associated with a proven survival benefit. Methods: Data was analysed from TRACC, a multisite Australian registry collecting prospective demographic, tumour, treatment and outcome data for mCRC. We identified all patients diagnosed across 11 hospitals and explored variation by site with regards to patient and tumour characteristics, first-line chemotherapy administration and resection of oligometastatic disease. Log-rank testing and Kaplan-Meier curves compare overall survival (OS) between sites, and Pearson correlation was used to assess associations with each QI. Results: We examined data from 3132 patients diagnosed with mCRC between July 2009 – April 2021. Median age was 66 years (range 62 – 71 years by site), ECOG 0-1 81% (range 69 – 96% by site), and Charlson Comorbidity Index ≤2 43% (33 – 59% by site). Multivariate analysis confirmed association of known adverse prognostic factors with inferior OS (poor ECOG, right sided primary, KRAS or BRAF mutation, all p <0.05). Median OS for entire cohort was 26.2 months (95%CI 24.9 – 27.3 months), and varied by hospital site from 20.1 – 36.1 months (p<0.001). Of the QI evaluated, rate of triplet chemotherapy (FOLFOXIRI) administration (2.8 – 13.2% by site) was very strongly correlated with OS (R2 = 0.851), rate of liver resection (9.8 – 23.2% by site) was moderately correlated (R2 = 0.523), and rates of active treatment with first-line chemotherapy (63 - 90% by site) were weakly correlated (R2 = 0.209). Other proposed QI such as rates of lung metastases resection or chemotherapy administration in the elderly showed significant variation by site, but did not correlate with survival. Conclusions: There is significant variation in OS for patients with mCRC in these Australian hospitals, with major differences in treatment approaches. Treatment strategies known to improve survival outcomes, such as triplet FOLFOXIRI chemotherapy and resection of liver metastases, may be potential QI to benchmark and track quality improvement over time. Further analysis will determine the impact of baseline patient populations between sites, and to correlate these QI with other quality measures.

Chemotherapy at the end of life in colorectal and pancreatic cancer: Real-world data and trends over time

e18723

Background: Chemotherapy at the end of life (CEOL) is widely accepted as an indicator of aggressive care. However, the evidence is limited primarily to single-centre experiences, with no consensus regarding acceptable benchmarks for CEOL, nor how this may be changing over time and with novel treatment options. We describe ‘real world’ CEOL in two large, multisite Australian registries of metastatic colorectal cancer (mCRC) and both locally advanced and metastatic pancreatic cancer (PC). Methods: Data was analysed from the TRACC and PURPLE registries, two large prospective multisite Australian cancer registries collecting prospective demographic, tumour, treatment and outcome data for mCRC and PC respectively. We identified all decedents between May 2009 and November 2020, determined the proportion who died within 14 or 30 days of chemotherapy (14D / 30D), or within 30D after a new line of therapy, defined as the first cycle of a new treatment regimen. Using univariate analysis, we compared baseline demographic and clinicopathological variables and trends over time. Results: 1505 mCRC and 602 PC decedents were identified. 20.9% of decedents (21.6% mCRC and 19.3% PC) received chemotherapy within 30D, and 11.5% within 14D (12.3% mCRC and 9.6% PC). There were lower rates of 30D CEOL after the first cycle of a new line of therapy (4.3% mCRC and 5.7% PC). Rates of CEOL decreased over the study period for mCRC (median rate of initial 3-year period 28% versus 15% in last 3-year period), but remained largely static for PC (18.9% versus 17.9%). 30D CEOL was more likely with palliative than curative intent treatment (mCRC OR 1.6, 95% CI 1.14-2.25, p = 0.007, PC OR 5.3, 95% CI 1.6-17.8), and advanced rather than local disease in PC (PC OR 2.59, 95%CI 1.6-4.1, p < 0.001). There was a trend towards CEOL and poorer performance status (ECOG) across all groups, only significant for 30D CEOL mCRC (OR 1.51, 95% CI 1.04-2.2). There was no association between CEOL and patient age, gender, Charlson comorbidity score or lines of therapy. Conclusions: Real-world rates of CEOL are higher in our cohorts of mCRC and PC patients than historical benchmarks but comparable to contemporary reports, which may be due to a wider array of available active treatments. Overall rates are decreasing over time for mCRC but static for PC, which may reflect the poorer overall survival for PC and lack of new effective therapies. The lower rates of death after new lines of therapy may signify that CEOL is more likely with existing treatment regimens and that clinicians are less likely to initiate a new chemotherapeutic regimen at EOL.

177Lu-DOTATATE peptide receptor radionuclide therapy (PRRT) in patients with somatostatin-expressing neuroendocrine tumors: A real-world retrospective review of efficacy and safety from an Australian tertiary cancer Center

e16198

Background: There is relatively limited data guiding optimal treatment algorithms for inoperable or metastatic neuroendocrine tumours (NETs). The randomised NETTER 1 study established PRRT as a standard of care in midgut NET. The aim of this study was to retrospectively review the efficacy and safety of patients treated with 177Lu-DOTATATE at Royal Brisbane and Women’s Hospital. Methods: Consecutive patients with unresectable or metastatic NET who received their first 177Lu-DOTATATE therapy between 2010 and 2017 were identified. Paragangliomas were excluded. Response rates, progression free survival (PFS) and overall survival (OS) were assessed via a combination of chart review, post-therapeutic 68Ga DOTATATE PET/CT and multi-disciplinary meeting decisions. Gr3/4 haematological and renal safety data was analysed via chart review. Progression free survival (PFS) and overall survival (OS) were assessed via the Kaplan-Meier method. Results: 123 consecutive patients were included in the analysis (57% male, median age 61). Primary site: 41% gastroenteric, 34% pancreatic, 10% lung, 6% colon/rectum, 9% other. Grade 1 (42%), Grade 2 (31%) Grade 3 (6%), unknown (21%). 76% of patients had previously received somatostatin analogues. 80% received four cycles (with persistent myelosuppression the most common cause of ceasing therapy early). 20% of patients received concurrent chemotherapy (10% capecitabine, 10% capecitabine and temozolomide). For the entire cohort, median follow up was 51 months. Median PFS, 34.2 months (95% CI, 30.3 – 38.17) and median OS, 61 months (95% CI, 54.2– 67.8). There was no significant difference in PFS or OS between small bowel and pancreatic patients. Response rates to PRRT included complete response (n = 4, 3%), stable or responding disease (n = 96, 78%), progressive disease (n = 22, 18%), not assessable (1, 1%). Grade 3 or 4 thrombocytopenia, neutropenia or anaemia occurred in less than 1% of patients. Confirmed acute leukaemia occurred in three patients. One patient had grade 3 renal toxicity secondary to treatment. 26 patients received retreatment with 177Lu-DOTATATE at progression with median time to re-treatment of 35 months. Conclusions: This single centre, retrospective audit confirms the efficacy of 177Lu-DOTATATE in the treatment of unresectable and metastatic NETs. Treatment was associated with low rates of severe haematological or renal adverse events. Our results are comparable to similar published reviews.

Outcomes of isolated distant lymph node metastases in colorectal cancer

84

Background: The optimal management of isolated distant lymph node metastases (IDLNM) in metastatic colorectal cancer (mCRC) is not clearly established. Small case series and prior data from the TRACC (Treatment of Recurrent and Advanced Colorectal Cancer) registry support the use of radical treatment with curative intent (local resection, chemo-radiation or stereotactic radiotherapy), which may lead to better outcomes in mCRC patients with IDLNM. Aims: This study investigates the clinical characteristics and outcomes of mCRC patients with IDLNM treated with systemic therapies plus locoregional therapy with curative intent versus systemic therapies with palliative intent. Methods: Clinical data were collected and reviewed from the TRACC registry, a prospective, comprehensive registry for mCRC from multiple tertiary hospitals across Australia from 01/07/2009 to 30/06/2020. Clinicopathological characteristics, treatment modalities and survival outcomes were analyzed in patients with IDLNM and compared to patients with other organ metastases. Fisher exact test was used for significance tests and Kaplan Meier curves for survival analyses. Results: Of 3408 mCRC patients with a median follow-up of 38.0 months, 93 (2.7%) were found to have IDLNM. Compared to mCRC with other organ metastases, patients with IDLNM were younger (mean age: 62.1 vs 65.6 years, p=0.0200), more likely to have metachronous disease (57.0% vs 38.9%, p=0.0005), be KRAS wild-type (74.6% vs 53.9%, p=0.0012) and BRAF mutant (12.9% vs 6.2%, p=0.0100). There was no overall survival difference between with IDLNM and those with other organ metastases (median OS 27.24 vs 25.92 months, p=0.2300). Twenty-four patients (25.8%) with IDLNM received treatment with curative intent, with a trend towards improved overall survival compared to those with other organ metastases treated with curative intent (73.5 vs 62.7 months, p=0.8200). Amongst mCRC patients with IDLNM, those who received treatment with curative intent had a significantly better overall survival than those treated with palliative intent (73.5months vs 23.2 months, p=0.0070). Conclusions: Our findings suggest that there are differences in the patterns of presentation of IDLNM and other organ metastases. Radical treatment with curative intent options should be considered for mCRC patients with IDLNM where appropriate.

Investigating real-world treatment sequencing outcomes in advanced pancreatic cancer: A purple translational registry analysis

386

Background: Current standard combination first line palliative chemotherapy regimens in advanced pancreatic ductal adenocarcinoma (PDAC) have not been compared in head-to-head trials. Data on optimum treatment sequencing is also lacking. Methods: To assess whether first line (1L) treatment with gemcitabine/nab-paclitaxel (Gem/Nab-P)(SEQ1) or FOLFIRINOX (SEQ2) in the palliative treatment setting impacts survival outcomes, data for patients receiving palliative intent combination chemotherapy between 2016 and May 2020 was extracted from PURPLE, a prospective pancreatic cancer registry enrolling consecutive patients across multiple institutions. Results: Of 637 patients, 180 (28%) who received 1L single agent therapy and/or palliative radiotherapy were excluded. Of the remaining 449 patients, 132 (29%) had locally advanced disease (LA PDAC), 67 had local recurrence (15%), and 250 (56%) had de novo metastatic disease (mPDAC). Patients receiving 1L Gem/Nab-P (n=376, 84%) were older (median 67 vs 59 years, P<0.001), had a higher Charlson Comorbidity Index (CCI) (CCI ≥2: 18% vs 3%, Odds Ratio [OR] 8.0, P=0.002), and poorer performance status (ECOG≥2: 10% vs 1%, OR 8.4, P=0.01) compared to the 1L FOLFIRINOX group (n=73, 16%). 140 (37%) patients receiving 1L Gem/Nab-P (SEQ1) and 32 (44%) patients receiving 1L FOLFIRINOX (SEQ2) received second line (2L) chemotherapy. SEQ1 2L regimens included FOLFIRINOX (n=14), FOLFIRI (n=49), FOLFOX (n=35) and 5FU alone (n=3). SEQ2 2L regimens included Gem/Nab-P (n=19), Gem/5FU (n=4), Gem/Cisplatin (n=1) and gemcitabine alone (n=5). Median progression free survival (mPFS) did not differ between patients receiving 1L Gem/Nab-P vs 1L FOLFIRINOX (5.7 vs 5.1 months, P=0.54); nor did median overall survival (mOS; 11.3 vs 12.3 months, P=0.37). In the subset of patients who went on to receive 2L chemotherapy, mPFS in 2L was shorter for SEQ1 compared to SEQ2 (2.9 vs 5.2 months, Hazard Ratio [HR] 1.3, P=0.03) but mOS did not differ (15.9 vs 17.3 months P=0.91). In the subset with LA PDAC, mPFS in 2L was comparable (2.9 vs 3.3 months, P=0.55) but mOS was significantly longer with SEQ1 (22.5 vs 13.8 months, HR 0.50, P=0.01). In mPDAC, mPFS in 2L was shorter with SEQ1 (2.3 vs 5.6 months, HR 1.64, P=0.03), but mOS did not differ by sequence (13 vs 17 months, P=0.23). Conclusions: There were no significant differences in survival outcomes between 1L choices of chemotherapy, despite patients offered 1L FOLFIRINOX (SEQ2) being younger and fitter. Survival differences observed for LA PDAC versus mPDAC will be further explored. Given the multiple potential confounders, randomised clinical trials are needed to make firmer conclusions regarding the optimal initial treatment for each patient subset.

Prognostic significance of postsurgery circulating tumor DNA in nonmetastatic colorectal cancer: Individual patient pooled analysis of three cohort studies

Studies in multiple solid tumor types have demonstrated the prognostic significance of ctDNA analysis after curative intent surgery. A combined analysis of data across completed studies could further our understanding of circulating tumor DNA (ctDNA) as a prognostic marker and inform future trial design. We combined individual patient data from three independent cohort studies of nonmetastatic colorectal cancer (CRC). Plasma samples were collected 4 to 10 weeks after surgery. Mutations in ctDNA were assayed using a massively parallel sequencing technique called SafeSeqS. We analyzed 485 CRC patients (230 Stage II colon, 96 Stage III colon, and 159 locally advanced rectum). ctDNA was detected after surgery in 59 (12%) patients overall (11.0%, 12.5% and 13.8% for samples taken at 4-6, 6-8 and 8-10 weeks; P = .740). ctDNA detection was associated with poorer 5-year recurrence-free (38.6% vs 85.5%; P < .001) and overall survival (64.6% vs 89.4%; P < .001). The predictive accuracy of postsurgery ctDNA for recurrence was higher than that of individual clinicopathologic risk features. Recurrence risk increased exponentially with increasing ctDNA mutant allele frequency (MAF) (hazard ratio, 1.2, 2.5 and 5.8 for MAF of 0.1%, 0.5% and 1%). Postsurgery ctDNA was detected in 3 of 20 (15%) patients with locoregional and 27 of 60 (45%) with distant recurrence (P = .018). This analysis demonstrates a consistent long-term impact of ctDNA as a prognostic marker across nonmetastatic CRC, where ctDNA outperforms other clinicopathologic risk factors and MAF further stratifies recurrence risk. ctDNA is a better predictor of distant vs locoregional recurrence.

Australasian Gastrointestinal Trials Group (AGITG) CONTROL NET Study: Phase II study evaluating the activity of 177Lu-Octreotate peptide receptor radionuclide therapy (LuTate PRRT) and capecitabine, temozolomide CAPTEM)—First results for pancreas and updated midgut neuroendocrine tumors (pNETS, mNETS)

4608

Background: CAPTEM is an accepted regimen for patients (pts) with advanced pNETs. Single agent 177Lu-Octreotate PRRT is now a standard of care for progressive WHO Grade (G) 1/2 mNETs. High activity was seen with LuTate/CAPTEM in a single arm Phase I/II trial. This study was undertaken to determine the relative activity of adding CAPTEM to LuTate PRRT in pts with mNETs and pNETs. Methods: Non-comparative randomised open label parallel group phase II trial with 2:1 randomisation to PRRT/CAPTEM (experimental arm) vs. PRRT (mNETs control) and CAPTEM (pNETS control). PRRT/CAPTEM: 7.8GBq LuTate day(D) 10, 8 weekly (wkly) x 4, with b.i.d. oral CAP 750mg/m2 D1-14 & TEM 75mg/m2D10-14, 8 wkly x 4; PRRT: 8 wkly x 4; CAPTEM 8 wkly x 4. Primary endpoint: Progression free survival (PFS). mNETS- at 15 months (mo) assuming 15mo PFS 66.4% in control arm, aiming for PFS ³ 80%; pNETS- at 12mo assuming 12mo PFS 60% in control arm, aiming for PFS ³ 75%. Secondary endpoints: Objective tumour response rate (complete or partial) (OTRR), clinical benefit rate (OTRR, stable disease) (CBR), toxicity, quality of life. Results: 75 pts enrolled (Dec 2015 – Nov 2018): mNETs 33 PRRT/CAPTEM and 14 PRRT; pNETS 19 PRRT/CAPTEM and 9 CAPTEM. mNETS: Median follow-up 35mo; 15mo PFS was 90% (95% CI: 73-97%) v 92% (95% CI: 57-99%); OTRR 31% vs 15%; and CBR 97% vs 92% for PRRT/CAPTEM v PRRT respectively. Treatment related adverse events (AEs): 24/32 PRRT/CAPTEM pts had at least one G3 event (75%) vs 5/13 (38%, PRRT); and 4/32 pts at least one G4 event (13%) v 1/13 (8%) respectively, mostly haematologic (haem). Only one patient failed to complete therapy (PRRT/CAPTEM). pNETS: Median follow-up 34mo; 12mo PFS was 76% (95% CI: 48-90%) v 67% (95% CI: 28-88%); OTRR 68% vs 33%; and CBR 100% vs 100% for PRRT/CAPTEM v CAPTEM respectively. Treatment related AEs: 5/18 PRRT/CAPTEM pts had at least one G3 event (28%) vs 3/9 (33%) CAPTEM; 3/18 pts at least one G4 event (17%) v 1/9 (11%) respectively. Conclusions: CAPTEM/PRRT is active, meeting its target landmark PFS for CAPTEM/PRRT (12mo pNETs; 15mo mNETs) with numerically greater OTRR in both pNETs and mNETs, but with more haem toxicity in mNETs. As activity was high in both control arms longer follow up is required to determine if the relative activity of PRRT/CAPTEM is sufficient to warrant Phase III evaluation. Clinical trial information: ACTRN12615000909527 .

Real-world survival outcomes of using neoadjuvant chemotherapy in pancreatic cancer patients: Findings from the PURPLE clinical registry

e16755

Background: Only a minority of pancreatic cancer patients (pts) are surgical candidates at presentation. Neoadjuvant chemotherapy (NAT) is proposed to increase the proportion of surgical candidates. This study investigates the impact of NAT in routine care of pancreatic cancer. Three cohorts were analysed, patients with early-stage resectable (ER), borderline resectable (BR) and locally advanced (LA) pancreatic cancer. Within these groups, survival outcomes of those undergoing immediate resection (IR) was compared to those receiving NAT with nab-paclitaxel and Gemcitabine (nabPGem) and NAT with FOLFIRINOX. Methods: The PURPLE registry consists of 1492 pancreatic cancer pts from 27 hospitals in Australia, New-Zealand and Singapore, collated between 2016-2019. After exclusion of LA unresectable and metastatic pts (n = 809), 683 pts were included. Kaplan-Meir curves estimated survival between groups with 95% confidence intervals. Multivariable cox proportional hazards models adjusted for age, gender and ECOG performance status. Results: Of 683 pts, 107 received NAT and 576 underwent IR. Those in the NAT group had favourable characteristics, including younger age (mean 63 vs. 66 yrs, p < 0.01) and higher proportion of ECOG 0 vs. ≥1 (64% vs 46%) than those undergoing IR. Of those that received NAT, 64 received FOLFIRINOX and 35 nabPGem. Those receiving FOLFIRINOX were younger (mean: 60 vs. 67 yrs, p < 0.01) and were more likely ECOG 0 compared to those receiving nabPGem (72% vs. 46%, p = 0.02). Resection rates for pts undergoing IR vs. NAT were 88% vs. 50% in ER and 16% vs. 43% in BR. Rates of R0 resection margins in pts undergoing IR vs. NAT were 54% vs. 25% in ER and 6% vs. 21% in BT. Comparing ER to BR, mOS was 29.9 vs. 20.3 mths (HR: 0.54, p < 0.01). Within BR, mOS was 20.3 vs. 17.2 mths for NAT vs. IR (HR: 1.11, p = 0.74). Comparing those receiving FOLFIRINOX vs. nabPGem over all groups, mOS was 22 mths vs. 12 mths (HR: 0.31, p < 0.01). Conclusions: Real-world data confirms the use of NAT remains infrequent in this Asia-Pacific population. The use of FOLFIRINOX was associated with better survival than nabPGem based on this observational study. Improved methods for treatment selection are required. Potential biomarkers including circulating tumor DNA are being explored in the DYNAMIC-Pancreas clinical trial.

Pembrolizumab monotherapy for patients with advanced MSI-H colorectal cancer: Longer-term follow-up of the phase II, KEYNOTE-164 study

4032

Background: Pembrolizumab provides effective antitumor immunity and durable responses in patients (pts) with advanced, colorectal cancer (CRC) with microsatellite instability-high (MSI-H) tumors. We present data on antitumor immunity with pembrolizumab in pts from the phase 2, KEYNOTE-164 study who had approximately 3 years of follow-up, and in pts re-treated after disease progression. Methods: KEYNOTE-164 enrolled pts with metastatic MSI-H CRC, MSI-H status confirmed locally by IHC or PCR, and ≥2 (cohort A) or ≥1 (cohort B) prior lines of therapy (fluoropyrimidine, oxaliplatin, irinotecan, or anti VEGF/EGFR). Eligible pts received pembrolizumab 200 mg Q3W for 2y (35 administrations) or until progression, unacceptable toxicity, or withdrawal. Pts who stopped pembro due to a confirmed CR or after completing 2y of treatment and who progressed after stopping were eligible for re-treatment with up to 17 administrations in the second-course phase, at investigator discretion. Tumor response was assessed Q9W per RECIST v1.1 by independent review. The primary endpoint was ORR. Secondary endpoints included DOR, PFS, OS, and safety. The data cutoff date was Sep 9, 2019. Results: At data cutoff, the median follow-up was 31.4 mo (range, 0.2-47.8) for 61 pts in cohort A and 36.1 mo (0.1-39.3) for 63 pts in cohort B. ORR was 32.8% (3CR, 17PR; 95% CI% 21.3-46.0) for cohort A and 34.9% (8CR, 14PR; 95% CI 23.3-48.0) in cohort B. Median DOR was not reached (NR [range, 6.2-41.3+]) and not reached (range, 3.9+ to 37.1+), respectively. Fifteen pts in cohort A and 17 in cohort B had ongoing responses at data cutoff. Median PFS was 2.3 mo (95% CI 2.1-8.1) with 3-yr PFS rate of 31% in cohort A and was 4.1 mo (2.1-18.9) with 3-yr PFS rate of 34% in cohort B. Median OS was 31.4 mo (21.4-NR) with 3-yr OS rate of 49% in cohort A and was not reached (19.2-NR) with 3-yr OS rate of 52% in cohort B. Nine pts (6 in cohort A, 3 in cohort B) had a second course of treatment. The best response in second course was PR in 1 patient each in cohort A and B. Grade 3-4 drug-related adverse events occurred in 10 (16%) pts in cohort A and 8 (13%) pts in cohort B. No grade 5 drug-related events occurred. Conclusions: After approximately 3 y of follow-up, pembrolizumab continues to provide effective long-term antitumor immunity with durable responses, with small numbers of drug-related adverse events and no drug-related deaths in pts with advanced, MSI-H CRC. Clinical trial information: NCT02460198 .

Pretherapeutic 68Ga-DOTATATE PET SUV predictors of survival of radionuclide therapy for metastatic neuroendocrine tumors

4609

Background: Peptide receptor radionuclide therapy (PRRT) is an effective treatment option in patients with advanced neuroendocrine tumours (NETs). Patients are pre-selected based on 68Gallium-DOTA-(0-Tyr3)-octreotate Positron Emission Tomography (68Ga-DOTATATE PET) uptake. The level of uptake in tumour on the baseline 68Ga-DOTATATE PET scan has been explored as a predictor of response in NETs with inconclusive evidence. The aim of this study is to determine the correlation between 68Ga-DOTATATE PET SUV parameters to survival outcomes. Methods: We retrospectively analysed 142 lesions (up to five lesions per patient) in 73 patients with NET undergoing PRRT with 177Lutetium octreotate (8.0-8.3GBq) and pretherapeutic 68Ga-DOTATATE PET/CT in a single institution. Standardised uptake values (SUVs) max and mean were correlated with progression-free survival (PFS) and overall survival (OS). Results: A total of 73 patients were included in the analysis. The median age was 63 (28-89) years. NET origin was gastroenteric (49%), pancreatic [pNET] (38%), bronchial (10%) and other (3%). Ki-67 proliferation index ( < 3%: 36%, 3-20%: 36%, > 20%- < 50%: 8%, unknown: 21%) was seen. Pretherapeutic SUV max but not SUV mean was higher in pNETs (P = 0.04). No difference was seen with Ki-67 index. The median PFS was 32 (95%CI: 26-38) months. Median PFS was reduced with increasing ECOG performance status [PS] (P= 0.029), increasing tumour grade (P = 0.003), increasing Ki-67 proliferation index (P = 0.013), reduced SUV max (P= 0.003), reduced SUV mean (P= 0.001). Multivariate analysis confirmed SUV mean (HR =-1.71 [95%CI: -2.66- -0.80]; P<0.01) and Ki-67 index (HR = 1.11 [1.06-1.17]; P < 0.01) as maintaining significance when incorporating ECOG PS (HR = 1.96 [0.68-5.47]; P = 0.22). The mean OS was 40 [37-44] months. A higher SUV max (SUV max < 30: 34 [30-40] months vs SUV max > 30: 48 [44-51]; P<0.01) and higher SUV mean (SUV mean < 20: 33 [28-39] months vs SUV mean > 20: 47 [43-51]; P<0.01) were associated with improved mean OS. Mean OS was not affected by ECOG performance status (P = 0.896), primary site of origin (P = 0.567) and Ki-67 index (P = 0.110). Conclusions: 68Ga-DOTATATE PET SUV measures correlated with an improved PFS on multivariate analysis as well as improved OS in this select group of patients suitable for PRRT. Those patients with lower SUV mean may benefit from escalation of therapy such as increasing administered therapeutic activity.

PET Imaging Quantifying 68Ga-PSMA-11 Uptake in Metastatic Colorectal Cancer

At diagnosis, 22% of colorectal cancer (CRC) patients have metastases, and 50% later develop metastasis. Peptide receptor radionuclide therapy (PRRT), such as ¹⁷⁷Lu-PSMA-617, is used to treat metastatic prostate cancer. ¹⁷⁷Lu-PSMA-617 targets prostate-specific membrane antigen (PSMA), a cell-surface protein enriched in prostate cancer and the neovasculature of other solid tumors, including CRC. We performed ⁶⁸Ga-PSMA-11 PET/CT imaging of 10 patients with metastatic CRC to assess metastasis avidity. Eight patients had lesions lacking avidity, and 2 had solitary metastases exhibiting very low avidity. Despite expression of PSMA in CRC neovasculature, none of the patients exhibited tumor avidity sufficient to be considered for ¹⁷⁷Lu-PSMA-617 PRRT.

First results for Australasian Gastrointestinal Trials Group (AGITG) control net study: Phase II study of 177Lu-octreotate peptide receptor radionuclide therapy (LuTate PRRT) +/- capecitabine, temozolomide (CAPTEM) for midgut neuroendocrine tumors (mNETs)

604

Background: Single agent 177Lu-octreotate peptide receptor radionuclide therapy is now a standard of care for progressive mNETS. High activity was seen with LuTate and concurrent CAPTEM chemotherapy in a single arm Phase I/II trial. This study was undertaken to determine the relative activity of adding CAPTEM to LuTate PRRT in patients with mNETs. Methods: Non-comparative randomised open label phase II trial of PRRT +/- CAPTEM in patients with mNETs, with 2:1 randomisation: PRRT /CAPTEM (experimental arm) vs. PRRT (control). PRRT /CAPTEM: 7.8GBq LuTate day(D) 10, 8 weekly (wkly) x 4, with b.i.d. oral CAP 750mg/m2 D1-14 & TEM 75mg/m2 D10-14, 8 wkly x 4, vs. PRRT 8 wkly x 4. Primary endpoint: progression free survival (PFS) at 15 months assuming 15 month PFS of 66.4% in the control arm, aiming for PFS rate > 80%; secondary endpoints: objective tumour response rate (complete or partial response) (OTRR), clinical benefit rate (complete or partial response, stable disease) (CBR), toxicity, and QOL. Results: 47 patients enrolled (Dec 2015 - Feb 2018): 33 PRRT/CAPTEM and 14 PRRT. Two patients withdrew prior to treatment. Patient characteristics were balanced except gender (female 58% vs. 14%). Two patients received 2 prior systemic regimens. After a median follow-up of 32 months, the 15 month PFS was 90% (95% CI: 73-97%) v 92% (95% CI: 57-99%); OTRR 25% vs 15%; and CBR 97% vs 92% for PRRT/CAPTEM v PRRT respectively. For treatment related adverse events 22/32 CAPTEM patients experienced one Grade 3 event (69%) vs 5/13 (38%, PRRT); 4/32 pts experienced one Grade 4 event (13%) v 1/13 (8%) respectively. Only one patient failed to complete therapy due to toxicity (PRRT/CAPTEM). Conclusions: This initial planned analysis demonstrates similarly high 15 month PFS for CAPTEM/PRRT relative to PRRT alone. OTRR is numerically higher but at the cost of greater toxicity. Longer follow up is required to determine if the activity of PRRT/CAPTEM is sufficient to warrant Phase III evaluation. Clinical trial information: ACTRN12615000909527.

MONARCC: A randomized phase II study of panitumumab monotherapy and panitumumab (pan) plus 5 Fluorouracil (FU) as first-line therapy for RAS and BRAF wild type metastatic colorectal cancer (mCRC): An AGITG clinical trial

TPS271

Background: Pan added to combination chemotherapy is established first-line therapy for RAS and BRAF wild type mCRC. Elderly patients are not well represented in clinical trials and may be more suited to treatment protocols with lower toxicity risks; FU plus bevacizumab (bev) is commonly used. Treatment related efficacy, toxicity, impact on quality of life and other outcomes of pan based regimens in an elderly population have not been well studied. Methods: A prospective non-comparative randomized phase 2 study. Australian Clinical Trials Registry Number: ACTRN 12618000233224. Main inclusion criteria include: Untreated patients aged 70 years or older; RAS and BRAF wild type; ECOG performance 0-2. Randomisation 1:1, stratified by primary tumour side, performance status, number of metastatic sites; to pan 6mg/kg 2 weekly or panitumumab plus FU 400 mg/m² bolus; leucovorin 200mg/m²; FU 2400mg/m² 48 hour infusion 2 weekly. Primary endpoint is 6-month progression-free survival (PFS). Sample size is 80 patients based on expected 6-month PFS rate of 73% with FU and bev. Using the method of Metha-Cain, if 24 or more patients are progression free at 6 months, the one sided 95% confidence interval includes 73% and we declare similar activity. Secondary endpoints include overall survival; toxicity and overall treatment utility, a composite measure of treatment benefit based on radiology, clinical progress, toxicity and patient reflection of the impact of treatment on their daily lives. Patients undergo a comprehensive health assessment at baseline and limited health assessment at 4 months. Physical activity trackers are worn for 2 weeks at treatment commencement and again at week 16. Tumour tissue and blood samples (at baseline, cycle 3 day 1 and at 24 weeks) will be collected for translational research. First site opened in June 2018. Twelve patients have been recruited to date from 9 sites in Australasia. Eighteen sites were open as at September 2019. Clinical trial information: 12618000233224.

A pooled analysis of multicenter cohort studies of post-surgery circulating tumor DNA (ctDNA) in early stage colorectal cancer (CRC)

3518

Background: Studies in multiple tumor types have demonstrated the prognostic impact of ctDNA analysis after curative intent surgery. Emerging data suggest that conversion of ctDNA from detectable to undetectable after adjuvant chemotherapy may reflect treatment efficacy. Further review of existing study data could increase the precision of ctDNA guided adjuvant therapy decision making. Methods: We combined individual patient (pt) data from three independent prospective cohort studies that enrolled 485 pts with stage II or III CRC. Clinicians were blinded to ctDNA results. We evaluated pt outcomes over a 5-year follow-up period (median, 47.2 months). Plasma samples were collected 4 to 10 weeks after surgery. Mutations in ctDNA were assayed using Safe-SeqS. Results: ctDNA was detected after surgery in 59 (12%) pts overall (11.0%, 12.5% and 13.8% respectively for samples taken at 4-6, 6-8 and 8-10 weeks; P = 0.740). ctDNA detection was associated with nodal status; 8.7%, 16.7% and 32.4% in N0, N1 and N2 disease (P < 0.001), but remained an independent adverse prognostic factor in multivariable analysis. ctDNA detection was associated with poor overall survival for pts treated (mortality ratio, 3.0; P = 0.026) or not treated with adjuvant chemotherapy (mortality ratio, 5.17; P < 0.001). The median MAF (mutant allele frequency) in pts with detectable ctDNA was 0.046%. For pts not treated with adjuvant chemotherapy, 3 year recurrence free survival (RFS) was 9% in pts with a MAF > 0.046% vs 33% with a MAF ≤ 0.046% (HR, 2.7; P = 0.032). For chemotherapy treated pts, 3 year RFS was 25% in pts with a MAF > 0.046% vs 70% with a MAF ≤ 0.046 (HR, 3.1; P = 0.025). In 90 pts with recurrence, ctDNA had been detected post surgery in 3 of 20 (15%) with locoregional recurrence, 27 of 60 (45%) with distant recurrence and 5 of 10 (50%) with both (P = 0.044). Conclusions: Where samples for ctDNA analysis were collected 4 to 10 weeks post surgery, sampling timing may not significantly impact detection rates. The prognostic significance of ctDNA detection can be further stratified by MAF level, but MAF level may not impact adjuvant treatment benefit. ctDNA analysis is most sensitive for detecting minimal residual disease at distant sites.

Metastasectomy and BRAF mutation: An analysis of survival outcome in metastatic colorectal cancer

3531

Background: Surgical resection of oligometastases improves survival in metastatic colorectal cancer (mCRC). It is unclear whether such benefit is consistently observed for BRAF V600E mutant (MT) and wild type (WT) mCRC. We conducted a retrospective analysis to explore the influence of BRAF mutation status on survival outcomes after metastasectomy. Methods: Data collected from two large prospective population databases in Australia (Treatment of Recurrent and Advanced Colorectal Cancer (TRACC) and South Australian cancer registry). Overall survival (OS) and recurrence free survival (RFS) for BRAF MT and WT mCRC were evaluated by Kaplan-Meier method and compared by log-rank test. Results: 513 patients who had undergone metastasectomy were identified, 6% were BRAF MT. Median age 63. Metastasectomy rate was lower in BRAF MT (13 v 27%). In BRAF WT, 4% underwent resection of metastases (mets) in >1 organ at diagnosis and 5% had 3 or 4 metastasectomies versus none in BRAF MT. Median OS in BRAF MT v WT: 25.7 v 48.5 months (HR 1.95; 1.18-3.22). In a multivariate model adjusting for variables which were significant on univariate analysis, OS differences were not statistically significant. Right primary tumor, intact primary, >1 metastatic sites at diagnosis, non R0 resection, peritoneal mets and synchronous mets were independent predictors of worse OS. Among 364 patients with RFS data there was no difference between BRAF MT and WT (16 v 19 months, p=0.09). Rate of downsizing was higher with triplet chemo than doublet +/- bevacizumab or doublet/EGFR in BRAF WT (50 v 30%) as well as MT (33 v 11%). Conclusions: Median OS was > 2 years in BRAF MT V600E after metastasectomy in this study consistent with an OS benefit. OS did not differ after metastasectomy between BRAF MT and WT in a multivariate model. Presence of BRAF MT should not impact patient selection for metastasectomy.[Table: see text]

Answering real-world questions using real-world data: Understanding dynamic treatment decisions and outcomes in metastatic colorectal cancer (mCRC)

e18061

Background: The wide range of possible combinations and sequences available for mCRC treatment presents a major challenge to clinicians, who need to determine the optimal approach for an individual patient or patient subset. In the absence of clinical trial evidence, real world data are an increasingly valuable resource that can be utilized not only to understand treatment patterns and outcomes in routine practice, but also to define an optimal treatment strategy for individual patients across multiple lines of therapy. Methods: Real world data from an Australian mCRC registry were used to develop an interactive data visualization tool that displays treatment variation, customizable to different levels of detail and specific patient subsets, based on patient and disease characteristics. Next, a discrete event simulation model was developed to predict progression-free (PFS) and overall survival (OS) for first line palliative treatment with doublet chemotherapy alone or with bevacizumab, based on data of 867 patients that were treated accordingly. Results: Of 2694 Australian patients enrolled, 2057 (76%) started 1st line treatment with chemotherapy and/or a biologic agent, 1087 (40%) and 428 (16%) received 2nd and 3rd line therapy, respectively. Combined, these 3 lines of treatment accounted for 733 unique sequences. After recoding treatment to the most intensive chemotherapy and the first exposed biologic, 472 unique sequences remained. In exploratory analyses, the simulation model estimated that median 1st line PFS (95% CI) of 219 (25%) patients could be improved from 175 (156, 199) to 269 days (247, 293) by targeting a different treatment. Conclusions: This was an initial exploration of the potential for data visualization and simulation modeling to inform understanding of practice in mCRC and to guide clinical decision making. Such tools allow clinicians and health system providers to define variation in practice patterns and to identify opportunities to improve care and outcomes. Ultimately, the aim is to improve the delivery of personalized cancer care, where other applications such as conditional survival and cost-effectiveness analyses may be useful.

Left- versus right-side metastatic colorectal cancer: Teasing out clinicopathologic drivers of disparity in survival

623

Background: Patients with metastatic colorectal cancer (mCRC) arising from a right sided colorectal cancer (RC) have an inferior survival versus patients with a left sided colorectal cancer (LC). Previous analyses have suggested that multiple factors contribute. Methods: The Treatment of Recurrent and Advanced Colorectal Cancer (TRACC) Registry has prospectively captured comprehensive data on consecutive mCRC patients since 2009. For this analysis LC were defined as primary tumors at or distal to the splenic flexure and included rectal cancers. We analysed patient, tumor, treatment and outcome data. Results: Of 2306 patients enrolled from July 2009–March 2018, median age 67 years (range 18–100), 1559 (67.6%) had a LC. Patients with a RC were older (median 71 vs 65 years, p < 0.001), more likely to be female (50.2% vs 36.9%, p < 0.001) and to have a Charlson score ≥ 3 (67.5% vs 54.6%, p < 0.001). More common in RC, where tested, were BRAF mutations (22.3% vs 7.2%, p < 0.001) and deficient mismatch repair (6.4% vs 1.5%, p < 0.001), but not RAS mutations. Peritoneal metastases were more frequent in RC (29.7% vs 14.9%, p < 0.001). RC patients were more likely to have their primary tumor resected (77.9% vs 70.6%, p < 0.001) and to receive less first-line chemotherapy (72.2% vs 78.7%, p = 0.001). Progression-free survival on first-line systemic therapy was inferior for RC patients (8.1 vs 10.8 months, HR 1.38, p < 0.001). Median overall survival (OS) for all RC patients was inferior (19.6 vs 27.5 months, HR 1.44, p < 0.001). Primary side remained a significant factor for OS in multivariate analysis. Conclusions: Our data confirms the poor prognosis associated with RC, contributed to by multiple factors. However, primary side remains significant even when adjusting for these, indicating the existence of further as yet undefined differences. [Table: see text]

RNF43 is mutated less frequently in Lynch Syndrome compared with sporadic microsatellite unstable colorectal cancers

The WNT signaling pathway is commonly altered during colorectal cancer development. The E3 ubiquitin ligase, RNF43, negatively regulates the WNT signal through increased ubiquitination and subsequent degradation of the Frizzled receptor. RNF43 has recently been reported to harbor frequent truncating frameshift mutations in sporadic microsatellite unstable (MSI) colorectal cancers. This study assesses the relative frequency of RNF43 mutations in hereditary colorectal cancers arising in the setting of Lynch syndrome. The entire coding region of RNF43 was Sanger sequenced in 24 colorectal cancers from 23 patients who either (i) carried a germline mutation in one of the DNA mismatch repair genes (MLH1, MSH6, MSH2, PMS2), or (ii) showed immunohistochemical loss of expression of one or more of the DNA mismatch repair proteins, was BRAF wild type at V600E, were under 60 years of age at diagnosis, and demonstrated no promoter region methylation for MLH1 in tumor DNA. A validation cohort of 44 colorectal cancers from mismatch repair germline mutation carriers from the Australasian Colorectal Cancer Family Registry (ACCFR) were sequenced for the most common truncating mutation hotspots (X117 and X659). RNF43 mutations were found in 9 of 24 (37.5%) Lynch syndrome colorectal cancers. The majority of mutations were frameshift deletions in the G659 G7 repeat tract (29%); 2 cancers (2/24, 8%) from the one patient harbored frameshift mutations at codon R117 (C6 repeat tract) within exon 3. In the ACCFR validation cohort, RNF43 hotspot mutations were identified in 19/44 (43.2%) of samples, which was not significantly different to the initial series. The proportion of mutant RNF43 in Lynch syndrome related colorectal cancers is significantly lower than the previously reported mutation rate found in sporadic MSI colorectal cancers. These findings identify further genetic differences between sporadic and hereditary colorectal cancers. This may be because Lynch Syndrome cancers commonly arise in colorectal adenomas already bearing the APC mutation, whereas sporadic microsatellite unstable colorectal cancers arise from serrated polyps typically lacking APC mutation, decreasing the selection pressure on other WNT signaling related loci in Lynch syndrome.

The prognostic role of inflammatory markers in patients with metastatic colorectal cancer treated with bevacizumab

719

Background: Bevacizumab (BEV) with fluoropyrimidine-containing chemotherapy is well-established for patients (pts) with mCRC. There is a need for reproducible, validated, inexpensive and accessible prognostic markers to aid in treatment selection. Pre-treatment levels of systemic inflammatory markers including the neutrophil to lymphocyte ratio (NLR) have been shown retrospectively to be prognostic in many tumors. This study aimed to prospectively evaluate the relationship between NLR and treatment outcomes in previously untreated mCRC pts receiving BEV-based therapy. Methods: ASCENT (NCT01588990) is an open-label, single arm, phase-IV, multi-center study. Pts received 1st-line BEV+XELOX or mFOLFOX6 in phase A (PhA) with planned continuation of BEV+FOLFIRI beyond 1st progression in phase B (PhB). Primary analysis: association of NLR with progression free survival (PFS). Secondary analyses: overall survival (OS) and safety. Sub-study: safety in pts with intact primary in situ (PIS) and resected primary tumor (RPT). Results: 128 pts started PhA; median age 63.5 years (range: 26-84), 70(55%) females, 51(40%) PIS, 71(56%) and 56(44%) were PS 0 and ≥ 1; respectively. 53(41%) pts entered PhB. The median baseline (b) NLR was 3.2 (range: 1.5-20.4) with 32(25%) pts having a baseline level > 5. The hazard ratio (HR) for PFS by NLR > 5 vs ≤5 was 1.4 (95% CI: 0.9-2.2; p = 1.01). Median PFS was 9.2 months (mo; 95% CI: 7.9-10.8) for PhA and 6.7 mo (95% CI: 3.0-8.2) for PhB. HR for OS based on baseline NLR > 5 vs ≤5 was 1.6 (95% CI: 1.0-2.7; p = 0.052). Based on a multivariate model, the 12-month predicted OS probability for pts with PS 0, no metastatic liver disease and ≤ 3 sites of other metastatic disease, was 87% in pts with bNLR ≤ 5 and 79% in pts with bNLR > 5. Median OS was 25 mo (95% CI: 19.2-29.7) for the full analysis set and 14.9 mo for PhB. Treatment related toxicities were numerically and qualitatively consistent with prior experience. There were 4(3%) instances of GI perforation, of which, 3(6%) occurred in the PIS group. Conclusions: Consistent with previous studies, the trend was towards worse PFS and OS in pts with higher bNLR, but the association was not proven. Clinical trial information: NCT01588990.

Copy number profiles of paired primary and metastatic colorectal cancers

Liver metastasis is the major cause of death following a diagnosis of colorectal cancer (CRC). In this study, we compared the copy number profiles of paired primary and liver metastatic CRC to better understand how the genomic structure of primary CRC differs from the metastasis. Paired primary and metastatic tumors from 16 patients and their adjacent normal tissue samples were analyzed using single nucleotide polymorphism arrays. Genome-wide chromosomal copy number alterations were assessed, with particular attention to 188 genes known to be somatically altered in CRC and 24 genes that are clinically actionable in CRC. These data were analyzed with respect to the timing of primary and metastatic tissue resection and with exposure to chemotherapy. The genomic differences between the tumor and paired metastases revealed an average copy number discordance of 22.0%. The pairs of tumor samples collected prior to treatment revealed significantly higher copy number differences compared to post-therapy liver metastases (P = 0.014). Loss of heterozygosity acquired in liver metastases was significantly higher in previously treated liver metastasis samples compared to treatment naive liver metastasis samples (P = 0.003). Amplification of the clinically actionable genes ERBB2, FGFR1, PIK3CA or CDK8 was observed in the metastatic tissue of 4 patients but not in the paired primary CRC. These examples highlight the intra-patient genomic discrepancies that can occur between metastases and the primary tumors from which they arose. We propose that precision medicine strategies may therefore identify different actionable targets in metastatic tissue, compared to primary tumors, due to substantial genomic differences.

The potential of circulating tumor DNA (ctDNA) to guide adjuvant chemotherapy decision making in locally advanced rectal cancer (LARC)

3521

Background: The optimal approach to adjuvant chemotherapy for rectal cancer is keenly debated. Routine practice and clinical guidelines vary widely. After pre-operative chemoradiation (CRT), a pathologic complete response (pCR) or nodal involvement (pN+) are prognostic markers that can guide clinical decision-making, but markers that better define the patients (pts) that are likely or unlikely to benefit from chemotherapy are urgently needed. We investigated the potential role of ctDNA as a biomarker to guide therapy. Methods: We conducted a prospective, multi-centre study in pts with LARC (T3/T4 and/or N+) planned for CRT and curative resection. Serial plasma samples were collected pre-CRT, post-CRT, and 4-10 weeks after surgery. Somatic mutations in individual pts’ tumor were identified via sequencing of 15 genes commonly mutated in colorectal cancers. We then designed personalized assays to quantify ctDNA in plasma samples. Pts received adjuvant therapy at clinician discretion. Results: 200 pts were enrolled between Apr-2012 and Dec-2015. Median age was 62 years (range 28-86), 67% were male and 159 pts had pre-CRT and post-op ctDNA samples available for analysis. Of these, 122 (77%) pts had detectable ctDNA prior to therapy. After surgery, 19 pts had detectable ctDNA and 11 of these 19 (58%) have recurred during a median follow up of 22 months. Recurrence occurred in only 12 of 140 (8.6%) with negative ctDNA (HR 12, p < 0.001). One hundred and two (64%) pts received adjuvant chemotherapy. Post-op ctDNA detection was predictive of recurrence irrespective of adjuvant chemotherapy (chemo: HR 10, p < 0.001; no chemo: HR 16, p < 0.001). Thirty-four pts (21%) achieved a pCR, 43 (27%) had pN+ disease. pCR (vs non-pCR) was associated with a trend for lower recurrence risk (HR 0.31, p = 0.089) and pN+ (vs pN0) with a higher recurrence risk (HR 4.2, p < 0.001). ctDNA detection remained predictive of recurrence among pts with a pCR (HR 14, p = 0.014) or with pN+ disease (HR 11, p < 0.001). Conclusions: Post-op ctDNA analysis stratifies pts with LARC into very high and low risk groups. ctDNA analysis remains strongly predictive of recurrence among pts with both lower risk (pCR) and higher risk (pN+) disease.

Bevacizumab and its impact on survival for patients receiving subsequent anti-EGFR therapy: Updated results from the SA metastatic CRC registry

3569

Background: Debate exists as to whether first line bevacizumab effects subsequent sensitivity to anti-EGFR therapy. Authors hypothesize that initial anti-VEGF therapy may induce biological changes that then increase the risk of acquired resistance to subsequent EGFR inhibitors. Methods: A retrospective cohort study was performed to compare the characteristics and survival of patients who were treated with an anti-EGFR therapy 2nd line and beyond by two groups defined by the first line therapy; 1. chemotherapy (chemo) plus bevacizumab (bev) and 2. chemo alone. Survival for this analysis is from the time of commencing first line chemotherapy and secondly from anti-EGFR therapy. Pearson chi test analysis was performed to determine whether receiving first line bev was associated with worse overall survival (OS). Results: 348 mCRC patients who received chemo with or without bev and then an anti-EGFR therapy were studied. Patient characteristics are summarised in the table below. The significant differences between group 1. Vs. 2. were as follows; median age 63.8 years v 67.9 years (p = 0.005), lower use of single agent FU 6.4% v 19.2%, KRAS status not tested (reflecting the practice changes over time) 19.3% v 39.2%, KRAS MT 2% v 4%, and where BRAF MT status was known (11%); BRAF MT rate 23% v 0. Median OS for the 2 groups was 34.2 months, and 28.2 months respectively (p = 0.12) from first line therapy. Median OS for patients who underwent single agent anti-EGFR as subsequent therapy was also not significantly different, 31.1 months group 1 (n = 60) versus 27.7 months group 2 (n = 85), p = 0.52. Results based on commencement of anti-EGFR therapy are under way. Conclusions: Survival was not significantly different between the two groups, and the trend was towards higher OS with chemo plus bev suggesting that in our registry population, bev administration in first line therapy with chemo did not lead to a worse outcome overall for those patients subsequently receiving anti-EGFR therapy, either with chemotherapy or as a single agent. Updated results from commencement of anti-EGFR therapy will give further insights and will be presented at the meeting.

The utility of genomics and functional imaging to predict irinotecan pharmacokinetics and pharmacodynamics: The PREDICT IR study

2564

Background: BSA-based dosing of Irinotecan (IR), does not account for its pharmacokinetic (PK) and pharmacodynamic (PD) variability. Given IR’s unique metabolism, functional hepatic nuclear imaging (HNI) with probes for hepatic transporters correlated with its PK. This study further evaluated the utility of HNI combined with extensive excretory/metabolic/PD pharmacogenomics (PG) to predict IR PK and PD in patients (pts) treated with FOLFIRI to enable dose individualization. Methods: Eligible pts had advanced colorectal cancer, suitable for 1st/2nd-line FOLFIRI± Bevacizumab. Pts had blood analyzed by Affymetrix DMET™ Plus Array and additional SNPs were genotyped. For HNI, pts were given IV 250MBq 99mTc-IDA and imaging data analyzed for hepatic extraction/excretion parameters (clearance [CL], 1hour retention [1hRET], deconvulutional CL [DeCL], hepatic extraction fraction [HEF]). Pts treated with chemotherapy, q2-weekly, and restaged after 4 cycles. Blood taken for IR and metabolite (SN38, SN38G) analysis on day 1 cycle 1, PK parameters derived by non-compartmental analysis. Statistical correlations were evaluated between (i) IDA HNI and (2) PGs, with IR PK, toxicity, objective response (ORR) and progression-free survival (PFS). Results: 32 pts analysed, 31 pts completed 4 cycles. (1) PK correlates: (a) HNI CL and 1hRET with SN38 Metabolic CL, (P = 0.04) and (b) HNI DeCL with IR AUC(0-∞) (P = 0.04). (2) Grade 3+ diarrhea (N = 4, 13%) predicted by SN38 AUC(0-∞) and Metabolic CL (P = 0.04), and gene variants for SCL22A2 and -28A3, ABCC2, UGT2B17, CYP2C18 and DPYD (P < 0.05). (3) Grade 3+ neutropenia (N = 9, 28%) predicted by SN38 PK exposure (P < 0.02), HNI CL and 1hRET (P < 0.0001) and variants for SLC7A7-, SLC22A2-, CHST1-, UGT1A1-, -2B7, ABCB1. (4) ORR (N = 6, 20%) predicted by Methylene tetrahydrofolate reductase (MTHFR) 677C > T (P = 0.002), SN38 exposure (P < 0.003), and variants in metabolic/transporter genes (P < 0.05). (5) PFS by SN38 PK exposure, MTHFR 677C > T, HNI CL, HNI HEF and variants in PK genes (P < 0.05). Conclusions: Hepatic functional imaging with extensive pharmacogenomics correlate with Irinotecan PK and PD enabling the development of nomograms to individualize dosing. Clinical trial information: ACTRN12610000898055.

Potential role of circulating tumor DNA (ctDNA) in the early diagnosis and post-operative management of localised pancreatic cancer

4101

Background: Pancreatic cancer remains a devastating disease, with the diagnosis typically being made late. ctDNA has shown promise as a screening test for various tumor types. The detection of ctDNA post curative intent surgery has been associated with a high risk of recurrence in multiple solid tumors. We explored the potential of ctDNA to improve pancreatic cancer outcomes. Methods: Data from separate US and Australian series were combined. Plasma samples were collected prior to surgery in both studies and post-operative samples were collected in Australia from cases undergoing curative intent surgery. Clinicians were blinded to ctDNA results and adjuvant therapy was at clinician discretion. Tissue samples from both series were analyzed at Johns Hopkins University. Next generation sequencing was used to search for somatic KRAS mutations in the primary tumors and in cell-free DNA in the plasma. Clinico-pathologic, treatment and outcome data were collected. Results: 119 pts had a ctDNA sample at diagnosis (median age 67 years, 56.3% male). Sixty six pts (55.5 %) had detectable ctDNA, including 3/7 (42.9%) with stage I disease, 54/99 (54.5%) with stage II disease, 4/8 (50%) with stage III disease and 5/5 (100%) with metastases. Specific codon 12 KRAS (G12D, G12V or G12R) mutations were identified in the tumor tissue of 12/16 (75%) patients who had a ctDNA sample collected post-surgery. At a median follow-up of 15.2 months, 7/12 (58.3%) pts had recurred, including 3/8 (37.5%) with no detectable ctDNA and 4/4 (100%) with detectable ctDNA post-surgery (HR 4.9, p = 0.04). Detectable ctDNA post-surgery was significantly associated with poor overall survival (HR 6.93, p = 0.006), with a median of 8 months for pts with detectable ctDNA. Conclusions: ctDNA shows promise as a pancreatic cancer screening test, being detectable in a high proportion of pts with early stage disease. The detection of ctDNA post operatively predicts a very high risk of recurrence. The clinical utility of ctDNA to guide adjuvant therapy decision making, and its potential as a real-time marker of treatment effect, are being explored in further studies. Clinical trial information: ACTRN12612000763842.

Impact of clinical and molecular features on risk of recurrence following curative intent resection of metastases in metastatic colorectal cancer

785

Background: Resection of metastases with curative intent is an integral component of mCRC management. However, relapse rates are high and identifying patients most likely to benefit from this approach is of considerable importance. Among patients with mCRC, mutations (mt) in RAS and BRAF genes portent a worse prognosis. Our hypothesis, therefore, is that patients harbouring these mutations may have a higher relapse rate after resection of metastases. We also wished to analyse clinical predictors of relapse, including site of metastases. Methods: We interrogated the TRACC database of patients undergoing resection with curative intent who had mutation status available. The frequency of RAS and BRAF mt was established and their association with clinical parameters determined. Relapse free (RFS) and overall survival (OS), from the date of resection, was estimated for the mt and wild type (wt) groups using the Kaplan Meier method. Multivariate analysis is planned to investigate factors associated with RFS, including stage of the primary tumour, synchronous metastases, site and number of metastases, CEA, peri-operative chemotherapy use, site of the primary (left v right) and RAS and BRAF mutation status. Results: 188 patients were identified. 89 were KRAS/BRAF wt, 92 KRAS mt and 7 BRAF mt. 40% had presented with metastatic disease and 27% had a right sided primary. 76%, 22% and 2% underwent resection of liver, lung or both metastases. Microscopic resection margin was involved in 6%. Resection was performed prior to any chemotherapy in 48%. No difference was seen in relapse free or overall survival between the mt and wt groups. Conclusions: We found no difference in relapse free or overall survival by mutation subgroup suggesting this should not influence suitability for curative intent resection, but analyses is planned on a much larger cohort once data is available. A multivariate analysis, adjusting for important prognostic variables, is planned.

Bevacizumab and its impact on survival for patients receiving subsequent anti-EGFR therapy: Results from the South Australian metastatic colorectal cancer registry

754

Background: Questions over the impact of 1st line bevacizumab on subsequent sensitivity to anti-EGFR therapy have been raised with authors hypothesizing that up-front anti-VEGF agent’s use influences biological changes that increase the risk of acquired resistance to subsequent EGFR inhibitors. Methods: Aretrospective cohort study was performed to compare the characteristics and survival of patients who were treated with an anti-EGFR therapy 2nd line and beyond by two groups defined by the 1st line therapy; 1. chemotherapy (chemo) plus bevacizumab (bev) and 2. chemo alone. Pearson chi test analysis was performed to determine whether receiving 1st line bev was associated with worse OS. Results: 348 mCRC patients who received chemo with or without bev and then an anti-EGFR therapy were studied. Patient characteristics are summarised in table. The significant differences between group 1. Vs. 2. were as follows; median age, lower use of single agent FU, KRAS status not tested, and where BRAF MT status was known (11%); BRAF MT rate. Median OS for the 2 groups was 34.2 mths, and 28.2 mths respectively (p = 0.12). Conclusions: Survival was not significantly different between the two groups, and the trend was towards higher OS with chemo plus bev suggesting that in our registry population, bev administration in first line therapy with chemo did not lead to a worse outcome for those patients subsequently receiving anti-EGFR therapy. [Table: see text]

Left versus right sided colorectal cancer: Teasing out drivers of disparity in outcomes in metastatic disease

682

Background: Patients (pts) with metastatic colorectal cancer (mCRC) arising from a right sided colon cancer (RCC) have an inferior survival versus pts with a left sided primary (LCC). Previous analyses have suggested that multiple factors contribute. Methods: The Treatment of Recurrent and Advanced Colorectal Cancer (TRACC) Registry has prospectively collected data on consecutive mCRC pts at 32 sites (Australia and Hong Kong) from July 2009. LLC were defined as those distal to the splenic flexure. We analysed patient, tumour, treatment and outcome data. Results: Of 1763 patients, median age 67 years (range 18 – 100), 1201 patients (68.1%) had a LCC. RCC were more common in pts ≥ 70 years old (53.9% vs 38.5%, p < 0.01) and females (50.7%, p < 0.01). Where tested, KRAS (50.9% vs 41.5%, p < 0.01) and BRAF (19.1% vs 6.4%, p < 0.01) mutations were more common in RCC. Where tested, dMMR was more common in RCC (11.0% vs 3.2%, p < 0.01). Pts with a RCC more often had peritoneal metastases (25.9% vs 12.5%%, p < 0.01) and were less likely to have liver metastases (69.8% vs 77.9%, p < 0.01) or resection of metachronous metastases (31.8% vs 43.6%, p < 0.01). LCC more commonly received chemotherapy in the 1st (86.1% vs 80.3%, p = 0.02) and 2nd (49.7% vs 43.0%, p = 0.04) line setting. There was no difference in ECOG performance status or co-morbidity for LCC vs RCC. Overall survival for pts with a RCC was 20.6 months vs 27.9 months for LCC L-sided tumours (HR 1.395, p < 0. 0001). Conclusions: Our data confirms that pts with mCRC arising from a RCC have an inferior survival. This appears multifactorial due to associations with adverse clinical and molecular features, metastatic pattern, and less therapy, some of which are related. A multivariate analysis is being performed.

Effectiveness of bevacizumab and cetuximab in metastatic colorectal cancer across selected public hospitals in Queensland

AIM

Metastatic colorectal cancer has a large burden of disease in Australia. Medical therapy is fundamental to extending survival and improving quality of life. The benefits of two costly medicines, bevacizumab and cetuximab, used in Australia remain unclear. The aim of this study was to retrospectively examine the use of these two medicines in metastatic colorectal cancer across five public hospitals in south east Queensland and to compare clinical outcomes to those of published clinical trials.

METHODS

We extracted data from the chemotherapy prescribing database for patients planned for bevacizumab or cetuximab therapy between 2009 and 2013. Median overall survival was estimated using Kaplan-Meier methods.

RESULTS

There were 490 bevacizumab-containing protocols planned and 292 patients received at least one dose of bevacizumab. Median overall survival was 17.2 months (95% confidence interval [CI], 15.4-19.3). Of 208 planned cetuximab-containing protocols, 134 patients received at least one dose of cetuximab. Median overall survival was 9.1 months (95% CI, 7.6-12.0). Thirty-day mortality rates from date of first dose were 0.7% for bevacizumab and 7.5% for cetuximab.

CONCLUSION

Overall survival of patients receiving bevacizumab and cetuximab was consistent with clinical trials, providing some assurance that benefits seen in trials are observed in usual practice. This study provides a methodology of using routinely collected health data for clinical monitoring and research. Because of the high cost of these medicines and the lack of toxicity data in this study, further analysis in the postmarketing setting should be explored.

Effects of a Structured Exercise Program on Physical Activity and Fitness in Colon Cancer Survivors: One Year Feasibility Results from the CHALLENGE Trial

BACKGROUND

There is strong interest in testing lifestyle interventions to improve cancer outcomes; however, the optimal methods for achieving behavior change in large-scale pragmatic trials are unknown. Here, we report the 1-year feasibility results for exercise behavior change in the Canadian Cancer Trials Group CO.21 (CHALLENGE) Trial.

METHODS

Between 2009 and 2014, 273 high-risk stage II and III colon cancer survivors from 42 centers in Canada and Australia were randomized to a structured exercise program (SEP; n = 136) or health education materials (HEM; n = 137). The primary feasibility outcome in a prespecified interim analysis was a difference between randomized groups of ≥5 metabolic equivalent task (MET)-hours/week in self-reported recreational physical activity (PA) after at least 250 participants reached the 1-year follow-up. Secondary outcomes included health-related fitness.

RESULTS

The SEP group reported an increase in recreational PA of 15.6 MET-hours/week compared with 5.1 MET-hours/week in the HEM group [mean difference = +10.5; 95% confidence interval (CI) = +3.1-+17.9; P = 0.002]. The SEP group also improved relative to the HEM group in predicted VO2max (P = 0.068), 6-minute walk (P < 0.001), 30-second chair stand (P < 0.001), 8-foot up-and-go (P = 0.004), and sit-and-reach (P = 0.08).

CONCLUSIONS

The behavior change intervention in the CHALLENGE Trial produced a substantial increase in self-reported recreational PA that met the feasibility criterion for trial continuation, resulted in objective fitness improvements, and is consistent with the amount of PA associated with improved colon cancer outcomes in observational studies.

IMPACT

The CHALLENGE Trial is poised to determine the causal effects of PA on colon cancer outcomes. Cancer Epidemiol Biomarkers Prev; 25(6); 969-77. ©2016 AACR.

Use and impact of bevacizumab in patients undergoing liver resection for metastatic colorectal cancer in routine clinical practice

698

Background: In metastatic colorectal cancer (mCRC) patients with isolated liver metastases, surgical resection offers the greatest likelihood of cure. Whilst for mCRC patients treated with palliative intent the addition of bevacizumab to the chemotherapy backbone is of proven benefit, whether to use bevacizumab in the resectable or potentially resectable population is a clinical dilemma. Methods: Consecutive patients who underwent resection of liver metastases were identified from a prospective Australian mCRC registry that captures comprehensive data on patient and tumor characteristics (including resectability), treatment and outcome. The use of bevacizumab in this setting was examined and the impact on outcomes was explored. Results: From a total mCRC population of 1,700 patients, 543 patients with liver-only mCRC were identified, of which 217 patients (40%) underwent liver resection. Perioperative chemotherapy was administered to 185 patients (85.3%), with bevacizumab added to chemotherapy in 73 (39.5%) patients. There was a trend for bevacizumab treated patients to be younger (median age 60.4 vs 65.1 years, p = 0.07) and fitter (mean Charlson score 2.22 vs 2.64, p = 0.054). Patients that received bevacizumab with perioperative chemotherapy were considerably less likely to have disease regarded as resectable at diagnosis (39 of 73 (53.4%) vs 95 of 112 (84.8%), p =<.01). At 5 years, overall survival was similar for bevacizumab treated and non-treated patients (61.4% vs. 59.2%, HR 0.83, p=0.52). There were no deaths within 30 days of surgery in any patients. Conclusions: Despite limited evidence to support the use of bevacizumab in patients with resectable or potentially resectable liver-only mCRC, clinicians are not infrequently utilising this approach, particularly in younger and fitter patients and those not considered to have resectable disease at diagnosis. The addition of bevacizumab did not appear to impact survival outcomes. A multivariate analysis is underway to better define the impact of bevacizumab on survival outcomes.

Impact of first-line bevacizumab therapy on the efficacy of subsequent anti-epidermal growth factor antibodies (anti-EGFR) in metastatic colorectal cancer (mCRC)

685

Background: The FIRE3 study reported that cetuximab added to first line combination chemotherapy versus the addition of bevacizumab to the same chemotherapy backbone, with a planned switch to the alternative biologic in second line therapy, resulted in improved overall survival. One possible explanation is that the sequence of biologic administration in mCRC impacts on treatment benefit. Examination of the efficacy of anti-EGFR therapy in cohorts where patients have and have not received prior bevacizumab will inform this. Methods: We interrogated our prospective database (ACCORD), which records comprehensive detail on demographics, treatments received and patient outcomes for consecutive mCRC patients in real world practice across multiple institutions in Australia. We included patients receiving 2nd and subsequent line anti-EGFR monotherapy. Disease control rate (DCR) (partial response plus stable disease), progression free survival (PFS) and overall survival (OS) from the start of anti-EGFR were determined as a function of prior bevacizumab use. Results: From a total of 1511 patients treated between 01/10/2002 and 31/12/2014,129 patients treated with single agent anti- EGFR were identified, 75 had received prior bevacizumab. Demographics, performance status and tumour characteristics were similar between the 2 groups. 84% were confirmed KRAS exon 2 wild type. Prior treatment with bevacizumab did not impact DCR (35% vs 40%, p = 0.5065), median PFS (3.0 vs 3.0 months, p = 0.9460) or OS (8.6 vs 8.4 months, HR 0.95, p = 0.8073). Similarly, for bevacizumab treated patients, the time since last dose of bevacizumab to start of anti-EGFR ( < 3 months (n = 26) vs 3-6 months (n = 17) vs > 6 months (n = 32)) did not impact PFS (2.3 vs 3.6 vs 3 months) or OS (8.9 vs 10.1 vs 7.1 months, HR: 0.88, p = 0.82). Conclusions: There was no evident impact of previous bevacizumab treatment, or time from previous bevacizumab treatment, on the efficacy of single agent anti-EGFR therapy in a routine clinical practice cohort. Our data do not support the notion that prior bevacizumab renders tumors more or less sensitive to subsequent anti EGFR therapy.

Effects of a structured exercise program on physical activity and health-related fitness in colon cancer survivors: One year feasibility results from the NCIC CTG CO.21 (CHALLENGE) trial

648

Background: There is strong interest in testing lifestyle interventions to improve cancer outcomes, however, the optimal methods for achieving behaviour change in large scale pragmatic trials are unknown. We report the 1 year feasibility results for behaviour change in NCIC CTG CO.21 Colon Health and Life-Long Exercise Change (CHALLENGE) Trial. Methods: Between 2009 and 2014, 273 high risk stage II and III colon cancer survivors who had recently completed chemotherapy at 50 centers in Canada and Australia were randomized to a structured exercise program (SEP n = 136) or health education materials (HEM n = 137). The primary feasibility outcome in an a priordefined interim analysis was a difference between groups of ≥ 5 metabolic equivalent task (MET) hours/week in self-reported physical activity (PA) after at least 250 participants had reached 1 year. Secondary feasibility outcomes included cardiorespiratory fitness, body weight and circumferences, and objective physical functioning. Results: The SEP group reported a mean increase from baseline to 1 year PA of 15.6 MET hours/week vs a mean increase of 5.1 in the HEM group (+10.5 hrs/wk [95% CI = +3.1 to +17.9]; p = 0.002). SEP also improved relative to HEM in mean differences for predicted VO2max (+2.2ml/kg/min; p = 0.068), 6 minute walk (+29m; p < 0.001), 30 second chair stand (+1.6reps; p < 0.001), 8 foot up and go (-0.4s; p = 0.004), and the sit and reach (+2.1cm; p = 0.08). Conclusions: The behaviour change intervention in the CO.21 Trial produced an increase in self-reported PA that exceeded the feasibility criterion for trial continuation, resulted in objective fitness improvements, and is consistent with the amount of PA associated with improved colorectal cancer outcomes in observational studies. Recruitment to CO.21 remains ongoing. Clinical trial information: NCT00819208.

Impact of anti-VEGF therapy in metastatic colorectal cancer with an intact primary tumour

650

Background: The management of metastatic colorectal cancer (mCRC) with anti-VEGF therapy and a fluoropyrimidine-containing chemotherapy is well established in first and second line therapy. However the role, safety and benefits of anti-VEGF therapy (Bevacizumab) in specific subgroups including those with an intact primary tumour (IPT), are not well described. Methods: Consecutive mCRC pts enrolled from January 2009 were identified from a prospective multi-site Australian registry. Survival and key safety endpoints were explored for pts with an IPT versus pts with a resected primary (RPT). Differences in outcomes were compared using log-rank test and Kaplan-Meier curves. Results: Of 1,204 pts with mCRC, 826 (69%) eligible pts were identified. Anti-VEGF therapy use in the first-line setting was similar in pts with an IPT [200/313 (64%)] vs. a RPT [357/513 (70%)]. In both groups the addition of anti-VEGF therapy was associated with greater PFS; IPT 8.5 vs. 4.7 months (HR 0.73, p=0.017) and RPT 10.8 vs. 5.8 months (HR 0.67, p=0.0002). OS was longer in anti-VEGF treated pts; IPT 20 vs. 14.8 months (HR 0.66, p=0.0046) and RPT 24.4 vs. 17.3 months (HR 0.71, p=0.0039). A trend for more GI perforations with anti-VEGF therapy use was seen in pts with IPT (4.5% vs. 1.8%, p=0.33), but not RPT (1.7% vs. 1.9%). The rate of bleeding and thrombosis was lower in the IPT group treated with bev: 1.5% vs. 5.3%, p=0.05 and 1.5% vs. 2.7%, p=0.47 respectively. In comparison, RPT pts treated with and without anti-VEGF therapy had a bleeding rate of 3.4% vs. 0.6%, p=0.07, and a thrombosis rate of 5.9% vs. 0.6%, p=0.007. Conclusions: In routine clinical practice clinicians appear comfortable using anti-VEGF therapy in pts with an IPT. Superior PFS and OS outcomes were seen with addition of anti-VEGF therapy, with no evident impact of primary tumor status. There was a trend for a higher GI perforation rate in anti-VEGF treated IPT pts, but a lower rate of other adverse events. This may reflect clinicians avoiding anti-VEGF therapy in pts considered at increased risk of these complications. Further analysis is underway to explore this possibility.

A prospective study of the impact of fluorodeoxyglucose positron emission tomography with concurrent non-contrast CT scanning on the management of operable pancreatic and peri-ampullary cancers

BACKGROUND

The role of fluorodeoxyglucose (FDG) positron emission tomography (PET/CT) scanning in operable pancreas cancer is unclear. We, therefore, wanted to investigate the impact of PET/CT on management, by incorporating it into routine work-up.

METHODS

This was a single-institution prospective study. Patients with suspected and potentially operable pancreas, distal bile duct or ampullary carcinomas underwent PET/CT in addition to routine work-up. The frequency that PET/CT changed the treatment plan or prompted other investigations was determined. The distribution of standard uptake values (SUV) among primary tumours, and adjacent to biliary stents was characterised.

RESULTS

Fifty-six patients were recruited. The surgical plan was abandoned in 9 (16%; 95% CI: 6-26) patients as a result of PET/CT identified metastases. In four patients, metastases were missed and seven were inoperable at surgery, not predicted by PET/CT. Unexpected FDG uptake resulted in seven additional investigations, of which two were useful. Among primary pancreatic cancers, a median SUV was 4.9 (range 2-12.1). SUV was highest around the biliary stent in 17 out of 28 cases. PET/CT detected metastases in five patients whose primary pancreatic tumours demonstrated mild to moderate avidity (SUV < 5).

CONCLUSIONS

PET/CT in potentially operable pancreas cancer has limitations. However, as a result of its ability to detect metastases, PET/CT scanning is a useful tool in the selection of such patients for surgery.

Deficient mismatch repair in colorectal cancer: current perspectives on patient management and future directions

Abstract 

Molecular aberrations leading to colorectal cancer are diverse and heterogeneity exists both at a molecular level and in clinical behavior. Defective mismatch repair (dMMR) is a feature of 15% of colorectal cancers. These are hypermutated tumors, mostly right sided and histopathologically elicit a marked immune response. A proportion of these arise due to germline mutation of a mismatch repair gene giving rise to Lynch syndrome, while the majority arise sporadically due to somatic alteration of the MLH1 mismatch repair gene. Although dMMR is associated with an excellent patient prognosis, as tumor stage advances the frequency of dMMR declines and the association with improved prognosis dissipates. It is apparent that dMMR tumors do not represent a unique molecular subset. As the knowledge of the underlying biology evolves, the hope is for individualized therapy that goes well beyond the crude and oversimplified categorization of dMMR versus proficient MMR.

A multicenter, phase II trial of preoperative gemcitabine and nab-paclitaxel for resectable pancreas cancer: The AGITG GAP study

387

Background: Pancreatic cancer (PC) outcomes remain poor, recent series show approximately 37% of patients achieve R0 resection (all margins ≥ 1mm) for resectable disease. Recent data in metastatic disease shows that nab-paclitaxel (nab-PC) and gemcitabine (GEM) chemotherapy (CX) is an active regimen. We aimed to determine the feasibility and R0 resection rate with peri-operative nab-PC and GEM for resectable PC. Methods: A multicentre, single arm, phase 2 trial. Patients with operable PC received 2 cycles of neo-adjuvant GEM 1000mg/m2 and nab-PC 125mg/m2CX on days 1, 8 and 15 of a 28 day cycle followed by surgical resection and then 4 cycles of post-operative CX. The primary endpoint was R0 resection with the aim to assess if a rate of 85% could be achieved. Secondary endpoints included feasibility, toxicity, pathological response (College of American Pathologists Cancer Reporting Protocol for Exocrine Pancreatic Cancer, 2009), recurrence and survival. Results: Forty-one patients were enrolled from 2012-14. Median age was 65 (range 43-79), 41% male, 49% stage I and 51% stage II. Twenty nine (71%) patients underwent resection with evaluable cancer: 1 did not have proven cancer on central pathology review; 5 did not go to surgery (2 disease progression, 2 refused surgery, 1 due to cholangitis); and in 6 patients surgery was abandoned (unresectable disease). No deaths were surgery-related. Median tumor size was 27.5mm (IQR = 25-35). Pre-operative GEM and nab-PC produced an R0 rate of 52% (15/29) (95% CI 34-69%). 15/29 (52%) of resected tumors showed grade 0-2 tumor regression grade. Twenty-nine patients (71%) experienced >1 grade 3/4 toxicities (total 52 events) during pre-operative CX, neutropenia (46%) being most frequent with 8% febrile neutropenia. 39/41 (95%) patients received 2 cycles of pre-operative CX. Conclusions: Peri-operative GEM and nab-PC CX is feasible and well tolerated. The pre-operative regimen was associated with an R0 resection rate comparable or higher than surgical series without pre-operative therapy, although the primary endpoint of 85% could not be met. The value of this regimen will also depend on longer term outcomes related to recurrence and survival. Clinical trial information: 12611000848909.