07/23/2017: Articles Summaries - GI & GU Cancers

View the most clinically relevant journal articles of the week. Updated each Sunday!

Please click on the tumour site to jump to summary



Prognostic Survival Associated With Left-Sided vs Right-Sided Colon Cancer: A Systematic Review and Meta-analysis
Petrelli F et al., JAMA Oncol. 2016 Oct 27.
PubMed ID: 27787550


Studies on colorectal cancer (CRC) have revealed different molecular and biologic characteristics, clinical presentation and prognosis for left-sided CRC (LCC) and right-sided CRC (RCC). In the NO147 trial that randomized patients with metastatic CRC to either cetuximab with FOLFOX vs. FOLFOX, patients with LCC had significantly better DFS compared to RCC (J Clin Oncol. 2013;31(29):3664-3672).
Italian researchers synthesized current evidence with a systematic review and meta-analysis on CRC and the importance of tumor location on prognosis.


  • Objective: determine prognostic role of tumor site in CRC
  • Patients: histologically or cytologically confirmed CRC
  • Study selection: studies that reported CRC tumor side (left or right-sided) as a variable with OS data (hazard ratios [HR] reported from multivariate cox regression analysis) and published in English
  • Endpoint: overall survival (OS) by tumor side


  • 66 studies met criteria and included 1,437,846 patients and a median follow-up of 65 mos.
  • Quality of evidence: NOS scale range 5-9, 75% high quality (NOS score range 7-9)
  • Analysis of OS using random effects model (heterogeneity I2; p<0.001) revealed HR 0.82 (0.79-0.84); p<0.001 with survival benefit in favour of LCC compared to RCC.
  • OS advantage with LCC was observed across subgroups including race, prospective vs. retrospective design, study quality and year of publication.
  • In 20 studies that looked at stage IV patients alone vs. 25 studies on stage 1-III, a similar trend was observed with LCC (HR 0.73 vs. HR 0.84; p<0.001).
  • Meta-regression revealed effect size did not depend on stage and adjuvant treatment. There was no evidence of publication bias on funnel plot and Egger test.


  • This systematic review and meta-analysis of 66 CRC studies over two decades revealed tumor site to be a significant prognostic factor. LCC reduced the risk of death by 18% compared to RCC. This advantage was independent of race, stage, study design and year of publication. The magnitude of survival benefit with LCC was less pronounced in stage 1-III CRC compared to metastatic disease.
  • This difference in prognosis of LCC compared to RCC can likely be explained by evidence that these are distinct in terms of molecular characteristics and tumor biology. RCC is usually associated with defective mismatch repair (dMMR) genes, BRAF and KRAS mutations (World J Gastroenterol. 2015;21(21):6470-6478). In the QUASAR trial where CRC patients were randomized to chemotherapy (5-FU with folinic acid) vs. no chemotherapy, KRAS mutant tumors had slightly higher risk for recurrence compared to KRAS wild-type tumors (27% vs. 21%; p=0.002) [J Clin Oncol. 2011 Apr 1;29(10):1261-70].
  • There is also evidence that the survival benefit of different treatment strategies may vary by tumor location. While surgery appears to confer similar survival benefit for both LCC and RCC, chemotherapy regimens and targeted agents have varying effects based on tumor location. In early stage CRC, RCC is more responsive to FOLFIRI compared to FOLFOX, where the benefit is slightly better than LCC (World J Gastroenterol. 2015;21(21):6470-6478). In advanced CRC the FIRE3 and CALGB/SWOG 80405 trial subgroup analyses showed that RCCs have decreased benefit from anti-EGFR therapy (J Clin Oncol. 2016;34(suppl, abstr 3504); J Clin Oncol. 2014;32:5s (suppl abstr, 3600)).
  • Study limitations include lack of analysis by RAS mutation status as well as treatment(s) received, which would have provided important information regarding therapeutic strategies for RCC compared to LCC. Significant heterogeneity between studies was another limitation, but this was adjusted for using a random effects model and meta-regression..

More GI Cancer Updates



Nivolumab in metastatic urothelial carcinoma after platinum therapy (CheckMate 275): a multicentre, single-arm, phase 2 trial
Sharma P et al, Lancet Oncol. 2017 Mar;18(3):312-322.
PubMed ID: 28131785

This article features an accompanying Editorial written by Dr. Brandon Bernard (Medical Oncologist, University of Colorado Anschutz Medical Campus). 
Click here to view


First line platinum-based chemotherapy is standard of care in metastatic urothelial cancer. In around 20-30%, PD-L1 expression is present in tumor tissue samples. Patients with advanced urothelial cancer who received 2nd line anti-PD-L1 antibody atezolizumab achieved higher overall response rate compared to historical controls (Lancet 2016;387: 1909–20). A phase 1/2 study of anti-PD1 inhibitor nivolumab revealed that 24.4% of patients with metastatic urothelial cancer achieved an objective response (Lancet Oncol 2016; 17: 1590–98). Based on promising data, researchers further investigated the efficacy of nivolumab in the CheckMate 275 trial.


  • Patients: histologically confirmed metastatic or locally advanced, unresectable urothelial cancer, progressed on at least 1 platinum-based treatment for metastatic disease or within 12 mos. of perioperative platinum treatment for muscle-invasive disease, ECOG 0-1, no active brain metastases
  • Study design: phase II single-arm trial
  • Intervention: Nivolumab 3mg/kg IV q2w
  • Endpoints:
    • Primary: ORR in all treated patients, in patients with tumors >/=1% PD-L1 expression, and patients with tumors >/=5% PD-L1 expression
    • Secondary: PFS, OS, investigator-assessed objective response


  • 270 patients enrolled
  • Baseline patient characteristics: 45% age <65 yrs., 78% male, 86% white, 54% ECOG 0, 84% visceral mets, 82% baseline Hgb ≥ 100 g/L, 66% previous platinum regimen (metastatic), 42% at least 1 previous chemotherapy
  • 79% with tissue samples for PD-L1 analysis
  • Treatment discontinuation due to toxicity: 5%

Table 1: Efficacy Outcomes
Table 1

  • Safety
    • Most common any grade adverse event: fatigue (17%)
    • Most common grade 3-4 adverse events: fatigue and diarrhea (2% each)
    • Most common immune-related any grade adverse events: skin (17%) and endocrine (14%)

Table 2: Adverse Events
Table 2

  • QOL
    • QOL scores were stable and health status improved at week 41.
  • Gene expression analysis
    • Patients with higher 25-gene interferon-γ signature were more likely to respond to nivolumb compared to those with lower signature (p=0.0003).
    • Basal 1 subtype had the strongest 25-gene interferon-γ signature, 12-chemokine signature and the highest responders of all subtypes.
    • CXCL9 or CXCL10 expression was highest in basal 1 subtype, nivolumab responders and PD-L1 ≥5%.
    • Nivolumab responders had the highest 12-chemokine signature and CD8 expression


  • Nivolumab yielded response rates that were durable and clinically meaningful in advanced urothelial cancer after disease progression on platinum-based chemotherapy. Higher tumor cell PD-L1 expression appeared to correlate with better response and longer survival.
  • Nivolumab demonstrated tolerable safety profile consistent with previous literature. Adverse events were mostly low grade and resolved with anti-immunomodulatory drugs. Furthermore, QOL improved with time, despite incidence of grade 3-4 adverse events in less than 20% of the study population.
  • Although gene expression analysis revealed better outcomes for responders, high 25-gene interferon-γ signature and basal 1 subtype, the sample size limits its interpretation and caution is advised with respect to analysis of specific treatment effects. Further studies are warranted to confirm these findings.
  • In a phase 2 study of atezolizumab by Rosenberg et al, 15% of patients with advanced urothelial cancer who failed platinum-based regimen achieved ORR (Lancet 2016;387: 1909–20). Higher immune cell PD-L1 expression was associated with higher response rate. However, no association was observed in tumor cell PD-L1 expression.
  • The Pembrolizumab KEYNOTE randomized control trial of Pembrolizumab vs investigator's choice of paclitaxel, vinflunine or docetaxel was positive for OS.  Thus, more positive data is emerging for PD(L)-1 inhibitors in metastatic urothelial cancer.
  • Study limitations include lack of comparator arm, short follow-up (6 mos.), use of archival tissue, lack of immune cell PD-L1 expression analysis and small sample size.).

More Updates in GU Cancer