01/29/2017: Articles Summaries - GI & Breast Cancer

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

Please click on the tumour site to jump to summary



Pazopanib plus best supportive care versus best supportive care alone in advanced gastrointestinal stromal tumours resistant to imatinib and sunitinib (PAZOGIST): a randomised, multicentre, open-label phase 2 trial
Mir O. et al. Lancet Oncol 2016; 17: 632–41
PMID: 27068858


Pazopanib is a multi-targeted TKI that inhibits KIT, VEGFR1, 2, and 3, PDGFRα, and PDGFRβ. In a phase II trial, 48% of patients with advanced GIST who failed imatinib or sunitinib achieved best response with a median PFS of 1.9 mos. and OS of 10.7 mos. (Ann Oncol 2014; 25: 236–40). The PAZOGIST trial compared pazopanib with best supportive care (BSC) vs. BSC alone in advanced GIST.


  • Patients: histologically confirmed unresectable, metastatic, or locally advanced GIST that progressed on or unable to tolerate imatinib or sunitinib, ECOG ≤ 2
  • Study design: open-label, phase II RCT
  • Intervention
    • Experimental: Pazopanib 800 mg. PO in 4 wk. cycles + BSC
    • Control: BSC
  • Outcomes
    • Primary: PFS
    • Secondary: ORR, OS, safety


  • 81 patients randomized to pazopanib + BSC (n=40) and BSC alone (n=41)

Table 1. Baseline patient characteristics (selected)
Table 1

  • 88% in BSC alone group crossed over to pazopanib with median PFS of 3.5 mos. and median OS of 9.8 mos.
  • 1 partial response in BSC alone group

Table 2. Efficacy Outcomes
Table 2

Table 3. Adverse events summary
Table 3

  • Most common AEs with BSC: hypertension (12%), anemia (10%)
  • Pazopanib-related deaths: pulmonary embolism (n=2), hepatic cytolysis (n=1)
  • SAE's in BSC group after crossing over to pazopanib: hyperammonaemic encephalopathy, pneumopathy, respiratory failure

Genetic changes

  • 64% with genetic changes
  • Pazopanib + BSC who received pazopanib at the end of study: KIT exon 11 mutations (n=2), Asp842Val PDGFRA mutation (n=1), KIT and PDGFRA wild-type (n=1)
  • No KIT or PDGFRA mutation was detected for the patient with a partial response in the BSC group.
  • Median trough pazopanib concentration: 23.6 μg/mL in pazopanib + BSC group vs. 22.8 μg/mL in BSC alone group after switch
  • 4-month PFS: 42.5% in patients with trough pazopanib concentration < 20 μg/mL and 71.1% in patients with trough pazopanib concentration ≥ 20 μg/mL (p=0.17)
  • Median trough pazopanib concentration in patients without progression at 4 mos. was 24.7 μg/mL vs. 16.7 μg/mL in patients with disease progression by 4 mos. (p=0.078)
  • Treatment group was an independent prognostic marker for PFS in non-gastrectomized patients.


  • Pazopanib with BSC improved PFS compared to BSC alone in advanced GIST after disease progression on imatinib or sunitinib.
  • Adverse events were more frequent in the pazopanib + BSC group although three deaths were not pazopanib-related.
  • Genetic mutations were detected in some patients. However, small numbers limited the analysis. Posthoc analysis of pazopanib trough levels revealed no association with clinical outcomes.
  • PAZOGIST trial limitations include BSC as comparator arm, open- label, median follow-up of less than 30 mos., small sample size, crossover from BSC to pazopanib in 88% of patients, and lack of statistical power to detect an OS difference between the two groups.
  • PAZOGIST trial demonstrated a PFS advantage for pazopanib compared to BSC alone. However, this clinical benefit was marginal at best, and was associated with more toxicity.

More GI Cancer Updates


Extending Aromatase-Inhibitor Adjuvant Therapy to 10 Years
Goss, PE, et al., N Engl J Med. 2016 Jul 21;375(3):209-19.
PMID: 27264120


For women with ER/PR positive breast cancer, current guidelines recommend 5-10 years of hormonal therapy (AI, tamoxifen or both). Despite lack of efficacy data, some patients opt to continue AI beyond 5 years. Ross et al investigated this in a trial looking at 10 yrs. of AI in postmenopausal women with early breast cancer.


  • Patients: postmenopausal women with hormone positive breast cancer who received 4.5-6 yrs. of adjuvant AI (most preceded by tamoxifen) and disease-free
  • Study design: double-blind, placebo-controlled, phase III RCT
  • Intervention
    • Experimental: Letrozole 2.5 mg. PO daily x 5 yrs.
    • Control: Placebo PO daily x 5 yrs.
  • Outcomes
    • Primary: DFS
    • Secondary: OS, contralateral breast cancer, QOL, safety


  • 1,918 patients randomized to letrozole (n=959) or placebo (n=959)
  • Median follow-up: 6.3 yrs.
  • Baseline patient characteristics: median age 65 yrs., median time from first diagnosis of cancer 10.6 yrs., 90.5% T1-2 at diagnosis, 47.5% N1, 46.6% N0, 98.8% ER/PR positive, 5 yrs. median treatment with tamoxifen, 70.6% received 4.5-6 yrs. of tamoxifen, 5 yrs. median duration of AI, 99% received 4.5-6 yrs. of AI
  • Treatment adherence: 62.5% (L) vs. 62.3% (P)

Table 1: Efficacy Outcomes
Table 1

  • Adverse events were similar in both groups except for bone-related side effects with letrozole.
  • BMD change was significant between groups and favored placebo (P <0.001).
  • More patients on letrozole had lumbar spine T score <2.5 at any time after baseline (10% vs. 7%; P=0.03).

Table 2: Adverse events (selected)
Table 2

  • Use of bone-protecting meds was similar in both groups (Ca 86.1%, vit. D 84.5%, SERM 0.3%, bisphosphonates 46.2 % vs. 46.6%)
  • Treatment discontinuation due to adverse events: 5.4% (L) vs. 3.7% (P) QOL
  • >85% completed QOL assessment in both groups
  • SF-36 summary scores and MENQOL symptom subscales were similar in both groups.
  • Bodily pain was significantly greater in the letrozole group at 12, 24 and 36 mos. but lower at 48 and 60 mos.
  • Letrozole group had better emotional functioning scores at 12,36 and 60 mos. but worse at 24 and 48 mos.


  • At a median follow-up of 6.3 yrs., extending hormonal therapy by 5 more yrs. with letrozole after 4.5-6 yrs. of adjuvant treatment significantly improved 5-yr. DFS by 4% (absolute difference) and reduced risk of annual incidence of contralateral breast cancer by 58%. This however, did not translate to better OS.
  • Letrozole was associated with significantly higher rates of bone pain, new-onset osteoporosis, and bone fractures compared to placebo. Side effects were consistent with that previously reported in the literature. Despite these adverse events, QOL scores were similar in both groups.
  • In this study, treatment adherence was 62% (probably close to real life scenario) and patients in both groups received similar percentage of bone-protecting agents. Nevertheless, biases in terms of patient selection and self-reporting could not be completely discounted.
  • The MA17R trial revealed that in a small population of women with hormone positive breast cancer who received 4.5-6 yrs. of adjuvant hormonal therapy, adding 5 yrs. of letrozole significantly reduced disease recurrence and contralateral breast cancer. Due to the small incremental benefit at the risk of bone fractures and new-onset osteoporosis, women with good baseline BMD, able to tolerate AIs, and at higher risk for recurrence will most likely benefit from this treatment regimen. Whether the women who received initial AI will benefit more with extended AI compared to those who initially received tamoxifen remains unclear and requires further study.

More Updates in Breast Cancer