Comparison of ColoPrint risk classification with clinical risk in the prospective PARSC trial

465

Background: The 18-gene expression profile, ColoPrint, has been developed and validated for identifying risk of recurrence in patients with early-stage colon cancer (CC). In a pooled stage II validation study ColoPrint identified 63% of patients as Low Risk with a 3-yr recurrence-free survival (RFS) of 93% while High Risk patients had a 3-yr RFS of 82% with a HR of 2.7 (p=0.001). PARSC is a prospective study for the assessment of recurrence risk in stage II CC patients using ColoPrint. ColoPrint classification is compared to NCCN risk classification. Methods: The study enrolled 468 patients with histologically proven stage II CC from 31 institutes in Europe, the United States, and Asia between October 2008 and May 2013. Synchronous tumors were excluded. ColoPrint results were not disclosed to the physician and patient. Treatment was at the discretion of the physician, adhering to NCCN approved regimens or a recognized alternative. A McNemars test is performed to compare ColoPrint with NCCN risk classification. A p value ≤ 0.05 indicates the two tests differ significantly. Results: ColoPrint classified 320 (68%) patients as Low Risk and 148 (32%) as High Risk. 89 patients (19%) received adjuvant chemotherapy. In the ColoPrint Low Risk group, 57 (18%) patients received adjuvant chemotherapy while 32 (22%) of ColoPrint High Risk patients received chemotherapy. According to NCCN high risk factors (T4, high grade (exclusive of MSI-H), lymphovascular/perineural invasion, perforation/obstruction, <12 nodes examined, positive margins) 234 (50%) patients were NCCN Low Risk and 234 were NCCN High Risk. 72 (31%) of the NCCN Low Risk patients are ColoPrint High Risk. 158 (68%) of the NCCN High Risk patients are ColoPrint Low Risk. MSI-status was assessed in 86 (18%) patients of which 29 were MSI high and 57 were MSS. All MSI high were classified as ColoPrint Low Risk. Conclusions: The PARSC study is the first prospective study to compare genomic and clinical risk assessment and we observed marked differences between NCCN risk classification and ColoPrint. The clinical validity of these methods will be based on the outcomes at 3 and 5 years. Clinical trial information: NCT00903565. [Table: see text]

Statin use is associated with a reduced incidence of colorectal adenomatous polyps

BACKGROUND

Hydroxymethylglutaryl coenzyme A reductase inhibitors (statins) have been shown to have potentially useful anticancer effects against colorectal cancers in experimental studies, but clinical studies have shown inconsistent results on colorectal cancer incidence. Most colorectal cancers are believed to develop through the polyp-cancer sequence. We hypothesized that statins may protect against the development of adenomatous polyps, and this may contribute to the apparent cancer-protective effects.

OBJECTIVE

This study aims to compare previous statin use in patients with newly diagnosed adenomatous polyps against a control group without polyps.

METHOD

A case-control study involving 264 patients attending for diagnostic colonoscopy at the Norfolk and Norwich University Hospital was used. Polyp cases were age and sex matched against controls with normal colonoscopies. Structured patient interviews and clinical notes were used to ascertain drug and risk factor. Logistic regression was used to compare statin exposure and correct for confounding factors.

RESULTS

There was a significant negative association between prior statin use and a diagnosis of adenomatous polyps [odds ratio (OR) = 0.40 (0.24-0.76)]. The association was significantly stronger with higher statin doses [≥40 mg simvastatin or equivalent; OR 0.33 (0.10-0.53)] or longer duration of use [>5 years; OR 0.36 (0.10-0.67)]. Statin use was negatively associated with both high- and low-risk polyps.

CONCLUSIONS

Statins may have a protective effect against the development of adenomatous polyps. The negative association between statin use and polyp incidence showed a significant dose and duration relationship.

Statin use is associated with a reduced incidence of colorectal cancer: a colonoscopy-controlled case-control study

BACKGROUND

The aetiology of colorectal cancer (CRC) remains elusive in the majority of cases. There is experimental evidence to show that HMG-CoA reductase inhibitors (statins) may inhibit proliferation and induce cause apoptosis in CRC cells and although some clinical studies have suggested that statins may protect against the development of CRC, this has not been a consistent finding. Therefore we have examined any potential protective effects of statins by comparing statin use in patients with colorectal cancer against a control group.

METHODS

This was a case-control study examining statin use in symptomatic patients attending for diagnostic colonoscopy. Statin use was compared between patients with CRC and a control group, who had all had normal colonoscopy. Structured interviews and clinical records notes were used to determine drug exposure. Logistic regression was used to compare statin exposure and correct for confounding factors.

RESULTS

There was a significant inverse association between previous statin use and a diagnosis of CRC (OR = 0.43 (95% confidence interval 0.25 - 0.80), p<0.01). This inverse association was stronger with higher statin doses (OR = 0.19 (0.07 - 0.47), p<0.01) and greater duration of statin use (statin use >years: OR = 0.18 (0.06 - 0.55), p<0.01).

CONCLUSIONS

Statins use was associated with a protective effect against the development of CRC. This effect is associated with a significant dose and duration response. These findings need to be repeated in other observational studies before an interventional study can be considered.

BNIP3 expression in follicular lymphoma

AIMS

To investigate the role of BNIP3, a 19-kDa interacting protein of the Bcl-2 family, alongside Bcl-2 in follicular lymphoma in comparison with reactive lymphoid hyperplasia. The results were compared with those from p53 and caspase-3 (apoptotic markers) and Ki67 (proliferation marker).

METHODS AND RESULTS

Immunohistochemistry using monoclonal antibodies showed BNIP3 to be strongly expressed in most follicular lymphomas but to be weak to negative in all of the reactive cases. There was also an inverse relationship with Bcl-2 expression. There was no correlation of BNIP3 immunoreactivity with proliferation and caspase and p53 were virtually negative in all follicular lymphomas and reactive lymphoid cases.

CONCLUSIONS

BNIP3 is strongly expressed in most follicular lymphomas, especially those that are Bcl-2 negative. BNIP3 may serve as a marker of more aggressive behaviour in follicular lymphoma and be useful diagnostically in the distinction from reactive lymphadenitis.