M. J. Moore

Association of hypomagnesemia with inferior survival in a phase III, randomized study of cetuximab plus best supportive care versus best supportive care alone: NCIC CTG/AGITG CO.17

BACKGROUND Cetuximab-induced hypomagnesemia has been associated with improved clinical outcomes in advanced colorectal cancer (CRC). We explored this relationship from a randomized clinical trial of cetuximab plus best supportive care (BSC) versus BSC alone in patients with pretreated advanced CRC. PATIENTS AND METHODS Day 28 hypomagnesemia grade (0 versus ≥1) and percent reduction (<20% versus ≥20%) of Mg from baseline was correlated with outcome. RESULTS The median percentage Mg reduction at day 28 was 10% (-42.4% to 63.0%) for cetuximab (N = 260) versus 0% (-21.1% to 25%) for BSC (N = 251) [P < 0.0001]. Grade ≥1 hypomagnesemia and ≥20% reduction from baseline at day 28 were associated with worse overall survival (OS) [hazard ratio, HR 1.61 (95% CI 1.12-2.33), P = 0.01 and 2.08 (95% CI 1.32-3.29), P = 0.002, respectively] in multivariate analysis including grade of rash (0-1 versus 2+). Dyspnea (grade ≥3) was more common in patients with ≥20% versus < 20% Mg reduction (68% versus 45%; P = 0.02) and grade 3/4 anorexia were higher in patients with grade ≥1 hypomagnesemia (81% versus 63%; P = 0.02). CONCLUSIONS In contrast to prior reports, cetuximab-induced hypomagnesemia was associated with poor OS, even after adjustment for grade of rash.

Preoperative investigations for metastatic staging of colon and rectal cancer across multiple centres – what is current practice?

Objective  The optimal strategy for elective distant staging of colorectal carcinoma (CRC) has yet to be defined, with current guidelines based on small and limited series. One specific issue requiring review is the value of routine computerized tomographic (CT) chest examination. Also lacking is data on current routine clinical practice.

Patient comorbidities and behaviour once diagnosed are major contributors to disparities in cancer health outcomes.

TO THE EDITOR: As broad progress is made in the treatment and outcomes of cancer, it becomes increasingly important that all patients have the opportunity to benefit from these advances, so we commend the American Society of Clinical Oncology (ASCO) for highlighting the significant issue of disparities in outcomes. The main focus of the policy statement, however, is the presumed major contributions of poor quality of care and limited access to treatment. While this is consistent with the available data, we believe too little attention is paid to the potential impact of differences in patient comorbidities and/or behavior on cancer outcomes. Based on our own data, we would argue that further study of attitudes to disease and treatment may permit tailored intervention that will reduce some of the observed differences between advantaged and disadvantaged populations. The Australian model of healthcare differs markedly from the United States in that there are parallel systems—a publicly funded universal health care system, known as Medicare, and a private system. Many Australians choose to pay private health insurance to enable choice of doctor and treatment location and to avoid long waiting lists for elective procedures. Of note, there are many instances where the public and private hospital systems are closely affiliated, with the same surgeons and medical oncologists providing care in both settings. For patients presenting with cancer, the same range of treatment options is available, with all standard therapy made available in both systems through the government-funded pharmaceutical benefits scheme. Thus the Australian model provides a unique opportunity to contrast the outcomes of privately insured patients (an advantaged population), with that of public patients (a more disadvantaged population), where any disparity in outcome is unlikely to be because of issues such as the quality of care received or a difference in access to care. To explore the differences between private and public patients, and how these could impact on the presentation, treatment, and outcome of cancer, we examined data from four hospitals, consisting of two sets of affiliated public and private centers. We identified 1,930 patients diagnosed with colorectal cancer between January 2000 (when prospective comprehensive data collection was initiated) and December 2007, 1,568 (81%) of whom were treated in the public system and 362 (19%) in the private sector. Table 1 illustrates some of the differences in patient characteristics between the two populations. The public patients are a median of 3 years older than the private patients (P .001). As expected, they have a greater degree of social disadvantage measured by the Index of Relative Social Advantage/Disadvantage score, which is calculated on a postcode basis using indicators such as education, level of unemployment, and median income. There are also clear differences between the two populations with respect to specific comorbidities (diabetes) and lifestyle choices (smoking). Both smoking and diabetes are significantly associated with premature death unrelated to cancer diagnosis, and in the case of diabetes, worse cancer-specific survival outcomes. Inferior overall scores for the public patients when assessed by the general measure of operative risk, the American Society of Anaesthesiology score, are also consistent with an excess of comorbidities and poorer general health among the public patients. Features related to the cancer diagnosis are presented in Table 2. The marked excess of emergency presentations in the public patients suggests these patients may delay seeking medical attention for their symptoms, and the overall differences in stage at presentation are consistent with this, including an excess of public patients presenting with metastatic disease. Adjuvant treatment data is also presented in Table 2. For the 623 patients with stage II (n 200) or III colon cancer (n 423), the medical oncologist recommended adjuvant treatment in a greater percentage of private patients than public (63% v 54%; P .003). A significantly greater proportion of the public patients (52 of 482; 10.79%) compared with private patients (seven of 141; 4.96%) decided not to pursue treatment against medical advice (P .0001). The greater number of private patients offered treatment is consistent with differences in the age and comorbidity profile of the two groups, whereas the lower acceptance of treatment advice may represent differences in knowledge, attitudes, and/or beliefs about cancer. Table 1. Patient and Tumor Characteristics