Prognostic indexes in breast cancer: comparison of the Nottingham and Adelaide indexes

Prognostic models for breast cancer: a systematic review

BACKGROUND

Breast cancer is the most common cancer in women worldwide, with a great diversity in outcomes among individual patients. The ability to accurately predict a breast cancer outcome is important to patients, physicians, researchers, and policy makers. Many models have been developed and tested in different settings. We systematically reviewed the prognostic models developed and/or validated for patients with breast cancer.

METHODS

We conducted a systematic search in four electronic databases and some oncology websites, and a manual search in the bibliographies of the included studies. We identified original studies that were published prior to 1st January 2017, and presented the development and/or validation of models based mainly on clinico-pathological factors to predict mortality and/or recurrence in female breast cancer patients.

RESULTS

From the 96 articles selected from 4095 citations found, we identified 58 models, which predicted mortality (n = 28), recurrence (n = 23), or both (n = 7). The most frequently used predictors were nodal status (n = 49), tumour size (n = 42), tumour grade (n = 29), age at diagnosis (n = 24), and oestrogen receptor status (n = 21). Models were developed in Europe (n = 25), Asia (n = 13), North America (n = 12), and Australia (n = 1) between 1982 and 2016. Models were validated in the development cohorts (n = 43) and/or independent populations (n = 17), by comparing the predicted outcomes with the observed outcomes (n = 55) and/or with the outcomes estimated by other models (n = 32), or the outcomes estimated by individual prognostic factors (n = 8). The most commonly used methods were: Cox proportional hazards regression for model development (n = 32); the absolute differences between the predicted and observed outcomes (n = 30) for calibration; and C-index/AUC (n = 44) for discrimination. Overall, the models performed well in the development cohorts but less accurately in some independent populations, particularly in patients with high risk and young and elderly patients. An exception is the Nottingham Prognostic Index, which retains its predicting ability in most independent populations.

CONCLUSIONS

Many prognostic models have been developed for breast cancer, but only a few have been validated widely in different settings. Importantly, their performance was suboptimal in independent populations, particularly in patients with high risk and in young and elderly patients.

Predicting and communicating the risk of recurrence and death in women with early-stage breast cancer: a systematic review of risk prediction models

BACKGROUND

It is a challenge for oncologists to distinguish patients with breast cancer who can forego adjuvant systemic treatment without negatively affecting survival from those who cannot. Risk prediction models (RPMs) have been developed for this purpose. Oncologists seem to have embraced RPMs (particularly Adjuvant!) in clinical practice and often use them to communicate prognosis to patients. We performed a systematic review of published RPMs and provide an overview of the prognosticators incorporated and reported clinical validity. Subsequently, we selected the RPMs that are currently used in the clinic for a more in-depth assessment of clinical validity. Finally, we assessed lay comprehensibility of the reports generated by RPMs.

METHODS

Pubmed, EMBASE, and Web of Science were searched. Two reviewers independently selected relevant articles and extracted data. Agreement on article selection and data extraction was achieved in consensus meetings.

RESULTS

We identified RPMs based on clinical prognosticators (N = 6) and biomolecular features (N = 14). Generally predictions from RPMs seem to be accurate, except for patients ≤ 50 years or ≥ 75 years at diagnosis, in addition to Asian populations. RPM reports contain much medical jargon or technical details, which are seldom explained in lay terms.

CONCLUSION

The accuracy of RPMs' prognostic estimates is suboptimal in some patient subgroups. This urgently needs to be addressed. In their current format, RPM reports are not conducive to patient comprehension. Communicating survival probabilities using RPM might seem straightforward, but it is fraught with difficulties. If not done properly, it can backfire and confuse patients. Evidence to guide best communication practice is needed.

Short-Term Prognostic Index for Breast Cancer: NPI or Lpi

Axillary lymph node involvement is an important prognostic factor for breast cancer survival but is confounded by the number of nodes examined. We compare the performance of the log odds prognostic index (Lpi), using a ratio of the positive versus negative lymph nodes, with the Nottingham Prognostic Index (NPI) for short-term breast cancer specific disease free survival. A total of 1818 operable breast cancer patients treated in the University Hospital of Leuven between 2000 and 2005 were included. The performance of the NPI and Lpi were compared on two levels: calibration and discrimination. The latter was evaluated using the concordance index (cindex), the number of patients in the extreme groups, and difference in event rates between these. The NPI had a significant higher cindex, but a significant lower percentage of patients in the extreme risk groups. After updating both indices, no significant differences between NPI and Lpi were noted.

Survival of invasive breast cancer according to the Nottingham Prognostic Index in cases diagnosed in 1990-1999

The Nottingham Prognostic Index (NPI) is a well established and widely used method of predicting survival of operable primary breast cancer.

AIMS

Primary: To present the updated survival figures for each NPI Group. Secondary: From the observations to suggest reasons for the reported fall in mortality from breast cancer.

METHODS

The NPI is compiled from grade, size and lymph node status of the primary tumour. Consecutive cases diagnosed and treated at Nottingham City Hospital in 1980-1986 (n=892) and 1990-1999 (n=2,238) are compared. Changes in protocols towards earlier diagnosis and better case management were made in the late 1980s between the two data sets.

RESULTS

Case survival (Breast Cancer Specific) at 10 years has improved overall from 55% to 77%. Within all Prognostic groups there are high relative and absolute risk reductions. The distribution of cases to Prognostic groups shows only a small increase in the numbers in better groups.

CONCLUSION

The updated survival figures overall and for each Prognostic group for the NPI are presented.

Reading the prognosis of the individual with breast cancer

AIM

To obtain better survival estimates for the individual than is provided by placement in an NPI group.

METHOD

Consecutive primary operable breast cancers treated at Nottingham City Hospital 1990-1999. Ten year % actuarial survivals plotted for 10 ranges of NPI from 2.0 to 6.9. There is an excellent inverse correlation between median NPI value for each range and survival at 10 years. To enable estimation of survival for all individual values of NPI, a curve fitting technique applied to these results (by G.B.) gave the formula to estimate survival from the individual's NPI score: 10 year % survival for the individual=-3.0079 x NPI(2)+12.30 x NPI+83.84. This gave an r(2) of 0.98.

RESULTS AND CONCLUSION

Greater accuracy in individual survival prediction is obtained by dividing women into 10 groups by NPI scores than in the originally described six groups; rank order of survival in relation to NPI score is preserved. A curve fitting technique has been applied to these data to give a formula for the prediction of 10 year survival for every 0.1 value of NPI.

Inhibition of the stress response to breast cancer surgery by regional anesthesia and analgesia does not affect vascular endothelial growth factor and prostaglandin E2

Angiogenesis is essential for breast cancer metastases formation and is mediated by vascular endothelial growth factor (VEGF) and prostaglandin E2 (PGE2). We hypothesized that serum levels of VEGF and PGE2 are increased by the stress response to breast cancer surgery and attenuated by paravertebral anesthesia and analgesia (PVAA). Thirty women undergoing mastectomy were enrolled in this prospective, randomized study, to receive general anesthesia (GA) and postoperative opioid analgesia (morphine 0.1 mg/kg bolus and patient-controlled infusion) or GA and PVAA (72-h infusion). All patients received rectal diclofenac. Venous blood samples were taken preoperatively and at 4 and 24 h postoperatively for serum glucose, cortisol, C-reactive protein, VEGF, and PGE2. PVAA inhibited the surgical stress response, as indicated by significantly less plasma glucose, cortisol, and C-reactive protein. VEGF and PGE2 values did not differ significantly between the groups. Mean (SD) percentage change in VEGF at 4 and 24 h respectively were 3% +/- 44% versus 9% +/- 80%, P=0.29 and 5% +/- 43% versus -10% +/- 63%, P=0.41 for patients with combined general and PVAA and GA alone, respectively. Mean percentage change in postoperative PGE2 at 4 and 24 h respectively was 10% +/- 17% versus 11% +/- 69%, P=0.29 and 34% +/- 19% versus 47% +/- 18%, P=0.15. We conclude that despite inhibiting the surgical stress response, PVAA had no effect on serum levels of putative breast cancer angiogenic factors, VEGF and PGE2.