Critical characterization of the impact of clinical-pathological tumoral features on recurrence risk dynamics of NSCLS.

Proposal
1907

Title of Proposed Research

Lead Researcher

Tommaso De Pas

Affiliation

European institute of Oncology , Milan

Funding Source

None

Potential Conflicts of Interest

None

Data Sharing Agreement Date

13 September 2017

Lay Summary

Lung cancer is the leading cause of cancer death in the USA.Surgery is the initial treatment for patients with stage I to II(N0-N1).After surgery all patients had routine postsurgical surveillance with clinical assessment and imaging studies,including chest computed tomography,but frequency and choice of follow-up modality are not standardized.Indeed,no evidence from randomized trials is available to define optimal follow-up in treated stage I-III NSCLC patients.As a result,surveillance guidelines and practices vary greatly especially in imaging frequency and modality.Currently,experts can only extrapolate recommendations for follow-up strategies,either based on evidence from retrospective analysis of consecutive patients series treated in single insitution or follow-up policies in large published clinical trials The interpretation of previous retrospective studies is limited by their small sample sizes and by the heterogeneity of their surveillance methods used. Although some studies have suggested that surveillance CT scan is effective at detecting early second primary lung cancer,the data are less clear regarding efficacy of early detection of recurrence and the impact on patients survival. Furthermore,albeit available data show that the majority of recurrences are detected by scheduled surveillance CT scan, a not negligible percentage of recurrences are detected outside of the routine follow-up procedures, most commonly as a result of the development of new symptoms. This considerable high percentage of follow-up missed recurrences highlight the need to improve the follow-up strategies.A way to achieve this objective it may be represented by tailoring follow-up procedures and schedules on the specific distribution of the recurrence risk over time, according with the different clinical-pathological features of each tumor.So far,the reliable and consistent evidences available show that on average,a significant proportion(30%-60%)of patients who underwent resection for pathological stage IA-IIB NSCLC develop a locoregional or distant recurrence; that the hazard rate for disease recurrence is about 6%-7% per patient per year during the first 4 years, diminishing to2% per patient per year thereafter; and that, in addition, there is a smooth increase of the hazard rate for second primary cancer from 1% to 3% per patient per year during the first 3 years, which does not diminish over time.We have less evidences on whether and how the clinical-pathological features of tumors substantially modify the pattern of anatomical sites of recurrence as well as the distribution of the recurrence risk over time, as it has been largely proved in breast cancer. On the light of the limitations of the available data on the recurrences dynamics of NSCLCs that likely hamper the possibility of improve follow-up strategies, we propose to analyze the data of follow-up of patients enrolled in MAGRIT trial,to address specific questions

Study Data Provided

[{ "PostingID": 4790, "Title": "GSK-109493", "Description": "GSK1572932A Antigen-Specific Cancer Immunotherapeutic as adjuvant therapy in patients with resectable MAGE-A3 positive Non-Small Cell Lung Cancer" }]

Statistical Analysis Plan

1) For the first objective, (To Assess the anatomical sites of recurrences, with the aim to identify potential different recurrence patterns according with relevant clinical and pathological features of tumors) we will analyze the potential interactions between clinical and pathological tumors features and anatomical sites of recurrences. Features of tumors analyzed will be: histology (adenocarcinoma vs squamous) , grade, genomic status (EGFR mutated vs ALK translocated vs KRas mutated vs wildtype), stage (Ib vs II vs IIIa), previous treatment received (Type of surgery, type of lymphadenectomy, adjuvant chemo and/or radiotherapy ), sex, smoke status.For each patients, we will analyze the type of recurrence (distant, regional, or local). Local recurrence will be defined as any new lesion adjacent to a staple line, to the bronchial stump, or in the residual lobe (in cases of sublobar resection). Regional recurrences involves lymph node stations 1 to 14 or the ipsilateral lung. Distant metastasis will consist of extrathoracic disease, metastasis to the contralateral lung, pleural metastasis, pleural effusion, or pericardial effusion Second primary lung cancer will be defined according to the criteria of Martini and Melamed. In particular, a new, distinct pulmonary malignancy is considered a second primary cancer if it fulfilled any 1 of the following 3 criteria: (1) histologic results different from those of the index tumor; (2) same histologic results as the index tumor but diagnosed 2 years after the primary tumor; or (3) same histologic results as the index tumor, diagnosed within 2 years of the primary tumor, but located in different lobes or segments, with no positive intervening lymph nodes and no evidence of metastasis 2) For the second objective, (To Asses the distribution of second primary (SP), distant (DM) and locoregional (LR) recurrence risk overtime, with the aim to identify potential different recurrence dynamics according with relevant clinical pathological features of tumors specified above)we will analyze for each patient the time to treatment failure, defined as the interval between the date of surgery to date of disease recurrence (LR or DM) or development of an SP. Only first events will be considered. The recurrence event dynamics will be studied by estimating the discrete hazard rate for the considered event with the life-table method, i.e., the conditional probability of manifesting the event in a time interval, given that the patient did not previously experience it at the beginning of the interval. A discretization of the time axis in 3-month units will be applied, and all hazard rate levels will be measured as “events/patients at risk per 3-month interval.” 3) For the third objective, (To Assess the mode of detection of distant, locoregional and second primary recurrence, in particular with the aim to quantify the percentage of each type of recurrences missed by scheduled follow-up procedures, as well as the percentage of recurrences identified by follow-up and subsequently treated with curative intent) we will analyze the method of detection, and whether the recurrence event was detected at a scheduled appointment, as part of routine surveillance, or through an unscheduled visit outside of the follow-up protocol. The manner of detection—whether by scheduled CT scan, presentation of new symptoms, or other means—will be defined as the modality that led to the initiation of further workup and diagnosis of treatment failureFinally, if possible, we also calculate the percentage of False-Positive Results from follow-up Surveillance: i.e asymptomatic patients with had abnormal or indeterminate surveillance CT scan results and underwent additional testing that did not result in a diagnosis of recurrence or second primary cancer.

Publication Citation

https://doi.org/10.1016/j.ejca.2019.11.005