TABLE 1.
Summary of studies included in the systematic review and meta-analysis evaluating the association between total vegetable and fruit consumption and aerodigestive cancer prognosis1
| Cancer site | Outcome n (cases) | Follow-up (years) | Gender age (years) | Dietary assessment | Exposure categorization | HR/RR (95% CI) timeframe(2,3) | Adjustments | Author, year (ref.) (country) |
|---|---|---|---|---|---|---|---|---|
| Head and neck | ACM 2202 (445) | 3.2 ± 1.2 (mean ± SD) | M/F≥16 | FFQ | V: T3 (>1 p/d) vs. T1 (<5 p/wk)F: T3 (>1 p/d) vs. T1 (<1 p/wk) | 2 RR: 0.79 (0.61, 1.03)2RR: 0.91 (0.67, 1.23) | Age, sex, site, stage, comorbidity, treatment intent, education, relation status, income, smoking, alcohol, and fried food | Lang et al., 2019 (52) (UK) |
| Nasopharyngeal | ACM 1533 (243) | 3.3 (<1,5)median (min, max) | M/F46.1 (mean) | FFQ | F: daily or more vs. fewer than monthly servings | 2 HR: 0.78 (0.53, 1.14) | Age, sex, marital status, education level, clinical stage, smoking status, alcohol intake, and BMI | Shen et al., 2012 (53) (China) |
| Head and neck | ACM 504 (166) | 2.7 (<1,5.5) median (min, max) | M/F58.8 (mean) | FFQ | V: lowest (≤4 s/wk) vs. highest (≥5 s/wk)F: lowest (≤3 s/mo) vs. highest (≥1 s/wk) | 2 HR: 0.82 (0.59, 1.15)2HR: 1.26 (0.88, 1.81) | Age, sex, race, sleep score, educational level, marital status, cancer site, tumor stage, comorbidities, treatment received, smoking status, alcohol consumption, and physical activity | Duffy et al., 2009 (36) (USA) |
| Oral cavity and oropharynx | ACM146 (74)CCSM146 (49)R 146 (47) | 3.1 (mean)3.1 (mean)3.6 (mean) | M/Fcategories: <50, 50–59, 60–69 and ≥70 | FFQ | V: T3 (≥8) vs. T1 (≤4) s/wkF: T3 (≥8) vs. T1 (≤4) s/wkV: T3 (≥8) vs. T1 (≤4) s/wkF: T3 (≥8) vs. T1 (≤4) s/wkV: T3 (≥8) vs. T1 (≤4) s/wkF:T3 (≥8) vs. T1 (≤4) s/wk | 2 RR: 0.54 (0.30, 0.98)3RR: 0.14 (0.04, 0.50)2RR: 1.26 (0.73, 2.18)3RR: 0.77 (0.36, 1.64)1RR: 0.61 (0.29, 1.31)3RR: 0.14 (0.03, 0.69)2RR: 1.04 (0.52, 2.10)3RR: 0.89 (0.34, 2.32)2RR: 0.56 (0.27, 1.15)3RR: 0.43 (0.16, 1.18)2RR: 0.75 (0.37, 1.52)3RR: 0.68 (0.29, 1.59) | Age, sex, clinical stage, and tumor site | Sandoval et al., 2009 (56) (Spain) |
| Laryngeal/ hypopharyngeal | ACM931 (755) | 8 (mean)21 (max) | M/Fcategories: <50, 50–59, 60–69 and ≥70 | DQ | V: Q4 (>328.6) vs. Q1 (<148.6) g/dF: Q4 (>241) vs. Q1 (<56) g/d | 2 HR: 0.8 (0.6, 0.9)2HR: 1.0 (0.7, 1.2) | Age, sex, center, site of primary tumor, alcohol drinking, cigarette smoking, caloric intake without alcohol, and vegetable or fruit intake | Dikshit et al., 2005 (57) (Switzerland, France, Italy, and Spain) |
| Laryngeal | ACM215 (136) | 8–10 | Males59 (median) | DQ | V: T3 (<281.1) vs. T1 (<202.1) g/d | 2 HR: 0.57 (0.35, 0.94) | Age at diagnosis, clinical stage, occurrence of new primaries, and total calorie intake | Crosignani et al., 1996 (58) (Italy) |
| Gastric | ACM568 (345) | 1.2 (median)10 (max) | M/F63 (median) | FFQ | V+F: T1 (<2.7) vs. T3 (>4.3) s/d | 2 HR: 0.98 (0.75, 1.28) | Age, sex, education, extent of diseases, and total energy intake | Ferronha et al., 2012 (37) (Portugal) |
| Gastric | CCSM877 (241) | 10 (max) | M/F40–79 | DQ | F: highest (>3) vs. lowest (<3) times/wk | 2 HR: 0.98 (0.73, 1.31) | Age, sex, and pathological type and stage of cancer | Huang et al., 2000 (44) (Japan)4 |
| Colorectal | R1667 (738) | 0.5–10 | M/F55–74 | FFQ | V: T3 (≥1.5) vs. T1 (<1.1)5 F (excluding juice): T3 (≥0.9) vs. T1 (<0.5)5V+F: T3 (≥2.8) vs. T1 (<2.0)5 | 2 HR: 0.93 (0.69, 1.25)2HR: 0.96 (0.70, 1.32)2HR: 0.87 (0.64, 1.20) | Age, sex, center, race, energy intake, year of follow-up screening, adenoma at T0, T3, or T5, adequate screening at T0, T3, or T5, processed meat intake, red meat intake, calcium intake, smoking status, education, exercise, family history of colorectal cancer, use of NSAIDs, HRT, BMI, alcohol intake | Kunzmann et al., 2016 (45) (USA) |
| Colorectal | R87 (53) | 3 (max) | M/F65 (median) | 5d DR | V: highest (>110) vs. lowest (<110) g/dF+berries: highest (>200) vs. lowest (<200) g/d | 3 RR: 1.2 (0.5, 2.9)3RR: 0.9 (0.3, 2.2) | Colorectal cancer in a first-degree relative, BMI, and type of intervention | Almendingen et al., 2004 (46) (Norway) |
| Colorectal | ACM148 (46 at 5 y) | 10 (max)5 (RR ACM) | M/F30–79 | FFQ | V: T3 vs. T1F: T3 vs. T1 | 2 RR: 1.09 (0.49, 2.45)2RR: 0.84 (0.37, 1.88) | Age, sex, tumor stage, tumor location, and energy intake | Dray et al., 2003 (47) (France) |
| Lung | ACM1052 (869) | <1 (median) | Male≤80 | FFQ | V+F: frequent (≥1 s/d) vs. occasional (<1 s/d) | 3 HR: 0.86 (0.72, 1.02) | District of residence, age at diagnosis, BMI, cancer history in first-degree relatives, education level, family income, stage at diagnosis, smoking status, smoking pack-years, and treatment | Li et al., 2017 (48) (China) |
| Lung | ACM286 (ns) Current smokers | 11 (max) | M/F50–64 | FFQ | V: T3 (160–536) vs. T1 (16–88) g/dF: (excluding juices): T3 (143–671) vs. T1 (0–51) g/d | 2 HR: 0.84 (0.59, 1.21)2HR: 0.81 (0.58, 1.15) | Sex, age, extent of disease, duration of smoking, and potato and fruit/vegetable intake | Skuladottir et al., 2006 (49) (Denmark) |
1
Pre- and postdiagnosis dietary vegetables and fruit were not mixed in the meta-analysis. ACM, all-cause mortality; CCSM, cancer cause-specific mortality; DQ, dietary questionnaire; DR, dietary record; F, fruit; HRT, hormone replacement therapy; M/F, males and females; ns, not specified; NSAIDs, nonsteroidal anti-inflammatory drug; p/d, portion/day; p/wk, portion/week; R, cancer recurrence; ref., reference; RR, risk ratio; s/d, serving/day; s/mo, serving/month; s/wk, serving/week; V, vegetables; V+F, vegetables and fruit.
2
Prediagnosis.
3
Postdiagnosis.
4
Retrospective cohort study.
5
Food Patterns Equivalents Database (FPED) cup equivalents/1,000 kcal/d.