The purpose of this study is to (i) investigate the impact of various air gaps in conjunction with a range shifter of 7.5 cm water-equivalent-thickness (WET) on in-air spot size of a pencil proton beam at the isocenter and off-axis points, and (ii) compare the treatment planning system (TPS) calculated spot sizes against the measured spot sizes. A scintillation detector has been utilized to measure the in-air spot sizes at the isocenter. The air gap was varied from 0 to 35 cm at an increment of 5 cm. For each air gap, a single spot pencil proton beam of various energies (110–225 MeV) was delivered to the scintillation detector. By mimicking the experimental setup in RayStation TPS, proton dose calculations were performed using pencil beam (RS-PB) and Monte Carlo (RS-MC) dose calculation algorithms. The calculated spot sizes (RS-PB and RS-MC) were then compared against the measured spot sizes. For a comparative purpose, the spot sizes of each measured energy for different air gaps of (5–35 cm) were compared against that of 0 cm air gap. The results of the 5 cm air gap showed an increase in spot size by ≤ 0.6 mm for all energies. For the largest air gap (35 cm) in the current study, the spot size increased by 3.0 mm for the highest energy (225 MeV) and by 9.2 mm for the lowest energy (110 MeV). For the 0 cm air gap, the agreement between the TPS-calculated (RS-PB and RS-MC) and measured spot sizes were within ± 0.1 mm. For the 35 cm air gap, the RS-PB overpredicted spot sizes by 0.3–0.8 mm, whereas the RS-MC computed spot sizes were within ± 0.3 mm of measured spot sizes. In conclusion, spot size increment is dependent on the energy and air gap. The increase in spot size was more pronounced at lower energies (< 150 MeV) for all air gaps. The comparison between the TPS calculated and measured spot sizes showed that the RS-MC is more accurate (within ± 0.3 mm), whereas the RS-PB overpredicted (up to 0.8 mm) the spot sizes when a range shifter (7.5 cm WET) and large air gaps are encountered in the proton beam path. © 2019, Australasian College of Physical Scientists and Engineers in Medicine.